WO2020172919A1 - Ai intelligent process abnormality recognition closed-loop control method, host and device system - Google Patents

Ai intelligent process abnormality recognition closed-loop control method, host and device system Download PDF

Info

Publication number
WO2020172919A1
WO2020172919A1 PCT/CN2019/078376 CN2019078376W WO2020172919A1 WO 2020172919 A1 WO2020172919 A1 WO 2020172919A1 CN 2019078376 W CN2019078376 W CN 2019078376W WO 2020172919 A1 WO2020172919 A1 WO 2020172919A1
Authority
WO
WIPO (PCT)
Prior art keywords
abnormality
state
data
level
frequency
Prior art date
Application number
PCT/CN2019/078376
Other languages
French (fr)
Chinese (zh)
Inventor
何成鹏
周建
张雨军
刘学森
Original Assignee
武汉三工智能装备制造有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 武汉三工智能装备制造有限公司 filed Critical 武汉三工智能装备制造有限公司
Publication of WO2020172919A1 publication Critical patent/WO2020172919A1/en

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Definitions

  • This application relates to the technical field of solar photovoltaic module manufacturing, and in particular to a closed-loop control method, host and equipment system for AI intelligent process abnormality recognition.
  • the main purpose of this application is to provide an AI (Artificial Intelligence, Artificial Intelligence) Intelligence) Process anomaly identification closed-loop control method, host computer and equipment system are designed to solve the problems of low yield and low production efficiency due to untimely, difficult to quickly identify and fast mark in the manual operation process in the prior art.
  • AI Artificial Intelligence, Artificial Intelligence
  • Intelligence Process anomaly identification closed-loop control method, host computer and equipment system are designed to solve the problems of low yield and low production efficiency due to untimely, difficult to quickly identify and fast mark in the manual operation process in the prior art.
  • this application provides a closed-loop control method, host and equipment system for AI intelligent process abnormality recognition, wherein the AI intelligent process abnormality recognition closed-loop control method includes the following steps:
  • a corresponding exception handling strategy is performed according to the first exception level.
  • the first abnormality level includes severe abnormality, major abnormality, and minor abnormality
  • the abnormality handling strategy includes generating a shutdown information instruction
  • the abnormality handling strategy includes:
  • Second abnormality handling information includes an instruction to generate shutdown information
  • the abnormality handling strategy includes:
  • the third abnormality processing information is generated according to the third abnormality level, and the third abnormality processing information includes an instruction to generate warning information.
  • the method further includes:
  • the state-related data is matched with the abnormal level database, and when the matching is unsuccessful, the frequency data of the state-related data is counted;
  • the fourth abnormality processing information is generated by the fourth abnormality level.
  • the method further includes:
  • the state-related data is recorded and updated to the abnormal level database regularly.
  • the present application also provides a control host, the control host including: a memory, a processor, and an AI intelligent process abnormality recognition closed-loop control program stored on the memory and running on the processor, the AI intelligent process
  • the abnormality recognition closed-loop control program is configured to implement the steps of the AI intelligent process abnormality recognition closed-loop control method, and the AI intelligent process abnormality recognition closed-loop control method includes:
  • a corresponding exception handling strategy is performed according to the first exception level.
  • the AI intelligent process abnormality recognition closed-loop control equipment system includes a control host and an identification component electrically connected to the control host And executive components, where:
  • the identification component is configured to obtain state-related data of the solar cell, and send the state-related data to the control host;
  • the execution component is configured to perform abnormal processing after receiving the first abnormal processing information of the control host;
  • the control host includes: a memory, a processor, and an AI intelligent process abnormality recognition closed-loop control program stored on the memory and running on the processor, and the AI intelligent process abnormality recognition closed-loop control program is configured to realize AI
  • the steps of a closed-loop control method for intelligent process abnormality recognition includes:
  • a corresponding exception handling strategy is performed according to the first exception level.
  • the identification component includes a plurality of identification sensors, and the plurality of identification sensors includes an image identification sensor, a temperature sensor, and a photoelectric sensor.
  • the solar cell module production line has a plurality of workstations, and the plurality of workstations include a picking station, a transmission station, a layup station, a string welding station, a bus welding station, and a layout station;
  • a plurality of the identification sensors are correspondingly distributed at a plurality of the workstations, and are configured to obtain the state-related data of the solar cells at the corresponding workstations.
  • it further includes a marking component electrically connected to the control host;
  • the control host is also configured to generate abnormal distribution data information and marking information from the state-related data, and send the marking information to the marking component, and the abnormal distribution data information is set to guide the work of the repairer;
  • the marking component is configured to receive the marking information and perform marking processing on the abnormal solar cell.
  • the marking component includes a laser coder, configured to perform coding processing on the abnormal solar cell sheet and/or the glass substrate of the solar cell component.
  • This application receives state-related data of solar cells in real time and compares the state-related data with normal state data to determine the state of the solar cells.
  • the solar cells are in an abnormal state, pass the state
  • the relevant data is matched with the abnormality level database to obtain the first abnormality level, and the corresponding abnormality processing strategy is carried out according to the first abnormality level, which can intelligently monitor the status of the solar cell module production line in real time, according to the In different states, the operation of the solar cell module production line is automatically guided and controlled, which reduces human involvement, improves the production efficiency of the solar cell module production line, and also improves the product yield.
  • FIG. 1 is a schematic diagram of a server structure of a hardware operating environment involved in a solution of an embodiment of the present application
  • FIG. 2 is a schematic flowchart of a first embodiment of a closed-loop control method for anomaly recognition of AI intelligent process according to the application;
  • FIG. 3 is a schematic flowchart of a second embodiment of a closed-loop control method for identifying abnormalities in an AI intelligent process according to this application;
  • FIG. 4 is a schematic flowchart of a third embodiment of a closed-loop control method for anomaly recognition of AI intelligent process according to the application;
  • FIG. 5 is a schematic flowchart of a fourth embodiment of a closed-loop control method for identifying abnormalities in an AI intelligent process according to the application;
  • Fig. 6 is a schematic structural diagram of an embodiment of a closed-loop control system for identifying abnormalities in AI intelligent process of the application;
  • FIG. 7 is a schematic structural diagram of an embodiment of a solar cell module production line (partial structure) applying the AI intelligent process abnormality recognition closed-loop control system in FIG. 6 in this application;
  • FIG. 8 is a schematic structural diagram of an embodiment of an identification component of the AI intelligent process abnormality identification closed-loop control system in FIG. 6;
  • FIG. 9 is a schematic structural diagram of an embodiment of the marking component of the closed-loop control system for AI intelligent process abnormality recognition in FIG. 6.
  • Label name 1000 AI intelligent process abnormal recognition closed-loop control equipment system 20 Recognition sensor a Solar cell module production line 300 Executive component 100 Control host 400 Marking components 200 Identify components 40 Laser Coder
  • Figure 1 is a schematic diagram of the structure of the control host of this application.
  • the control host may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005.
  • the communication bus 1002 is configured to realize connection and communication between these components.
  • the user interface 1003 may include a display screen (Display) and an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface.
  • the network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).
  • the memory 1005 can be a high-speed RAM memory or a stable memory (non-volatile memory), such as disk storage.
  • the memory 1005 may also be a storage device independent of the foregoing processor 1001.
  • FIG. 1 does not constitute a limitation on the control host, and may include more or less components than shown in the figure, or a combination of certain components, or different component arrangements.
  • the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a control program of the AI intelligent process abnormality recognition closed-loop control host 100.
  • the network interface 1004 is mainly set to connect to terminal devices and perform data communication with the terminal devices;
  • the user interface 1003 is mainly set to receive input instructions from the administrator;
  • the server calls the memory 1005 through the processor 1001
  • the stored AI intelligent process abnormally recognizes the control program of the closed-loop control host 100, and performs the following operations:
  • a corresponding exception handling strategy is performed according to the first exception level.
  • processor 1001 may call the control program of the AI intelligent process abnormal recognition closed-loop control host 100 stored in the memory 1005, and also perform the following operations:
  • the first abnormality level includes severe abnormality, major abnormality and minor abnormality
  • the abnormality handling strategy includes generating a shutdown information instruction
  • the abnormality handling strategy includes:
  • Second abnormality handling information includes an instruction to generate shutdown information
  • the abnormality handling strategy includes:
  • the third abnormality processing information is generated according to the third abnormality level, and the third abnormality processing information includes an instruction to generate warning information.
  • processor 1001 may call the control program of the AI intelligent process abnormal recognition closed-loop control host 100 stored in the memory 1005, and also perform the following operations:
  • the method further includes:
  • the state-related data is matched with the abnormal level database, and when the matching is unsuccessful, the frequency data of the state-related data is counted;
  • the fourth abnormality processing information is generated by the fourth abnormality level.
  • processor 1001 may call the control program of the AI intelligent process abnormal recognition closed-loop control host 100 stored in the memory 1005, and also perform the following operations:
  • the method further includes:
  • the state-related data is recorded and updated to the abnormal level database regularly.
  • This application receives state-related data of solar cells in real time and compares the state-related data with normal state data to determine the state of the solar cells.
  • the solar cells are in an abnormal state, pass the state
  • the relevant data is matched with the abnormality level database to obtain the first abnormality level, and the corresponding abnormality processing strategy is carried out according to the first abnormality level, which can intelligently monitor the status of the solar cell module production line in real time, according to the In different states, the operation of the solar cell module production line is automatically guided and controlled, which reduces human involvement, improves the production efficiency of the solar cell module production line, and also improves the product yield.
  • FIGS. 2 to 6 are embodiments of the closed-loop control method for AI intelligent process abnormality recognition provided by this application.
  • FIG. 2 is a schematic flowchart of the first embodiment of the AI intelligent process abnormality recognition closed-loop control method provided by this application.
  • the AI intelligent process abnormality recognition closed-loop control method includes the following steps:
  • Step S10 receiving real-time state-related data of the solar cell
  • the state-related data of the solar cell includes image data, temperature data, welding forming quality data, positioning data and other related data.
  • the above data are all related to the solar cell.
  • the quality of the battery slice is related.
  • the above status data is not limited, and a lot of status data can also be included.
  • Step S20 comparing the state-related data with normal state data, and judging the state of the solar cell
  • the normal state data of the solar cells are all stored in the control system, including image data of normal cells, temperature data, welding quality data, positioning data and other related data.
  • the state-related data is compared with the normal state data to determine the state of the solar cell. If it is a normal state, the detected solar cell is in a normal state.
  • the production line is operating normally.
  • Step S30 When the solar cell is in an abnormal state, match the state-related data with an abnormal level database to obtain a first abnormal level;
  • an abnormality level database will be established in the system.
  • the abnormality level database classifies different abnormal defects. According to different categories, the abnormality level is classified according to a preset rule. For example, it can be classified as a mapping table. Form, if the detected solar cell is in an abnormal state, the entire system determines that the detected solar cell is in an abnormal state, searches for the abnormality category paired with the abnormal state, and then obtains the corresponding abnormality level .
  • Step S40 Perform a corresponding abnormality processing strategy according to the first abnormality level
  • This application receives state-related data of solar cells in real time and compares the state-related data with normal state data to determine the state of the solar cells.
  • the solar cells are in an abnormal state, pass the state
  • the relevant data is matched with the abnormality level database to obtain the first abnormality level, and the corresponding abnormality processing strategy is performed according to the first abnormality level, which can intelligently monitor the status of the cells on the solar cell module production line a in real time.
  • the operation of the solar cell module production line a is automatically guided and controlled, which reduces human involvement, improves the production efficiency of the solar cell module production line a, and also improves the yield of products.
  • FIG. 3 is a schematic flowchart of the second embodiment of the AI intelligent process abnormality recognition closed-loop control method provided by this application.
  • the first abnormality level includes severe abnormality, major abnormality, and minor abnormality.
  • Step S40a When the first abnormality level is a serious abnormality, the abnormality handling strategy includes generating a shutdown information instruction;
  • the first abnormality level is classified into three levels: severe abnormality, major abnormality, and minor abnormality.
  • the first abnormality level may also be Differentiating multiple abnormal levels makes the monitoring of the production line more detailed. For example, when severe fragments occur and the area is large, the production quality of the entire product is seriously affected, and it is determined as a serious abnormality, and a general abnormality is determined as a major abnormality. When only slight scratches appear, and the production quality of the entire product is not affected or has a small effect, it is determined as a minor abnormality.
  • Step S40b When the first abnormality level is a major abnormality, the abnormality handling strategy includes the first step of counting the first frequency data of the state-related data in the first time period;
  • the first abnormality level is the main abnormality
  • the frequency of the main abnormality is too high, it often means that the equipment has a serious failure or the material has a serious failure, which will also seriously affect the entire product. Production quality, at this time, it is necessary to count the first frequency data of the state-related data in the first time period for further judgment.
  • Step S50b Calculate the first fault-tolerant frequency in the first time period based on the risk probability of serious incidents of internal equipment and/or products pre-stored in the abnormal level database;
  • the data on the probability of serious incidents of internal equipment and/or products is stored in the abnormality level database. Based on the probability, the first fault-tolerant frequency of abnormal occurrence during the effective working time can be calculated.
  • the frequency is used as a reference for judgment.
  • the frequency of the abnormal state data of the minor abnormality is N times within a period of time. As long as the frequency exceeds the fault tolerance frequency N times, it corresponds to an abnormal level, which is convenient for comparison and judgment;
  • Step S60b Obtain a second abnormality level of the state-related data by comparing the first frequency data with the first fault tolerance frequency;
  • the frequency of the abnormal state data of the main abnormality is N times, as long as the frequency of fault tolerance exceeds the frequency of N times, it corresponds to an abnormal level, and when the frequency of fault tolerance is within the frequency of N times Corresponding to another abnormality level, which can be handled differently according to different abnormality levels.
  • Step S70b The second abnormal handling information includes an instruction to generate shutdown information
  • second abnormality handling information is generated through the second abnormality level, such as stopping or continuing production.
  • Step S40c When the first abnormality level is a secondary abnormality, the abnormality processing strategy includes the first step of counting the second frequency data of the state-related data in the second time period;
  • the first abnormality level is a secondary abnormality
  • the frequency of the secondary abnormality is too high, it will also affect the production quality of the entire product. At this time, it is necessary to count the occurrence of the state-related data. The frequency data for further judgment.
  • Step S50c Calculate the second fault tolerance frequency in the second time period based on the repetitive batch accident probability of internal equipment and/or products pre-stored in the abnormal level database;
  • the data on the probability of repetitive batch accidents of internal equipment and/or products is stored in the abnormal level database. Based on the probability, the second fault-tolerant frequency of abnormal occurrence during effective working hours can be calculated.
  • the fault tolerance frequency is used as a reference for judgment.
  • the frequency of the abnormal state data of the minor abnormality is N times within a period of time, which is convenient for comparison and judgment;
  • Step S60c comparing the second frequency data with the second fault tolerance frequency to obtain the third abnormality level of the state-related data
  • the frequency of the abnormal state data of the minor abnormality is N times within a period of time. As long as the frequency of fault tolerance exceeds this frequency of N times, it corresponds to an abnormal level. When the frequency of fault tolerance is N times The inside corresponds to another abnormal level, which can be handled differently according to different abnormal levels.
  • Step S70c generating third abnormality processing information according to the third abnormality level, the third abnormality processing information including an instruction to generate warning information;
  • third abnormality handling information is generated through the third abnormality level, such as a warning alarm, reminding the operator to intervene in time and finding out the details of the abnormality.
  • the second factor that affects the product quality and efficiency of the entire production line that is, the frequency data of the state-related data
  • the two factors are comprehensively considered.
  • the yield rate and production efficiency of the product are significantly improved.
  • FIG. 4 is a schematic flowchart of a third embodiment of the closed-loop control method for AI intelligent process abnormality recognition provided by this application.
  • the state-related data is compared with the normal state data. After the step of judging the state of the solar cell, it further includes:
  • Step S80 When the solar cell is in an abnormal state, match the state-related data with the abnormal level database, and when the matching is unsuccessful, count the frequency data of the state-related data;
  • the abnormal level database is different quality defect data entered by operators based on experience in production, and the abnormal level database is a continuously improved database.
  • the frequency data needs to be considered for further judgment at this time.
  • Step S90 Obtain the second abnormality level of the state-related data by comparing the frequency data with the fault tolerance frequency;
  • a fault tolerance principle and a fourth abnormality level need to be set in the system.
  • a fault tolerance frequency is set.
  • the frequency of the abnormal state data of the minor abnormality is N times, as long as it exceeds This fault-tolerant frequency N times corresponds to an abnormality level.
  • the fault-tolerant frequency is N times, it corresponds to another abnormality level, which can be treated differently according to different abnormality levels.
  • Step S100 Generate fourth abnormality processing information according to the fourth abnormality level.
  • the fourth abnormality level is used to generate fourth abnormality handling information, such as stopping or continuing production, or directly jump out of the system dialog box and request the operator Intervention operation.
  • FIG. 5 is a flowchart of a fourth embodiment of the AI intelligent process abnormality recognition closed-loop control method provided by this application.
  • the frequency data is compared with the fault-tolerant frequency to obtain the state correlation
  • the step of the second abnormal level of the data it also includes:
  • Step S100a When the frequency data is greater than the fault-tolerant frequency, store the state-related data in the abnormality level database, and update the abnormality level database;
  • the frequency data is greater than the fault-tolerant frequency, it indicates that the state-related data is more important. At this time, the personnel are prompted to participate in the judgment, and the state-related data is stored in the abnormality level database and updated The abnormal level database makes the database more perfect.
  • Step S100b When the frequency data is less than the fault-tolerant frequency, record the state-related data and update it to the abnormal level database regularly.
  • the frequency data is less than the fault-tolerant frequency, it indicates that the state-related data is not important, and the system records the state-related data and regularly updates the abnormal level database.
  • the abnormality level database needs to be continuously improved.
  • the frequency data is greater than the fault tolerance frequency
  • the status-related data is stored in the abnormality level database, and the abnormality level database is updated
  • the frequency data is less than the fault-tolerant frequency
  • the state-related data is recorded and updated to the abnormality level database regularly, and the abnormality level database is updated in time through two different situations of emergency and non-emergency, so that the entire system Continuous learning and self-improvement to continuously improve product yield and production efficiency.
  • FIG. 6 is a schematic structural diagram of an embodiment of the AI intelligent process abnormality recognition closed-loop control equipment system provided by this application.
  • the AI intelligent process abnormal recognition closed-loop control equipment system 1000 is set as a solar cell module production line a, including a control host 100, and an identification component 200 electrically connected to the control host 100 and execution The component 300, wherein the identification component 200 is configured to obtain the state-related data of the solar cell and send the state-related data to the control host 100, and the execution component 300 is configured to receive the After the first abnormal processing information of the control host 100, exception processing is performed.
  • the control host 100 includes all the above technical solutions. Therefore, the AI intelligent process abnormal recognition closed-loop control equipment system 1000 also includes all the above technical solutions. , Also has the technical effects brought about by the above technical solutions, and will not be repeated here.
  • control host 100 is equivalent to the control end of the entire AI intelligent process abnormality recognition closed-loop control equipment system 1000
  • identification component 200 is equivalent to the entire AI intelligent process abnormality recognition closed-loop control equipment system 1000
  • execution component 300 is equivalent to the output end of the entire AI intelligent process abnormality recognition closed-loop control equipment system 1000.
  • the input end collects the relevant state-related data of the solar cell and transmits it to the control After the control terminal performs related processing, it is transmitted to the output terminal for execution.
  • the identification component 200 is equivalent to the input end of the entire AI intelligent process abnormal recognition closed-loop control equipment system 1000, and mainly collects state-related data of the solar cells. Specifically, in this embodiment, the identification component 200 It includes a plurality of identification sensors 20, which include image identification sensors 20, temperature sensors, and photoelectric sensors, which collect quality data such as images, temperature, and positioning of the solar cells, and compare the status Related data is transmitted to the control host 100.
  • identification sensors 20 include image identification sensors 20, temperature sensors, and photoelectric sensors, which collect quality data such as images, temperature, and positioning of the solar cells, and compare the status Related data is transmitted to the control host 100.
  • the solar cell module production line a has a plurality of stations, and the plurality of stations include a picking station, a transmission station, a layup station, a string welding station, a bus welding station, and a layout station.
  • the identification sensors 20 are correspondingly distributed at a plurality of the workstations, and are configured to obtain state-related data of the solar cells corresponding to the workstations, and the identification components 200 are respectively arranged at different workstations. , Can comprehensively understand the state-related data of the solar cell during the entire production process, can comprehensively monitor the production quality of the solar cell module, and use different processing strategies to deal with problems when problems occur, reducing human involvement.
  • the specific form of the execution component 300 is not limited.
  • it may be a controller that controls the operation of an active motor on a production line, or it may be a marking component that marks abnormal solar cells. 400.
  • the AI intelligent process abnormality recognition closed-loop control equipment system 1000 further includes a marking component 400 electrically connected to the control host 100, and the control host 100 is also configured to pass the state-related data Generate abnormal distribution data information and marking information, send the marking information to the marking component 400, the abnormal distribution data information is set to guide the repairer to work, the marking component 400 is set to receive the marking information, and The abnormal solar cell is marked, the marking information is convenient for the repairer to identify, and the abnormal distribution data information is set to guide the repairer to find the abnormal solar cell.
  • the marking component 400 includes a laser encoder 40, which is configured to perform coding processing on the abnormal solar cell sheet and/or the glass substrate of the solar cell component, and according to certain coding rules, Mark the abnormal solar cells, record the location and type of abnormality, etc., for the large solar cell modules, perform coding processing on the glass substrate of the solar cell module, and guide the repair personnel to perform the work , Which obviously improves the efficiency of rework.
  • a laser encoder 40 which is configured to perform coding processing on the abnormal solar cell sheet and/or the glass substrate of the solar cell component, and according to certain coding rules, Mark the abnormal solar cells, record the location and type of abnormality, etc., for the large solar cell modules, perform coding processing on the glass substrate of the solar cell module, and guide the repair personnel to perform the work , Which obviously improves the efficiency of rework.
  • the abnormal distribution data information is associated with the barcode on the glass substrate of each solar cell module and stored in the MES system.
  • the module barcode is identified, and the abnormal distribution is automatically called by the MES system.
  • the data information is displayed on the display.
  • the workers at the repair station can identify the abnormal battery slices on the one hand, and on the other hand, the abnormal distribution data information on the display is used to identify the two icons, which is more intuitive and simple. , Realize the fool-proof design, facilitate the repair of the operators, and improve the work efficiency.

Landscapes

  • Engineering & Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • General Factory Administration (AREA)
  • Photovoltaic Devices (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

Provided are an AI intelligent process abnormality recognition closed-loop control method, a host and a device system. The AI intelligent process abnormality recognition closed-loop control method comprises: receiving state-related data of a solar cell sheet in real time (S10); comparing the state-related data with normal state data, determining a state of the solar cell sheet (S20); when the solar cell sheet is in an abnormal state, obtaining a first abnormality level by means of matching the state-related data with an abnormality level database (S30); and performing a corresponding abnormality processing policy according to the first abnormality level (S40). According to the solution, a state of a cell sheet on a solar cell assembly production line is monitored intelligently and in real time; and according to different states of the cell sheet, the operation of the solar cell assembly production line is automatically guided and controlled, thereby reducing manual participation, improving the production efficiency of the solar cell assembly production line, and also improving the yield of the products.

Description

AI智能过程异常识别闭环控制方法、主机及装备*** AI intelligent process abnormal recognition closed-loop control method, host and equipment system To
相关申请Related application
本申请要求2019年02月28日的，申请号为201910153814.1，名称为“AI智能过程异常识别闭环控制方法、主机及装备***”的中国专利申请的优先权，在此将其全文引入作为参考。This application claims the priority of the Chinese patent application on February 28, 2019 with the application number 201910153814.1 titled "AI intelligent process abnormal recognition closed-loop control method, host and equipment system", the full text of which is hereby incorporated by reference.
技术领域Technical field
本申请涉及太阳能光伏组件制造技术领域，尤其涉及一种AI智能过程异常识别闭环控制方法、主机及装备***。This application relates to the technical field of solar photovoltaic module manufacturing, and in particular to a closed-loop control method, host and equipment system for AI intelligent process abnormality recognition.
背景技术Background technique
目前太阳能光伏制造和应用已经快速高速发展，近几年中国在光伏太阳能面板制造装备领域逐步成为国际的主要供应商，尤其高速自动化焊接装备领先的企业都在国内。但实际在高速焊接自动化领域，还是没有办法对过程的异常形成快速高效的闭环控制，都是需要人员介入，在线进行判断和识别，标记等。在后续返修工序都需要员工进行筛选区分。随着设备焊接速度的进一步提升，尤其达到当前的2倍或者3倍的速度之后，常规依靠人员在线操作，难度就非常高了。为此必须寻找新的设计方案来解决。作为过程主要依靠员工判断或者随机抽样，就给制程及产品带到隐患。操作人员实际都有工作积极性的个性差异，也有体能的差异，因此存在很多漏判误判或者操作出错等各种现象，对整体产品生产制造带来良率的影响。由于人员需要休息，吃饭等，对设备的连续生产也带到影响。停线将导致产出降低，浪费后道人员等待时间和成本。安排连续生产随机抽样，可能面临漏检或者出现小批量不良风险等机率。At present, solar photovoltaic manufacturing and application have developed rapidly. In recent years, China has gradually become a major international supplier in the field of photovoltaic solar panel manufacturing equipment, especially the leading high-speed automated welding equipment companies are all located in China. However, in the field of high-speed welding automation, there is still no way to form a fast and efficient closed-loop control of abnormal processes. It requires human intervention, online judgment, identification, and marking. In the subsequent repair process, employees are required to screen and distinguish. With the further improvement of equipment welding speed, especially after reaching the current speed of 2 times or 3 times, it is very difficult to rely on online operation by personnel. To this end, we must find new design solutions to solve. The process mainly relies on employee judgment or random sampling, which brings hidden dangers to the process and products. Operators actually have personality differences in work enthusiasm, as well as differences in physical fitness. Therefore, there are many phenomena such as missed judgments, misjudgments or operating errors, which have an impact on the overall product manufacturing yield. Because personnel need to rest, eat, etc., it also affects the continuous production of equipment. Stopping the line will result in a reduction in output, and waste of waiting time and costs for downstream personnel. Arranging random sampling for continuous production may face the possibility of missed inspection or the risk of small batch defects.
在当前光伏太阳能面板制造领域，实际自动化焊接过程面临的不良必须返修，这是有原材料，工艺，设备，多方面因素形成的，因此返修也是各家都必须面临的制程挑战。当前返修操作人员，就需要面临大量的重复性的工作任务，每片异常的单元标记和区分就比较重要。实际当前都是依靠自动化焊接工序操作人员的在线判断和操作来完成，但在线速度快和操作也不方便，标记难以细化和精准，需要返修员工进一步的筛选再次识别。In the current photovoltaic solar panel manufacturing field, the defects faced by the actual automated welding process must be repaired. This is formed by raw materials, processes, equipment, and many factors. Therefore, repairing is also a process challenge that everyone must face. The current repair operators need to face a large number of repetitive tasks, and the marking and distinguishing of each abnormal unit is more important. Actually, it is currently completed by the online judgment and operation of the operators of the automated welding process, but the online speed is fast and the operation is inconvenient, and the marking is difficult to be refined and accurate. It requires further screening and re-identification by the repairing staff.
上述内容仅设置为辅助理解本申请的技术方案，并不代表承认上述内容是现有技术。The above content is only set to assist the understanding of the technical solution of this application, and does not mean that the above content is recognized as prior art.
发明内容Summary of the invention
本申请主要目的在于提供一种AI智能（人工智能，Artificial Intelligence）过程异常识别闭环控制方法、主机及装备***，旨在解决现有技术中人工操作过程中不及时、难以快速识别、快速标记而导致良品率低以及生产效率低下的问题。The main purpose of this application is to provide an AI (Artificial Intelligence, Artificial Intelligence) Intelligence) Process anomaly identification closed-loop control method, host computer and equipment system are designed to solve the problems of low yield and low production efficiency due to untimely, difficult to quickly identify and fast mark in the manual operation process in the prior art.
为实现上述目的，本申请提供一种AI智能过程异常识别闭环控制方法、主机及装备***，其中，所述AI智能过程异常识别闭环控制方法包括以下步骤：In order to achieve the above objective, this application provides a closed-loop control method, host and equipment system for AI intelligent process abnormality recognition, wherein the AI intelligent process abnormality recognition closed-loop control method includes the following steps:
实时接收太阳能电池片的状态相关数据；Receive real-time data related to the status of solar cells;
将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态；Comparing the state-related data with normal state data to determine the state of the solar cell;
当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，获得第一异常等级；When the solar cell is in an abnormal state, matching the state-related data with the abnormal level database to obtain the first abnormal level;
根据所述第一异常等级进行相应的异常处理策略。A corresponding exception handling strategy is performed according to the first exception level.
可选地，所述第一异常等级包括严重异常、主要异常和次要异常；Optionally, the first abnormality level includes severe abnormality, major abnormality, and minor abnormality;
根据所述第一异常等级进行相应的异常处理策略的步骤中：In the steps of performing a corresponding exception handling strategy according to the first exception level:
当所述第一异常等级为严重异常时，所述异常处理策略包括生成停机信息指令；When the first abnormality level is a serious abnormality, the abnormality handling strategy includes generating a shutdown information instruction;
当所述第一异常等级为主要异常时，所述异常处理策略包括：When the first abnormality level is a major abnormality, the abnormality handling strategy includes:
统计第一时间段内所述状态相关数据出现的第一频次数据；Collect statistics on the first frequency of occurrence of the state-related data in the first time period;
通过所述异常等级数据库内预存的内部设备和/或产品出现严重事风险几率计算出所述第一时间段内的第一容错频次；Calculating the first fault-tolerant frequency in the first time period according to the risk probability of serious incidents of internal equipment and/or products prestored in the abnormal level database;
通过所述第一频次数据与所述第一容错频次进行对比，获得所述状态相关数据的第二异常等级；Obtaining the second abnormality level of the state-related data by comparing the first frequency data with the first fault tolerance frequency;
通过所述第二异常等级生成第二异常处理信息，所述第二异常处理信息包括生成停机信息指令；Generating second abnormality handling information through the second abnormality level, where the second abnormality handling information includes an instruction to generate shutdown information;
当所述第一异常等级为次要异常时，所述异常处理策略包括：When the first abnormality level is a secondary abnormality, the abnormality handling strategy includes:
统计第二时间段内所述状态相关数据出现的第二频次数据；Counting the second frequency data of the state-related data in the second time period;
通过所述异常等级数据库内预存的内部设备和/或产品出现重复性批量事故几率计算出所述第二时间段内的第二容错频次；Calculating the second fault-tolerant frequency in the second time period according to the probability of repetitive batch accidents of internal equipment and/or products pre-stored in the abnormal level database;
通过所述第二频次数据与所述第二容错频次进行对比，获得所述状态相关数据的第三异常等级；Obtaining the third abnormality level of the state-related data by comparing the second frequency data with the second fault tolerance frequency;
通过所述第三异常等级生成第三异常处理信息，所述第三异常处理信息包括生成警示报警信息指令。The third abnormality processing information is generated according to the third abnormality level, and the third abnormality processing information includes an instruction to generate warning information.
可选地，在所述将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态的步骤之后，还包括：Optionally, after the step of comparing the state-related data with normal state data to determine the state of the solar cell, the method further includes:
当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，匹配不成功时，统计所述状态相关数据出现的频次数据；When the solar cell is in an abnormal state, the state-related data is matched with the abnormal level database, and when the matching is unsuccessful, the frequency data of the state-related data is counted;
通过所述频次数据与容错频次进行对比，获得所述状态相关数据第四异常等级；Obtaining the fourth abnormality level of the state-related data by comparing the frequency data with the fault tolerance frequency;
通过所述第四异常等级生成第四异常处理信息。The fourth abnormality processing information is generated by the fourth abnormality level.
可选地，通过所述频次数据与容错频次进行对比，获得所述状态相关数据的第四异常等级的步骤之后，还包括：Optionally, after the step of obtaining the fourth abnormality level of the state-related data by comparing the frequency data with the fault tolerance frequency, the method further includes:
当所述频次数据大于所述容错频次时，将所述状态相关数据存入至所述异常等级数据库，并更新所述异常等级数据库；When the frequency data is greater than the fault tolerance frequency, store the state-related data in the abnormality level database, and update the abnormality level database;
当所述频次数据小于所述容错频次时，记录所述状态相关数据，并定期更新至所述异常等级数据库。When the frequency data is less than the fault-tolerant frequency, the state-related data is recorded and updated to the abnormal level database regularly.
本申请还提供一种控制主机，所述控制主机包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的AI智能过程异常识别闭环控制程序，所述AI智能过程异常识别闭环控制程序配置为实现AI智能过程异常识别闭环控制方法的步骤，所述AI智能过程异常识别闭环控制方法包括：The present application also provides a control host, the control host including: a memory, a processor, and an AI intelligent process abnormality recognition closed-loop control program stored on the memory and running on the processor, the AI intelligent process The abnormality recognition closed-loop control program is configured to implement the steps of the AI intelligent process abnormality recognition closed-loop control method, and the AI intelligent process abnormality recognition closed-loop control method includes:
实时接收太阳能电池片的状态相关数据；Receive real-time data related to the status of solar cells;
将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态；Comparing the state-related data with normal state data to determine the state of the solar cell;
当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，获得第一异常等级；When the solar cell is in an abnormal state, matching the state-related data with the abnormal level database to obtain the first abnormal level;
根据所述第一异常等级进行相应的异常处理策略。A corresponding exception handling strategy is performed according to the first exception level.
本申请还提供一种AI智能过程异常识别闭环控制装备***，设置为太阳能电池组件生产线，所述AI智能过程异常识别闭环控制装备***包括控制主机、以及与所述控制主机电性连接的识别组件和执行组件，其中：This application also provides an AI intelligent process abnormality recognition closed-loop control equipment system, which is set as a solar cell module production line. The AI intelligent process abnormality recognition closed-loop control equipment system includes a control host and an identification component electrically connected to the control host And executive components, where:
所述识别组件设置为获取所述太阳能电池片状态相关数据，并将所述状态相关数据发送至所述控制主机；The identification component is configured to obtain state-related data of the solar cell, and send the state-related data to the control host;
所述执行组件，设置为在接收到所述控制主机的第一异常处理信息后，进行异常处理；The execution component is configured to perform abnormal processing after receiving the first abnormal processing information of the control host;
所述控制主机包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的AI智能过程异常识别闭环控制程序，所述AI智能过程异常识别闭环控制程序配置为实现AI智能过程异常识别闭环控制方法的步骤，所述AI智能过程异常识别闭环控制方法包括：The control host includes: a memory, a processor, and an AI intelligent process abnormality recognition closed-loop control program stored on the memory and running on the processor, and the AI intelligent process abnormality recognition closed-loop control program is configured to realize AI The steps of a closed-loop control method for intelligent process abnormality recognition, the AI intelligent process abnormality recognition closed-loop control method includes:
实时接收太阳能电池片的状态相关数据；Receive real-time data related to the status of solar cells;
将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态；Comparing the state-related data with normal state data to determine the state of the solar cell;
当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，获得第一异常等级；When the solar cell is in an abnormal state, matching the state-related data with the abnormal level database to obtain the first abnormal level;
根据所述第一异常等级进行相应的异常处理策略。A corresponding exception handling strategy is performed according to the first exception level.
可选地，所述识别组件包括多个识别传感器，多个所述识别传感器包括图像识别传感器、温度传感器、光电传感器。Optionally, the identification component includes a plurality of identification sensors, and the plurality of identification sensors includes an image identification sensor, a temperature sensor, and a photoelectric sensor.
可选地，所述太阳能电池组件生产线上具有多个工位，多个所述工位包括取料工位、传输工位、铺片工位、串焊工位、汇流焊工位和排版工位；Optionally, the solar cell module production line has a plurality of workstations, and the plurality of workstations include a picking station, a transmission station, a layup station, a string welding station, a bus welding station, and a layout station;
多个所述识别传感器对应分布于多个所述工位处，设置为获取对应所述工位处的所述太阳能电池片的状态相关数据。A plurality of the identification sensors are correspondingly distributed at a plurality of the workstations, and are configured to obtain the state-related data of the solar cells at the corresponding workstations.
可选地，还包括与所述控制主机电性连接的标记组件；Optionally, it further includes a marking component electrically connected to the control host;
所述控制主机还设置为通过所述状态相关数据生成异常分布数据信息和标记信息，将所述标记信息发送至所述标记组件，所述异常分布数据信息设置为供指导返修人员作业；The control host is also configured to generate abnormal distribution data information and marking information from the state-related data, and send the marking information to the marking component, and the abnormal distribution data information is set to guide the work of the repairer;
所述标记组件设置为接收所述标记信息，并对异常的所述太阳能电池片进行标记处理。The marking component is configured to receive the marking information and perform marking processing on the abnormal solar cell.
可选地，所述标记组件包括激光打码器，设置为对异常的所述太阳能电池片和/或所述太阳能电池组件的玻璃基板进行喷码处理。Optionally, the marking component includes a laser coder, configured to perform coding processing on the abnormal solar cell sheet and/or the glass substrate of the solar cell component.
本申请通过实时接收太阳能电池片的状态相关数据，将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态，当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，获得第一异常等级，根据所述第一异常等级进行相应的异常处理策略，能智能实时监控所述太阳能电池组件生产线上的电池片的状态，根据电池片的不同状态，自动指导控制所述太阳能电池组件生产线的运转，减少了人为的参与，提高了所述太阳能电池组件生产线的生产效率，也提高了产品的良品率。This application receives state-related data of solar cells in real time and compares the state-related data with normal state data to determine the state of the solar cells. When the solar cells are in an abnormal state, pass the state The relevant data is matched with the abnormality level database to obtain the first abnormality level, and the corresponding abnormality processing strategy is carried out according to the first abnormality level, which can intelligently monitor the status of the solar cell module production line in real time, according to the In different states, the operation of the solar cell module production line is automatically guided and controlled, which reduces human involvement, improves the production efficiency of the solar cell module production line, and also improves the product yield.
附图说明Description of the drawings
图1是本申请实施例方案涉及的硬件运行环境的服务器结构示意图；FIG. 1 is a schematic diagram of a server structure of a hardware operating environment involved in a solution of an embodiment of the present application;
图2为本申请AI智能过程异常识别闭环控制方法第一实施例的流程示意图；2 is a schematic flowchart of a first embodiment of a closed-loop control method for anomaly recognition of AI intelligent process according to the application;
图3为本申请AI智能过程异常识别闭环控制方法第二实施例的流程示意图；3 is a schematic flowchart of a second embodiment of a closed-loop control method for identifying abnormalities in an AI intelligent process according to this application;
图4为本申请AI智能过程异常识别闭环控制方法第三实施例的流程示意图；4 is a schematic flowchart of a third embodiment of a closed-loop control method for anomaly recognition of AI intelligent process according to the application;
图5为本申请AI智能过程异常识别闭环控制方法第四实施例的流程示意图；FIG. 5 is a schematic flowchart of a fourth embodiment of a closed-loop control method for identifying abnormalities in an AI intelligent process according to the application;
图6为本申请AI智能过程异常识别闭环控制***的一实施例的结构示意图；Fig. 6 is a schematic structural diagram of an embodiment of a closed-loop control system for identifying abnormalities in AI intelligent process of the application;
图7为本申请中应用有图6中AI智能过程异常识别闭环控制***的太阳能电池组件生产线（局部结构）一实施例的结构示意图；FIG. 7 is a schematic structural diagram of an embodiment of a solar cell module production line (partial structure) applying the AI intelligent process abnormality recognition closed-loop control system in FIG. 6 in this application;
图8为图6中AI智能过程异常识别闭环控制***的识别组件的一实施例的结构示意图；8 is a schematic structural diagram of an embodiment of an identification component of the AI intelligent process abnormality identification closed-loop control system in FIG. 6;
图9为图6中AI智能过程异常识别闭环控制***的标记组件的一实施例的结构示意图。FIG. 9 is a schematic structural diagram of an embodiment of the marking component of the closed-loop control system for AI intelligent process abnormality recognition in FIG. 6.
附图标号说明：
标号 名称 标号 名称
1000 AI智能过程异常识别闭环控制装备*** 20 识别传感器
a 太阳能电池组件生产线 300 执行组件
100 控制主机 400 标记组件
200 识别组件 40 激光打码器
Description with icon number:
Label name Label name
1000 AI intelligent process abnormal recognition closed-loop control equipment system 20 Recognition sensor
a Solar cell module production line 300 Executive component
100 Control host 400 Marking components
200 Identify components 40 Laser Coder
本申请目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。The realization, functional characteristics, and advantages of the purpose of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings.
具体实施方式detailed description
应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不设置为限定本申请。It should be understood that the specific embodiments described here are only used to explain the application, and are not set to limit the application.
参照图1，图1为本申请控制主机的结构示意图。Referring to Figure 1, Figure 1 is a schematic diagram of the structure of the control host of this application.
如图1所示，该控制主机可以包括：处理器1001，例如CPU，通信总线1002、用户接口1003，网络接口1004，存储器1005。其中，通信总线1002设置为实现这些组件之间的连接通信。用户接口1003可以包括显示屏（Display）、输入单元比如键盘（Keyboard），可选用户接口1003还可以包括标准的有线接口、无线接口。网络接口1004可选的可以包括标准的有线接口、无线接口（如WI-FI接口）。存储器1005可以是高速RAM存储器，也可以是稳定的存储器（non-volatile memory），例如磁盘存储器。存储器1005可选的还可以是独立于前述处理器1001的存储装置。As shown in FIG. 1, the control host may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is configured to realize connection and communication between these components. The user interface 1003 may include a display screen (Display) and an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 can be a high-speed RAM memory or a stable memory (non-volatile memory), such as disk storage. Optionally, the memory 1005 may also be a storage device independent of the foregoing processor 1001.
本领域技术人员可以理解，图1中示出的结构并不构成对控制主机的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。Those skilled in the art can understand that the structure shown in FIG. 1 does not constitute a limitation on the control host, and may include more or less components than shown in the figure, or a combination of certain components, or different component arrangements.
如图1所示，作为一种计算机存储介质的存储器1005中可以包括操作***、网络通信模块、用户接口模块以及AI智能过程异常识别闭环控制主机100的控制程序。As shown in FIG. 1, the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a control program of the AI intelligent process abnormality recognition closed-loop control host 100.
在图1所示的服务器中，网络接口1004主要设置为连接终端设备，与终端设备进行数据通信；用户接口1003主要设置为接收管理员的输入指令；所述服务器通过处理器1001调用存储器1005中存储的AI智能过程异常识别闭环控制主机100的控制程序，并执行以下操作：In the server shown in FIG. 1, the network interface 1004 is mainly set to connect to terminal devices and perform data communication with the terminal devices; the user interface 1003 is mainly set to receive input instructions from the administrator; the server calls the memory 1005 through the processor 1001 The stored AI intelligent process abnormally recognizes the control program of the closed-loop control host 100, and performs the following operations:
实时接收太阳能电池片的状态相关数据；Receive real-time data related to the status of solar cells;
将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态；Comparing the state-related data with normal state data to determine the state of the solar cell;
当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，获得第一异常等级；When the solar cell is in an abnormal state, matching the state-related data with the abnormal level database to obtain the first abnormal level;
根据所述第一异常等级进行相应的异常处理策略。A corresponding exception handling strategy is performed according to the first exception level.
进一步地，处理器1001可以调用存储器1005中存储的AI智能过程异常识别闭环控制主机100的控制程序，还执行以下操作：Further, the processor 1001 may call the control program of the AI intelligent process abnormal recognition closed-loop control host 100 stored in the memory 1005, and also perform the following operations:
所述第一异常等级包括严重异常、主要异常和次要异常；The first abnormality level includes severe abnormality, major abnormality and minor abnormality;
根据所述第一异常等级进行相应的异常处理策略的步骤中：In the steps of performing a corresponding exception handling strategy according to the first exception level:
当所述第一异常等级为严重异常时，所述异常处理策略包括生成停机信息指令；When the first abnormality level is a serious abnormality, the abnormality handling strategy includes generating a shutdown information instruction;
当所述第一异常等级为主要异常时，所述异常处理策略包括：When the first abnormality level is a major abnormality, the abnormality handling strategy includes:
统计第一时间段内所述状态相关数据出现的第一频次数据；Collect statistics on the first frequency of occurrence of the state-related data in the first time period;
通过所述异常等级数据库内预存的内部设备和/或产品出现严重事风险几率计算出所述第一时间段内的第一容错频次；Calculating the first fault-tolerant frequency in the first time period according to the risk probability of serious incidents of internal equipment and/or products prestored in the abnormal level database;
通过所述第一频次数据与所述第一容错频次进行对比，获得所述状态相关数据的第二异常等级；Obtaining the second abnormality level of the state-related data by comparing the first frequency data with the first fault tolerance frequency;
通过所述第二异常等级生成第二异常处理信息，所述第二异常处理信息包括生成停机信息指令；Generating second abnormality handling information through the second abnormality level, where the second abnormality handling information includes an instruction to generate shutdown information;
当所述第一异常等级为次要异常时，所述异常处理策略包括：When the first abnormality level is a secondary abnormality, the abnormality handling strategy includes:
统计第二时间段内所述状态相关数据出现的第二频次数据；Counting the second frequency data of the state-related data in the second time period;
通过所述异常等级数据库内预存的内部设备和/或产品出现重复性批量事故几率计算出所述第二时间段内的第二容错频次；Calculating the second fault-tolerant frequency in the second time period according to the probability of repetitive batch accidents of internal equipment and/or products pre-stored in the abnormal level database;
通过所述第二频次数据与所述第二容错频次进行对比，获得所述状态相关数据的第三异常等级；Obtaining the third abnormality level of the state-related data by comparing the second frequency data with the second fault tolerance frequency;
通过所述第三异常等级生成第三异常处理信息，所述第三异常处理信息包括生成警示报警信息指令。The third abnormality processing information is generated according to the third abnormality level, and the third abnormality processing information includes an instruction to generate warning information.
进一步地，处理器1001可以调用存储器1005中存储的AI智能过程异常识别闭环控制主机100的控制程序，还执行以下操作：Further, the processor 1001 may call the control program of the AI intelligent process abnormal recognition closed-loop control host 100 stored in the memory 1005, and also perform the following operations:
在所述将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态的步骤之后，还包括：After the step of comparing the state-related data with normal state data to determine the state of the solar cell, the method further includes:
当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，匹配不成功时，统计所述状态相关数据出现的频次数据；When the solar cell is in an abnormal state, the state-related data is matched with the abnormal level database, and when the matching is unsuccessful, the frequency data of the state-related data is counted;
通过所述频次数据与容错频次进行对比，获得所述状态相关数据第四异常等级；Obtaining the fourth abnormality level of the state-related data by comparing the frequency data with the fault tolerance frequency;
通过所述第四异常等级生成第四异常处理信息。The fourth abnormality processing information is generated by the fourth abnormality level.
进一步地，处理器1001可以调用存储器1005中存储的AI智能过程异常识别闭环控制主机100的控制程序，还执行以下操作：Further, the processor 1001 may call the control program of the AI intelligent process abnormal recognition closed-loop control host 100 stored in the memory 1005, and also perform the following operations:
通过所述频次数据与容错频次进行对比，获得所述状态相关数据的第四异常等级的步骤之后，还包括：After the step of obtaining the fourth abnormality level of the state-related data by comparing the frequency data with the fault tolerance frequency, the method further includes:
当所述频次数据大于所述容错频次时，将所述状态相关数据存入至所述异常等级数据库，并更新所述异常等级数据库；When the frequency data is greater than the fault tolerance frequency, store the state-related data in the abnormality level database, and update the abnormality level database;
当所述频次数据小于所述容错频次时，记录所述状态相关数据，并定期更新至所述异常等级数据库。When the frequency data is less than the fault-tolerant frequency, the state-related data is recorded and updated to the abnormal level database regularly.
本申请通过实时接收太阳能电池片的状态相关数据，将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态，当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，获得第一异常等级，根据所述第一异常等级进行相应的异常处理策略，能智能实时监控所述太阳能电池组件生产线上的电池片的状态，根据电池片的不同状态，自动指导控制所述太阳能电池组件生产线的运转，减少了人为的参与，提高了所述太阳能电池组件生产线的生产效率，也提高了产品的良品率。This application receives state-related data of solar cells in real time and compares the state-related data with normal state data to determine the state of the solar cells. When the solar cells are in an abnormal state, pass the state The relevant data is matched with the abnormality level database to obtain the first abnormality level, and the corresponding abnormality processing strategy is carried out according to the first abnormality level, which can intelligently monitor the status of the solar cell module production line in real time, according to the In different states, the operation of the solar cell module production line is automatically guided and controlled, which reduces human involvement, improves the production efficiency of the solar cell module production line, and also improves the product yield.
基于上述硬件结构，图2至图6为本申请提供的AI智能过程异常识别闭环控制方法的实施例。Based on the above hardware structure, FIGS. 2 to 6 are embodiments of the closed-loop control method for AI intelligent process abnormality recognition provided by this application.
请参照图2，图2为本申请提供的AI智能过程异常识别闭环控制方法第一实施例的流程示意图，在本实施例中，所述AI智能过程异常识别闭环控制方法包括以下步骤：Please refer to Figure 2. Figure 2 is a schematic flowchart of the first embodiment of the AI intelligent process abnormality recognition closed-loop control method provided by this application. In this embodiment, the AI intelligent process abnormality recognition closed-loop control method includes the following steps:
步骤S10、实时接收太阳能电池片的状态相关数据；Step S10, receiving real-time state-related data of the solar cell;
需要说明的是，在太阳能电池组件的整个生产流程中，所述太阳能电池片的状态相关数据包括图像数据，温度数据，焊接成型质量数据，定位数据等等相关数据，以上数据都与所述太阳能电池片的质量相关，本实施例中，不限制以上的状态数据，还可以包括很多状态数据。It should be noted that in the entire production process of solar cell modules, the state-related data of the solar cell includes image data, temperature data, welding forming quality data, positioning data and other related data. The above data are all related to the solar cell. The quality of the battery slice is related. In this embodiment, the above status data is not limited, and a lot of status data can also be included.
步骤S20、将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态；Step S20, comparing the state-related data with normal state data, and judging the state of the solar cell;
需要说明的是，所述太阳能电池片的正常状态数据全部存储至所述控制***中，包括正常电池片的图像数据，温度数据，焊接成型质量数据，定位数据等等相关数据，当实时接收太阳能电池片的状态相关数据后，将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态，如果是正常状态，则该被检测的所述太阳能电池片为正常状态，整个生产线正常运转。It should be noted that the normal state data of the solar cells are all stored in the control system, including image data of normal cells, temperature data, welding quality data, positioning data and other related data. When receiving solar energy in real time After the state-related data of the cell, the state-related data is compared with the normal state data to determine the state of the solar cell. If it is a normal state, the detected solar cell is in a normal state. The production line is operating normally.
步骤S30、当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，获得第一异常等级；Step S30: When the solar cell is in an abnormal state, match the state-related data with an abnormal level database to obtain a first abnormal level;
需要说明的是，在***中会建立一个异常等级数据库，所述异常等级数据库将不同的异常缺陷分类，根据不同的类别，按一预设的规则进行异常等级分类，例如，可以以映射表的形式，如果被检测的所述太阳能电池片为非正常状态，则整个***判断被检测的所述太阳能电池片为异常状态，查找所述异常状态配对的异常类别，进而得到对应的所述异常等级。It should be noted that an abnormality level database will be established in the system. The abnormality level database classifies different abnormal defects. According to different categories, the abnormality level is classified according to a preset rule. For example, it can be classified as a mapping table. Form, if the detected solar cell is in an abnormal state, the entire system determines that the detected solar cell is in an abnormal state, searches for the abnormality category paired with the abnormal state, and then obtains the corresponding abnormality level .
步骤S40、根据所述第一异常等级进行相应的异常处理策略；Step S40: Perform a corresponding abnormality processing strategy according to the first abnormality level;
需要说明的是，为了使得整个所述太阳能电池组件生产线a能实现高效生产，需要根据不同的异常等级来选择不同的异常处理策略，对不同的异常进行区别处理，可以获得更精细化的控制。It should be noted that, in order to enable the entire solar cell module production line a to achieve high-efficiency production, different exception handling strategies need to be selected according to different exception levels, and different exceptions are treated differently to obtain more refined control.
本申请通过实时接收太阳能电池片的状态相关数据，将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态，当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，获得第一异常等级，根据所述第一异常等级进行相应的异常处理策略，能智能实时监控所述太阳能电池组件生产线a上的电池片的状态，根据电池片的不同状态，自动指导控制所述太阳能电池组件生产线a的运转，减少了人为的参与，提高了所述太阳能电池组件生产线a的生产效率，也提高了产品的良品率。This application receives state-related data of solar cells in real time and compares the state-related data with normal state data to determine the state of the solar cells. When the solar cells are in an abnormal state, pass the state The relevant data is matched with the abnormality level database to obtain the first abnormality level, and the corresponding abnormality processing strategy is performed according to the first abnormality level, which can intelligently monitor the status of the cells on the solar cell module production line a in real time. Under different conditions, the operation of the solar cell module production line a is automatically guided and controlled, which reduces human involvement, improves the production efficiency of the solar cell module production line a, and also improves the yield of products.
请参照图3，图3为本申请提供的AI智能过程异常识别闭环控制方法第二实施例的流程示意图，在本实施例中，所述第一异常等级包括严重异常、主要异常和次要异常，根据所述第一异常等级进行相应的异常处理策略的步骤中：Please refer to FIG. 3, which is a schematic flowchart of the second embodiment of the AI intelligent process abnormality recognition closed-loop control method provided by this application. In this embodiment, the first abnormality level includes severe abnormality, major abnormality, and minor abnormality. , In the steps of performing a corresponding exception handling strategy according to the first exception level:
步骤S40a、当所述第一异常等级为严重异常时，所述异常处理策略包括生成停机信息指令；Step S40a: When the first abnormality level is a serious abnormality, the abnormality handling strategy includes generating a shutdown information instruction;
需要说明的是，将所述第一异常等级分类为严重异常、主要异常和次要异常三个级别，当然，为了更精细的对所述生产线进行控制，还可以是将所述第一异常等级区分多个异常等级，使得对所述生产线的监控更加精细，例如，出现严重的破片，而且面积较大时，严重影响整个产品的生产质量，确定为严重异常，一般的异常确定为主要异常，当仅使出现轻微的刮痕时，对整个产品的生产质量无影响或者影响较小时，确定为次要异常。It should be noted that the first abnormality level is classified into three levels: severe abnormality, major abnormality, and minor abnormality. Of course, in order to control the production line more finely, the first abnormality level may also be Differentiating multiple abnormal levels makes the monitoring of the production line more detailed. For example, when severe fragments occur and the area is large, the production quality of the entire product is seriously affected, and it is determined as a serious abnormality, and a general abnormality is determined as a major abnormality. When only slight scratches appear, and the production quality of the entire product is not affected or has a small effect, it is determined as a minor abnormality.
步骤S40b、当所述第一异常等级为主要异常时，所述异常处理策略包括第一步骤为统计第一时间段内所述状态相关数据出现的第一频次数据；Step S40b: When the first abnormality level is a major abnormality, the abnormality handling strategy includes the first step of counting the first frequency data of the state-related data in the first time period;
需要说明的是，虽然所述第一异常等级为主要异常，但是如果该主要异常的异常出现的频率过高，往往是代表设备出现严重故障，或者材料出现严重故障，也会严重影响整个产品的生产质量，此时需要统计第一时间段内所述状态相关数据出现的第一频次数据，来进一步地判断。It should be noted that although the first abnormality level is the main abnormality, if the frequency of the main abnormality is too high, it often means that the equipment has a serious failure or the material has a serious failure, which will also seriously affect the entire product. Production quality, at this time, it is necessary to count the first frequency data of the state-related data in the first time period for further judgment.
步骤S50b、通过所述异常等级数据库内预存的内部设备和/或产品出现严重事风险几率计算出所述第一时间段内的第一容错频次；Step S50b: Calculate the first fault-tolerant frequency in the first time period based on the risk probability of serious incidents of internal equipment and/or products pre-stored in the abnormal level database;
需要说明的是，内部设备和/或产品出现严重事风险几率的数据存储在所述异常等级数据库内，根据该几率可以计算出有效工作时间内异常出现的第一容错频次，所述第一容错频次作为一个判断的参考，如一段时间内，所述次要异常的异常状态数据的频次为N次，只要是超过这个容错频次N次，则对应为一个异常等级，便于对比判断；It should be noted that the data on the probability of serious incidents of internal equipment and/or products is stored in the abnormality level database. Based on the probability, the first fault-tolerant frequency of abnormal occurrence during the effective working time can be calculated. The frequency is used as a reference for judgment. For example, the frequency of the abnormal state data of the minor abnormality is N times within a period of time. As long as the frequency exceeds the fault tolerance frequency N times, it corresponds to an abnormal level, which is convenient for comparison and judgment;
步骤S60b、通过所述第一频次数据与所述第一容错频次进行对比，获得所述状态相关数据的第二异常等级；Step S60b: Obtain a second abnormality level of the state-related data by comparing the first frequency data with the first fault tolerance frequency;
需要说明的是，如一段时间内，所述主要异常的异常状态数据的频次为N次，只要是超过这个容错频次N次，则对应为一个异常等级，当在所述容错频次N次之内的，对应为另一个异常等级，根据不同的异常等级可以区别处理。It should be noted that, for a period of time, the frequency of the abnormal state data of the main abnormality is N times, as long as the frequency of fault tolerance exceeds the frequency of N times, it corresponds to an abnormal level, and when the frequency of fault tolerance is within the frequency of N times Corresponding to another abnormality level, which can be handled differently according to different abnormality levels.
步骤S70b、所述第二异常处理信息包括生成停机信息指令；Step S70b: The second abnormal handling information includes an instruction to generate shutdown information;
需要说明的是，判断所述状态相关数据的第二异常等级后，通过所述第二异常等级生成第二异常处理信息，如停机或者继续生产等等。It should be noted that after the second abnormality level of the state-related data is determined, second abnormality handling information is generated through the second abnormality level, such as stopping or continuing production.
步骤S40c、当所述第一异常等级为次要异常时，所述异常处理策略包括第一步骤为统计第二时间段内所述状态相关数据出现的第二频次数据；Step S40c: When the first abnormality level is a secondary abnormality, the abnormality processing strategy includes the first step of counting the second frequency data of the state-related data in the second time period;
需要说明的是，虽然所述第一异常等级为次要异常，但是如果该次要异常的异常出现的频率过高，也会影响整个产品的生产质量，此时需要统计所述状态相关数据出现的频次数据，来进一步地判断。It should be noted that although the first abnormality level is a secondary abnormality, if the frequency of the secondary abnormality is too high, it will also affect the production quality of the entire product. At this time, it is necessary to count the occurrence of the state-related data. The frequency data for further judgment.
步骤S50c、通过所述异常等级数据库内预存的内部设备和/或产品出现重复性批量事故几率计算出所述第二时间段内的第二容错频次；Step S50c: Calculate the second fault tolerance frequency in the second time period based on the repetitive batch accident probability of internal equipment and/or products pre-stored in the abnormal level database;
需要说明的是，内部设备和/或产品出现重复性批量事故几率的数据存储在所述异常等级数据库内，根据该几率可以计算出有效工作时间内异常出现的第二容错频次，所述第二容错频次作为一个判断的参考，如一段时间内，所述次要异常的异常状态数据的频次为N次，便于对比判断；It should be noted that the data on the probability of repetitive batch accidents of internal equipment and/or products is stored in the abnormal level database. Based on the probability, the second fault-tolerant frequency of abnormal occurrence during effective working hours can be calculated. The fault tolerance frequency is used as a reference for judgment. For example, the frequency of the abnormal state data of the minor abnormality is N times within a period of time, which is convenient for comparison and judgment;
步骤S60c、通过所述第二频次数据与所述第二容错频次进行对比，获得所述状态相关数据的第三异常等级；Step S60c, comparing the second frequency data with the second fault tolerance frequency to obtain the third abnormality level of the state-related data;
需要说明的是，如一段时间内，所述次要异常的异常状态数据的频次为N次，只要是超过这个容错频次N次，则对应为一个异常等级，当在所述容错频次N次之内的，对应为另一个异常等级，根据不同的异常等级可以区别处理。It should be noted that, for example, the frequency of the abnormal state data of the minor abnormality is N times within a period of time. As long as the frequency of fault tolerance exceeds this frequency of N times, it corresponds to an abnormal level. When the frequency of fault tolerance is N times The inside corresponds to another abnormal level, which can be handled differently according to different abnormal levels.
步骤S70c、通过所述第三异常等级生成第三异常处理信息，所述第三异常处理信息包括生成警示报警信息指令；Step S70c, generating third abnormality processing information according to the third abnormality level, the third abnormality processing information including an instruction to generate warning information;
需要说明的是，判断所述状态相关数据的第三异常等级后，通过所述第三异常等级生成第三异常处理信息，如警示报警，提醒作业人员的及时介入，查明异常详细情况。It should be noted that after judging the third abnormality level of the state-related data, third abnormality handling information is generated through the third abnormality level, such as a warning alarm, reminding the operator to intervene in time and finding out the details of the abnormality.
本申请中，考虑到所述状态相关数据的第一异常等级后，综合考虑到影响整个生产线产品质量和效率的第二个因素，即状态相关数据出现的频次数据，将两个因数综合考量，显著地提高了所述产品的良品率及生产效率。In this application, after considering the first abnormal level of the state-related data, the second factor that affects the product quality and efficiency of the entire production line, that is, the frequency data of the state-related data, is comprehensively considered, and the two factors are comprehensively considered. The yield rate and production efficiency of the product are significantly improved.
请参照图4，图4为本申请提供的AI智能过程异常识别闭环控制方法第三实施例的流程示意图，在本实施例中，在所述将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态的步骤之后，还包括：Please refer to FIG. 4, which is a schematic flowchart of a third embodiment of the closed-loop control method for AI intelligent process abnormality recognition provided by this application. In this embodiment, the state-related data is compared with the normal state data. After the step of judging the state of the solar cell, it further includes:
步骤S80、当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，匹配不成功时，统计所述状态相关数据出现的频次数据；Step S80: When the solar cell is in an abnormal state, match the state-related data with the abnormal level database, and when the matching is unsuccessful, count the frequency data of the state-related data;
需要说明的是，本实施例中，所述异常等级数据库是作业人员根据生产中的经验录入的不同的质量缺陷数据，所述异常等级数据库是一个不断完善的数据库，当出现了所述状态相关数据与异常等级数据库匹配不成功的情形时，此时也需要考虑到频次数据，来进一步地判断。It should be noted that, in this embodiment, the abnormal level database is different quality defect data entered by operators based on experience in production, and the abnormal level database is a continuously improved database. When the data is not successfully matched with the abnormal level database, the frequency data needs to be considered for further judgment at this time.
步骤S90、通过所述频次数据与容错频次进行对比，获得所述状态相关数据第二异常等级；Step S90: Obtain the second abnormality level of the state-related data by comparing the frequency data with the fault tolerance frequency;
需要说明的是，***中需要设定一个容错原则与第四异常等级，例如，设定一个容错频次，如一段时间内，所述次要异常的异常状态数据的频次为N次，只要是超过这个容错频次N次，则对应为一个异常等级，当在所述容错频次N次之内的，对应为另一个异常等级，根据不同的异常等级可以区别处理。It should be noted that a fault tolerance principle and a fourth abnormality level need to be set in the system. For example, a fault tolerance frequency is set. For example, within a period of time, the frequency of the abnormal state data of the minor abnormality is N times, as long as it exceeds This fault-tolerant frequency N times corresponds to an abnormality level. When the fault-tolerant frequency is N times, it corresponds to another abnormality level, which can be treated differently according to different abnormality levels.
步骤S100、通过所述第四异常等级生成第四异常处理信息。Step S100: Generate fourth abnormality processing information according to the fourth abnormality level.
需要说明的是，判断所述状态相关数据的第四异常等级后，通过所述第四异常等级生成第四异常处理信息，如停机或者继续生产等等，或者直接跳出***对话框，要求作业人员介入操作。It should be noted that after judging the fourth abnormality level of the state-related data, the fourth abnormality level is used to generate fourth abnormality handling information, such as stopping or continuing production, or directly jump out of the system dialog box and request the operator Intervention operation.
本申请中，当所述状态相关数据不属于所述异常等级数据库内的异常类型时，考虑到状态相关数据出现的频次数据，必要时，作业人员介入操作，显著地提高了所述产品的良品率及生产效率。In this application, when the state-related data does not belong to the abnormal type in the abnormal level database, taking into account the frequency data of the state-related data, if necessary, the operator intervenes in the operation, which significantly improves the quality of the product Rate and production efficiency.
请参照图5，图5为本申请提供的AI智能过程异常识别闭环控制方法第四实施例的流程示意图，在本实施例中，通过所述频次数据与容错频次进行对比，获得所述状态相关数据的第二异常等级的步骤之后，还包括：Please refer to FIG. 5, which is a flowchart of a fourth embodiment of the AI intelligent process abnormality recognition closed-loop control method provided by this application. In this embodiment, the frequency data is compared with the fault-tolerant frequency to obtain the state correlation After the step of the second abnormal level of the data, it also includes:
步骤S100a、当所述频次数据大于所述容错频次时，将所述状态相关数据存入至所述异常等级数据库，并更新所述异常等级数据库；Step S100a: When the frequency data is greater than the fault-tolerant frequency, store the state-related data in the abnormality level database, and update the abnormality level database;
需要说明的是，当所述频次数据大于所述容错频次时，说明该状态相关数据比较重要，此时提示人员参与判断，并将所述状态相关数据存入至所述异常等级数据库，并更新所述异常等级数据库，使得数据库更加完善。It should be noted that when the frequency data is greater than the fault-tolerant frequency, it indicates that the state-related data is more important. At this time, the personnel are prompted to participate in the judgment, and the state-related data is stored in the abnormality level database and updated The abnormal level database makes the database more perfect.
步骤S100b、当所述频次数据小于所述容错频次时，记录所述状态相关数据，并定期更新至所述异常等级数据库。Step S100b: When the frequency data is less than the fault-tolerant frequency, record the state-related data and update it to the abnormal level database regularly.
需要说明的是，当所述频次数据小于所述容错频次时，说明该状态相关数据不太重要，***记录所述状态相关数据，并定期更新至所述异常等级数据库。It should be noted that when the frequency data is less than the fault-tolerant frequency, it indicates that the state-related data is not important, and the system records the state-related data and regularly updates the abnormal level database.
本申请中，所述异常等级数据库是需要不断完善的，当所述频次数据大于所述容错频次时，将所述状态相关数据存入至所述异常等级数据库，并更新所述异常等级数据库，当所述频次数据小于所述容错频次时，记录所述状态相关数据，并定期更新至所述异常等级数据库，通过紧急与非紧急两种不同的情况及时更新所述异常等级数据库，使得整个***不断学习与自我完善，不断提高产品的良品率及生产效率。In this application, the abnormality level database needs to be continuously improved. When the frequency data is greater than the fault tolerance frequency, the status-related data is stored in the abnormality level database, and the abnormality level database is updated, When the frequency data is less than the fault-tolerant frequency, the state-related data is recorded and updated to the abnormality level database regularly, and the abnormality level database is updated in time through two different situations of emergency and non-emergency, so that the entire system Continuous learning and self-improvement to continuously improve product yield and production efficiency.
本申请提供一种AI智能过程异常识别闭环控制装备***，图6为本申请提供的AI智能过程异常识别闭环控制装备***的一实施例的结构示意图。This application provides an AI intelligent process abnormality recognition closed-loop control equipment system. FIG. 6 is a schematic structural diagram of an embodiment of the AI intelligent process abnormality recognition closed-loop control equipment system provided by this application.
请参阅图6至图9，所述AI智能过程异常识别闭环控制装备***1000，设置为太阳能电池组件生产线a，包括控制主机100、以及与所述控制主机100电性连接的识别组件200和执行组件300，其中，所述识别组件200设置为获取所述太阳能电池片状态相关数据，并将所述状态相关数据发送至所述控制主机100，所述执行组件300，设置为在接收到所述控制主机100的第一异常处理信息后，进行异常处理，其中，所述控制主机100包含上述的所有技术方案，因此，所述AI智能过程异常识别闭环控制装备***1000也包含上述的所有技术方案，也具有上述技术方案带来的技术效果，此处不再一一赘述。6-9, the AI intelligent process abnormal recognition closed-loop control equipment system 1000 is set as a solar cell module production line a, including a control host 100, and an identification component 200 electrically connected to the control host 100 and execution The component 300, wherein the identification component 200 is configured to obtain the state-related data of the solar cell and send the state-related data to the control host 100, and the execution component 300 is configured to receive the After the first abnormal processing information of the control host 100, exception processing is performed. The control host 100 includes all the above technical solutions. Therefore, the AI intelligent process abnormal recognition closed-loop control equipment system 1000 also includes all the above technical solutions. , Also has the technical effects brought about by the above technical solutions, and will not be repeated here.
需要说明的是，所述控制主机100相当于整个所述AI智能过程异常识别闭环控制装备***1000的控制端，所述识别组件200相当于是整个所述AI智能过程异常识别闭环控制装备***1000的输入端，所述执行组件300相当于是整个所述AI智能过程异常识别闭环控制装备***1000的输出端，所述输入端采集到相关的所述太阳能电池片的状态相关数据后传输至所述控制端，所述控制端进行相关的处理后，传出至所述输出端执行。It should be noted that the control host 100 is equivalent to the control end of the entire AI intelligent process abnormality recognition closed-loop control equipment system 1000, and the identification component 200 is equivalent to the entire AI intelligent process abnormality recognition closed-loop control equipment system 1000. At the input end, the execution component 300 is equivalent to the output end of the entire AI intelligent process abnormality recognition closed-loop control equipment system 1000. The input end collects the relevant state-related data of the solar cell and transmits it to the control After the control terminal performs related processing, it is transmitted to the output terminal for execution.
所述识别组件200相当于是整个所述AI智能过程异常识别闭环控制装备***1000的输入端，主要是采集所述太阳能电池片的状态相关数据，具体地，本实施例中，所述识别组件200包括多个识别传感器20，多个所述识别传感器20包括图像识别传感器20、温度传感器、光电传感器，分别采集所述太阳能电池片的图像、温度、定位情况等等质量数据，并将所述状态相关数据传输至所述控制主机100。The identification component 200 is equivalent to the input end of the entire AI intelligent process abnormal recognition closed-loop control equipment system 1000, and mainly collects state-related data of the solar cells. Specifically, in this embodiment, the identification component 200 It includes a plurality of identification sensors 20, which include image identification sensors 20, temperature sensors, and photoelectric sensors, which collect quality data such as images, temperature, and positioning of the solar cells, and compare the status Related data is transmitted to the control host 100.
另外，所述太阳能电池组件生产线a上具有多个工位，多个所述工位包括取料工位、传输工位、铺片工位、串焊工位、汇流焊工位和排版工位，多个所述识别传感器20对应分布于多个所述工位处，设置为获取对应所述工位处的所述太阳能电池片的状态相关数据，将不同的工位处分别设置所述识别组件200，可以全面地了解整个生产过程中的所述太阳能电池片的状态相关数据，可以全面监控所述太阳能电池组件的生产质量，出现问题及时此采用不同的处理策略进行处理，减少了人为的参与。In addition, the solar cell module production line a has a plurality of stations, and the plurality of stations include a picking station, a transmission station, a layup station, a string welding station, a bus welding station, and a layout station. The identification sensors 20 are correspondingly distributed at a plurality of the workstations, and are configured to obtain state-related data of the solar cells corresponding to the workstations, and the identification components 200 are respectively arranged at different workstations. , Can comprehensively understand the state-related data of the solar cell during the entire production process, can comprehensively monitor the production quality of the solar cell module, and use different processing strategies to deal with problems when problems occur, reducing human involvement.
需要说明的是，本实施例中，不限制所述执行组件300的具体形式，例如，可以是控制生产线上的主动电机运作的控制器，也可以是对异常的所述太阳能电池标记的标记组件400，具体地，所述AI智能过程异常识别闭环控制装备***1000，其中，还包括与所述控制主机100电性连接的标记组件400，所述控制主机100还设置为通过所述状态相关数据生成异常分布数据信息和标记信息，将所述标记信息发送至所述标记组件400，所述异常分布数据信息设置为供指导返修人员作业，所述标记组件400设置为接收所述标记信息，并对异常的所述太阳能电池片进行标记处理，所述标记信息便于返修人员识别，所述异常分布数据信息设置为指导返修人员查找异常的所述太阳能电池片。It should be noted that in this embodiment, the specific form of the execution component 300 is not limited. For example, it may be a controller that controls the operation of an active motor on a production line, or it may be a marking component that marks abnormal solar cells. 400. Specifically, the AI intelligent process abnormality recognition closed-loop control equipment system 1000 further includes a marking component 400 electrically connected to the control host 100, and the control host 100 is also configured to pass the state-related data Generate abnormal distribution data information and marking information, send the marking information to the marking component 400, the abnormal distribution data information is set to guide the repairer to work, the marking component 400 is set to receive the marking information, and The abnormal solar cell is marked, the marking information is convenient for the repairer to identify, and the abnormal distribution data information is set to guide the repairer to find the abnormal solar cell.
本实施例中，所述标记组件400包括激光打码器40，设置为对异常的所述太阳能电池片和/或所述太阳能电池组件的玻璃基板进行喷码处理，按一定的编码规则，将所述异常的太阳能电池片进行标记，记录异常的位置以及异常的类型等等信息，对于大的所述太阳能电池组件，在所述太阳能电池组件的玻璃基板进行喷码处理，指导返修人员进行作业，明显地提高了返修的作业效率。In this embodiment, the marking component 400 includes a laser encoder 40, which is configured to perform coding processing on the abnormal solar cell sheet and/or the glass substrate of the solar cell component, and according to certain coding rules, Mark the abnormal solar cells, record the location and type of abnormality, etc., for the large solar cell modules, perform coding processing on the glass substrate of the solar cell module, and guide the repair personnel to perform the work , Which obviously improves the efficiency of rework.
另外，需要说明的是，异常分布数据信息与每块太阳能电池组件玻璃基板上的条码进行关联，并存入MES***，在返修工位，对组件条码的识别，并通过MES***自动调用异常分布数据信息通过显示器进行显示，返修工位作业人员一方面可以通过异常电池片上的标记信息进行识别，另一方面通过显示器的异常分布数据信息进行识别，对两个图示进行对比，更加直观和简单，实现防呆设计，便于作业人员的返修，提高了工作效率。In addition, it should be noted that the abnormal distribution data information is associated with the barcode on the glass substrate of each solar cell module and stored in the MES system. At the repair station, the module barcode is identified, and the abnormal distribution is automatically called by the MES system. The data information is displayed on the display. On the one hand, the workers at the repair station can identify the abnormal battery slices on the one hand, and on the other hand, the abnormal distribution data information on the display is used to identify the two icons, which is more intuitive and simple. , Realize the fool-proof design, facilitate the repair of the operators, and improve the work efficiency.
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。It should be noted that in this article, the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, It also includes other elements not explicitly listed, or elements inherent to the process, method, article, or device. If there are no more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other identical elements in the process, method, article or device that includes the element.
上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。The serial numbers of the foregoing embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.
以上仅为本申请的优选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。The above are only preferred embodiments of this application, and do not limit the scope of this application. Any equivalent structure or equivalent process transformation made using the content of the description and drawings of this application, or directly or indirectly used in other related technical fields , The same reason is included in the scope of patent protection of this application.

Claims (20)

  1. 一种AI智能过程异常识别闭环控制方法，设置为太阳能电池组件生产线，其中，包括以下步骤： An AI intelligent process anomaly recognition closed-loop control method, set as a solar cell module production line, which includes the following steps:
    实时接收太阳能电池片的状态相关数据；Receive real-time data related to the status of solar cells;
    将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态；Comparing the state-related data with normal state data to determine the state of the solar cell;
    当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，获得第一异常等级；When the solar cell is in an abnormal state, matching the state-related data with the abnormal level database to obtain the first abnormal level;
    根据所述第一异常等级进行相应的异常处理策略。A corresponding exception handling strategy is performed according to the first exception level.
  2. 如权利要求1所述的AI智能过程异常识别闭环控制方法，其中，所述太阳能电池片的状态相关数据包括图像数据、温度数据、焊接成型质量数据以及定位数据。The AI intelligent process abnormality recognition closed-loop control method according to claim 1, wherein the state-related data of the solar cell includes image data, temperature data, welding quality data, and positioning data.
  3. 如权利要求1所述的AI智能过程异常识别闭环控制方法，其中，所述第一异常等级包括严重异常、主要异常和次要异常；The AI intelligent process abnormality recognition closed-loop control method according to claim 1, wherein the first abnormality level includes severe abnormality, major abnormality, and minor abnormality;
    根据所述第一异常等级进行相应的异常处理策略的步骤中：In the steps of performing a corresponding exception handling strategy according to the first exception level:
    当所述第一异常等级为严重异常时，所述异常处理策略包括生成停机信息指令；When the first abnormality level is a serious abnormality, the abnormality handling strategy includes generating a shutdown information instruction;
    当所述第一异常等级为主要异常时，所述异常处理策略包括：When the first abnormality level is a major abnormality, the abnormality handling strategy includes:
    统计第一时间段内所述状态相关数据出现的第一频次数据；Collect statistics on the first frequency of occurrence of the state-related data in the first time period;
    通过所述异常等级数据库内预存的内部设备和/或产品出现严重事风险几率计算出所述第一时间段内的第一容错频次；Calculating the first fault-tolerant frequency in the first time period according to the risk probability of serious incidents of internal equipment and/or products prestored in the abnormal level database;
    通过所述第一频次数据与所述第一容错频次进行对比，获得所述状态相关数据的第二异常等级；Obtaining the second abnormality level of the state-related data by comparing the first frequency data with the first fault tolerance frequency;
    通过所述第二异常等级生成第二异常处理信息，所述第二异常处理信息包括生成停机信息指令；Generating second abnormality handling information through the second abnormality level, where the second abnormality handling information includes an instruction to generate shutdown information;
    当所述第一异常等级为次要异常时，所述异常处理策略包括：When the first abnormality level is a secondary abnormality, the abnormality handling strategy includes:
    统计第二时间段内所述状态相关数据出现的第二频次数据；Counting the second frequency data of the state-related data in the second time period;
    通过所述异常等级数据库内预存的内部设备和/或产品出现重复性批量事故几率计算出所述第二时间段内的第二容错频次；Calculating the second fault-tolerant frequency in the second time period according to the probability of repetitive batch accidents of internal equipment and/or products pre-stored in the abnormal level database;
    通过所述第二频次数据与所述第二容错频次进行对比，获得所述状态相关数据的第三异常等级；Obtaining the third abnormality level of the state-related data by comparing the second frequency data with the second fault tolerance frequency;
    通过所述第三异常等级生成第三异常处理信息，所述第三异常处理信息包括生成警示报警信息指令。The third abnormality processing information is generated according to the third abnormality level, and the third abnormality processing information includes an instruction to generate warning information.
  4. 如权利要求1所述的AI智能过程异常识别闭环控制方法，其中，在所述将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态的步骤之后，还包括：The AI intelligent process abnormality recognition closed-loop control method according to claim 1, wherein after the step of comparing the state-related data with normal state data to determine the state of the solar cell, the method further comprises:
    当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，匹配不成功时，统计所述状态相关数据出现的频次数据；When the solar cell is in an abnormal state, the state-related data is matched with the abnormal level database, and when the matching is unsuccessful, the frequency data of the state-related data is counted;
    通过所述频次数据与容错频次进行对比，获得所述状态相关数据第四异常等级；Obtaining the fourth abnormality level of the state-related data by comparing the frequency data with the fault tolerance frequency;
    通过所述第四异常等级生成第四异常处理信息。The fourth abnormality processing information is generated by the fourth abnormality level.
  5. 如权利要求4所述的AI智能过程异常识别闭环控制方法，其中，通过所述频次数据与容错频次进行对比，获得所述状态相关数据的第四异常等级的步骤之后，还包括：The AI intelligent process abnormality recognition closed-loop control method according to claim 4, wherein after the step of obtaining the fourth abnormality level of the state-related data by comparing the frequency data with the fault tolerance frequency, the method further comprises:
    当所述频次数据大于所述容错频次时，将所述状态相关数据存入至所述异常等级数据库，并更新所述异常等级数据库；When the frequency data is greater than the fault tolerance frequency, store the state-related data in the abnormality level database, and update the abnormality level database;
    当所述频次数据小于所述容错频次时，记录所述状态相关数据，并定期更新至所述异常等级数据库。When the frequency data is less than the fault-tolerant frequency, the state-related data is recorded and updated to the abnormal level database regularly.
  6. 一种控制主机，其中，包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的AI智能过程异常识别闭环控制程序，所述AI智能过程异常识别闭环控制程序配置为实现所述AI智能过程异常识别闭环控制方法，所述AI智能过程异常识别闭环控制方法包括如下步骤：A control host, which includes: a memory, a processor, and an AI intelligent process abnormality recognition closed-loop control program stored on the memory and running on the processor, and the AI intelligent process abnormality recognition closed-loop control program is configured In order to realize the closed-loop control method of AI intelligent process abnormality recognition, the AI intelligent process abnormality recognition closed-loop control method includes the following steps:
    实时接收太阳能电池片的状态相关数据；Receive real-time data related to the status of solar cells;
    将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态；Comparing the state-related data with normal state data to determine the state of the solar cell;
    当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，获得第一异常等级；When the solar cell is in an abnormal state, matching the state-related data with the abnormal level database to obtain the first abnormal level;
    根据所述第一异常等级进行相应的异常处理策略。A corresponding exception handling strategy is performed according to the first exception level.
  7. 如权利要求6所述的控制主机，其中，所述太阳能电池片的状态相关数据包括图像数据、温度数据、焊接成型质量数据以及定位数据。7. The control host according to claim 6, wherein the state-related data of the solar cell includes image data, temperature data, welding quality data, and positioning data.
  8. 如权利要求6所述的控制主机，其中，所述第一异常等级包括严重异常、主要异常和次要异常；The control host according to claim 6, wherein the first abnormality level includes severe abnormality, major abnormality and minor abnormality;
    根据所述第一异常等级进行相应的异常处理策略的步骤中：In the steps of performing a corresponding exception handling strategy according to the first exception level:
    当所述第一异常等级为严重异常时，所述异常处理策略包括生成停机信息指令；When the first abnormality level is a serious abnormality, the abnormality handling strategy includes generating a shutdown information instruction;
    当所述第一异常等级为主要异常时，所述异常处理策略包括：When the first abnormality level is a major abnormality, the abnormality handling strategy includes:
    统计第一时间段内所述状态相关数据出现的第一频次数据；Collect statistics on the first frequency of occurrence of the state-related data in the first time period;
    通过所述异常等级数据库内预存的内部设备和/或产品出现严重事风险几率计算出所述第一时间段内的第一容错频次；Calculating the first fault-tolerant frequency in the first time period according to the risk probability of serious incidents of internal equipment and/or products prestored in the abnormal level database;
    通过所述第一频次数据与所述第一容错频次进行对比，获得所述状态相关数据的第二异常等级；Obtaining the second abnormality level of the state-related data by comparing the first frequency data with the first fault tolerance frequency;
    通过所述第二异常等级生成第二异常处理信息，所述第二异常处理信息包括生成停机信息指令；Generating second abnormality handling information through the second abnormality level, where the second abnormality handling information includes an instruction to generate shutdown information;
    当所述第一异常等级为次要异常时，所述异常处理策略包括：When the first abnormality level is a secondary abnormality, the abnormality handling strategy includes:
    统计第二时间段内所述状态相关数据出现的第二频次数据；Counting the second frequency data of the state-related data in the second time period;
    通过所述异常等级数据库内预存的内部设备和/或产品出现重复性批量事故几率计算出所述第二时间段内的第二容错频次；Calculating the second fault-tolerant frequency in the second time period according to the probability of repetitive batch accidents of internal equipment and/or products pre-stored in the abnormal level database;
    通过所述第二频次数据与所述第二容错频次进行对比，获得所述状态相关数据的第三异常等级；Obtaining the third abnormality level of the state-related data by comparing the second frequency data with the second fault tolerance frequency;
    通过所述第三异常等级生成第三异常处理信息，所述第三异常处理信息包括生成警示报警信息指令。The third abnormality processing information is generated according to the third abnormality level, and the third abnormality processing information includes an instruction to generate warning information.
  9. 如权利要求6所述的控制主机，其中，在所述将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态的步骤之后，还包括：7. The control host according to claim 6, wherein after the step of comparing the state-related data with normal state data to determine the state of the solar cell, the method further comprises:
    当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，匹配不成功时，统计所述状态相关数据出现的频次数据；When the solar cell is in an abnormal state, the state-related data is matched with the abnormal level database, and when the matching is unsuccessful, the frequency data of the state-related data is counted;
    通过所述频次数据与容错频次进行对比，获得所述状态相关数据第四异常等级；Obtaining the fourth abnormality level of the state-related data by comparing the frequency data with the fault tolerance frequency;
    通过所述第四异常等级生成第四异常处理信息。The fourth abnormality processing information is generated by the fourth abnormality level.
  10. 如权利要求9所述的控制主机，其中，通过所述频次数据与容错频次进行对比，获得所述状态相关数据的第四异常等级的步骤之后，还包括：The control host according to claim 9, wherein after the step of obtaining the fourth abnormality level of the state-related data by comparing the frequency data with the fault tolerance frequency, the method further comprises:
    当所述频次数据大于所述容错频次时，将所述状态相关数据存入至所述异常等级数据库，并更新所述异常等级数据库；When the frequency data is greater than the fault tolerance frequency, store the state-related data in the abnormality level database, and update the abnormality level database;
    当所述频次数据小于所述容错频次时，记录所述状态相关数据，并定期更新至所述异常等级数据库。When the frequency data is less than the fault-tolerant frequency, the state-related data is recorded and updated to the abnormal level database regularly.
  11. 一种AI智能过程异常识别闭环控制装备***，设置为太阳能电池组件生产线，其中，包括控制主机、以及与所述控制主机电性连接的识别组件和执行组件，其中：An AI intelligent process abnormal recognition closed-loop control equipment system is set as a solar cell module production line, which includes a control host, and identification components and execution components electrically connected to the control host, wherein:
    所述识别组件设置为获取所述太阳能电池片状态相关数据，并将所述状态相关数据发送至所述控制主机；The identification component is configured to obtain state-related data of the solar cell, and send the state-related data to the control host;
    所述执行组件，设置为在接收到所述控制主机的第一异常处理信息后，进行异常处理；The execution component is configured to perform abnormal processing after receiving the first abnormal processing information of the control host;
    所述控制主机包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的AI智能过程异常识别闭环控制程序，所述AI智能过程异常识别闭环控制程序配置为实现所述AI智能过程异常识别闭环控制方法，所述AI智能过程异常识别闭环控制方法包括如下步骤：The control host includes: a memory, a processor, and an AI intelligent process anomaly recognition closed-loop control program stored on the memory and running on the processor, and the AI intelligent process anomaly recognition closed-loop control program is configured to implement all The AI intelligent process abnormality recognition closed-loop control method, the AI intelligent process abnormality recognition closed-loop control method includes the following steps:
    实时接收太阳能电池片的状态相关数据；Receive real-time data related to the status of solar cells;
    将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态；Comparing the state-related data with normal state data to determine the state of the solar cell;
    当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，获得第一异常等级；When the solar cell is in an abnormal state, matching the state-related data with the abnormal level database to obtain the first abnormal level;
    根据所述第一异常等级进行相应的异常处理策略。A corresponding exception handling strategy is performed according to the first exception level.
  12. 如权利要求11所述的AI智能过程异常识别闭环控制装备***，其中，所述太阳能电池片的状态相关数据包括图像数据、温度数据、焊接成型质量数据以及定位数据。The AI intelligent process abnormality recognition closed-loop control equipment system of claim 11, wherein the state-related data of the solar cell includes image data, temperature data, welding quality data, and positioning data.
  13. 如权利要求11所述的AI智能过程异常识别闭环控制装备***，其中，所述第一异常等级包括严重异常、主要异常和次要异常；The AI intelligent process abnormality recognition closed-loop control equipment system according to claim 11, wherein the first abnormality level includes severe abnormality, major abnormality, and minor abnormality;
    根据所述第一异常等级进行相应的异常处理策略的步骤中：In the steps of performing a corresponding exception handling strategy according to the first exception level:
    当所述第一异常等级为严重异常时，所述异常处理策略包括生成停机信息指令；When the first abnormality level is a serious abnormality, the abnormality handling strategy includes generating a shutdown information instruction;
    当所述第一异常等级为主要异常时，所述异常处理策略包括：When the first abnormality level is a major abnormality, the abnormality handling strategy includes:
    统计第一时间段内所述状态相关数据出现的第一频次数据；Collect statistics on the first frequency of occurrence of the state-related data in the first time period;
    通过所述异常等级数据库内预存的内部设备和/或产品出现严重事风险几率计算出所述第一时间段内的第一容错频次；Calculating the first fault-tolerant frequency in the first time period according to the risk probability of serious incidents of internal equipment and/or products prestored in the abnormal level database;
    通过所述第一频次数据与所述第一容错频次进行对比，获得所述状态相关数据的第二异常等级；Obtaining the second abnormality level of the state-related data by comparing the first frequency data with the first fault tolerance frequency;
    通过所述第二异常等级生成第二异常处理信息，所述第二异常处理信息包括生成停机信息指令；Generating second abnormality handling information through the second abnormality level, where the second abnormality handling information includes an instruction to generate shutdown information;
    当所述第一异常等级为次要异常时，所述异常处理策略包括：When the first abnormality level is a secondary abnormality, the abnormality handling strategy includes:
    统计第二时间段内所述状态相关数据出现的第二频次数据；Counting the second frequency data of the state-related data in the second time period;
    通过所述异常等级数据库内预存的内部设备和/或产品出现重复性批量事故几率计算出所述第二时间段内的第二容错频次；Calculating the second fault-tolerant frequency in the second time period according to the probability of repetitive batch accidents of internal equipment and/or products pre-stored in the abnormal level database;
    通过所述第二频次数据与所述第二容错频次进行对比，获得所述状态相关数据的第三异常等级；Obtaining the third abnormality level of the state-related data by comparing the second frequency data with the second fault tolerance frequency;
    通过所述第三异常等级生成第三异常处理信息，所述第三异常处理信息包括生成警示报警信息指令。The third abnormality processing information is generated according to the third abnormality level, and the third abnormality processing information includes an instruction to generate warning information.
  14. 如权利要求11所述的AI智能过程异常识别闭环控制装备***，其中，在所述将所述状态相关数据与正常状态数据进行对比，判断所述太阳能电池片的状态的步骤之后，还包括：The AI intelligent process abnormal recognition closed-loop control equipment system according to claim 11, wherein after the step of comparing the state-related data with the normal state data to determine the state of the solar cell, the method further comprises:
    当所述太阳能电池片为异常状态时，通过所述状态相关数据与异常等级数据库进行匹配，匹配不成功时，统计所述状态相关数据出现的频次数据；When the solar cell is in an abnormal state, the state-related data is matched with the abnormal level database, and when the matching is unsuccessful, the frequency data of the state-related data is counted;
    通过所述频次数据与容错频次进行对比，获得所述状态相关数据第四异常等级；Obtaining the fourth abnormality level of the state-related data by comparing the frequency data with the fault tolerance frequency;
    通过所述第四异常等级生成第四异常处理信息。The fourth abnormality processing information is generated by the fourth abnormality level.
  15. 如权利要求14所述的AI智能过程异常识别闭环控制装备***，其中，通过所述频次数据与容错频次进行对比，获得所述状态相关数据的第四异常等级的步骤之后，还包括：The AI intelligent process abnormality recognition closed-loop control equipment system according to claim 14, wherein after the step of obtaining the fourth abnormality level of the state-related data by comparing the frequency data with the fault tolerance frequency, the method further comprises:
    当所述频次数据大于所述容错频次时，将所述状态相关数据存入至所述异常等级数据库，并更新所述异常等级数据库；When the frequency data is greater than the fault tolerance frequency, store the state-related data in the abnormality level database, and update the abnormality level database;
    当所述频次数据小于所述容错频次时，记录所述状态相关数据，并定期更新至所述异常等级数据库。When the frequency data is less than the fault-tolerant frequency, the state-related data is recorded and updated to the abnormal level database regularly.
  16. 如权利要求11所述的AI智能过程异常识别闭环控制装备***，其中，所述识别组件包括多个识别传感器，多个所述识别传感器包括图像识别传感器、温度传感器、光电传感器。The AI intelligent process abnormality recognition closed-loop control equipment system of claim 11, wherein the recognition component includes a plurality of recognition sensors, and the plurality of recognition sensors include image recognition sensors, temperature sensors, and photoelectric sensors.
  17. 如权利要求16所述的AI智能过程异常识别闭环控制装备***，其中，所述太阳能电池组件生产线上具有多个工位，多个所述工位包括取料工位、传输工位、铺片工位、串焊工位、汇流焊工位和排版工位；The AI intelligent process anomaly recognition closed-loop control equipment system according to claim 16, wherein the solar cell module production line has a plurality of stations, and the plurality of stations includes a reclaiming station, a transmission station, and a layup station. Work station, string welding station, bus welding station and typesetting station;
    多个所述识别传感器对应分布于多个所述工位处，设置为获取对应所述工位处的所述太阳能电池片的状态相关数据。A plurality of the identification sensors are correspondingly distributed at a plurality of the workstations, and are configured to obtain the state-related data of the solar cells at the corresponding workstations.
  18. 如权利要求11所述的AI智能过程异常识别闭环控制装备***，其中，还包括与所述控制主机电性连接的标记组件；The AI intelligent process abnormality recognition closed-loop control equipment system of claim 11, further comprising a marking component electrically connected to the control host;
    所述控制主机还设置为通过所述状态相关数据生成异常分布数据信息和标记信息，将所述标记信息发送至所述标记组件，所述异常分布数据信息设置为供指导返修人员作业；The control host is also configured to generate abnormal distribution data information and marking information from the state-related data, and send the marking information to the marking component, and the abnormal distribution data information is set to guide the work of the repairer;
    所述标记组件设置为接收所述标记信息，并对异常的所述太阳能电池片进行标记处理。The marking component is configured to receive the marking information and perform marking processing on the abnormal solar cell.
  19. 如权利要求11所述的AI智能过程异常识别闭环控制装备***，其中，所述标记信息包括异常的位置信息以及异常的类型信息。The AI intelligent process abnormality recognition closed-loop control equipment system according to claim 11, wherein the marking information includes abnormal location information and abnormal type information.
  20. 如权利要求19所述的AI智能过程异常识别闭环控制装备***，其中，所述标记组件包括激光打码器，设置为对异常的所述太阳能电池片和/或所述太阳能电池组件的玻璃基板进行喷码处理。 The AI intelligent process abnormality recognition closed-loop control equipment system according to claim 19, wherein the marking component includes a laser encoder, which is configured to detect abnormal solar cells and/or glass substrates of the solar cell components. Perform coding processing. To
PCT/CN2019/078376 2019-02-28 2019-03-15 Ai intelligent process abnormality recognition closed-loop control method, host and device system WO2020172919A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910153814.1A CN109828545B (en) 2019-02-28 2019-02-28 AI intelligent process anomaly identification closed-loop control method, host and equipment system
CN201910153814.1 2019-02-28

Publications (1)

Publication Number Publication Date
WO2020172919A1 true WO2020172919A1 (en) 2020-09-03

Family

ID=66864969

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/078376 WO2020172919A1 (en) 2019-02-28 2019-03-15 Ai intelligent process abnormality recognition closed-loop control method, host and device system

Country Status (2)

Country Link
CN (1) CN109828545B (en)
WO (1) WO2020172919A1 (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111626546A (en) * 2020-04-07 2020-09-04 青岛奥利普自动化控制***有限公司 Exception management system (MES) -based exception management method and equipment
CN113642683A (en) * 2021-10-14 2021-11-12 深圳市信润富联数字科技有限公司 Prompting method and device, electronic equipment and computer readable storage medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009086907A (en) * 2007-09-28 2009-04-23 Sharp Corp Defective factor estimation method, defective factor estimation device, program and recording medium
WO2010110365A1 (en) * 2009-03-26 2010-09-30 シャープ株式会社 Factor analysis apparatus and factor analysis method
CN105302104A (en) * 2015-11-24 2016-02-03 新奥光伏能源有限公司 Solar cell production line and control system and method thereof
CN107507885A (en) * 2017-07-17 2017-12-22 北京大学 Manufacture of solar cells process monitoring method based on multichannel sensor data
CN207586732U (en) * 2017-08-22 2018-07-06 正泰集团股份有限公司 Solar energy electroplax intelligent production system
CN109759755A (en) * 2019-02-28 2019-05-17 武汉三工智能装备制造有限公司 AI intelligent process anomalous identification processing system and solar cell chip bonding machine

Family Cites Families (47)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0854768A1 (en) * 1995-10-06 1998-07-29 Brown University Research Foundation Soldering methods and compositions
WO2004108345A1 (en) * 2003-06-09 2004-12-16 Senju Metal Industry Co., Ltd. Solder paste
DE102004050463B3 (en) * 2004-10-16 2006-04-20 Manz Automation Ag Test system for solar cells
WO2010090774A1 (en) * 2009-02-07 2010-08-12 Tau Science Corporation High speed detection of shunt defects in photovoltaic and optoelectronic devices
JP2011034647A (en) * 2009-08-04 2011-02-17 Toray Ind Inc Biaxially oriented polyester film
JP2011068807A (en) * 2009-09-28 2011-04-07 Toray Ind Inc Biaxially oriented polyester film
CN102566475B (en) * 2010-12-17 2015-07-08 北京北方微电子基地设备工艺研究中心有限责任公司 Method and device for processing monitoring alarm and plasma processing device
CN102999015A (en) * 2011-09-16 2013-03-27 苏州索力旺光伏设备有限公司 Photovoltaic module flexible manufacturing control system based on field bus
CN202295899U (en) * 2011-09-28 2012-07-04 山东力诺太阳能电力股份有限公司 Zero clearing cell storage box for efficiency detection of solar cells
CN103071626B (en) * 2011-10-25 2014-04-16 英稳达科技股份有限公司 Error proofing method for sorting of solar cells
TWI432551B (en) * 2011-11-11 2014-04-01 Eternal Chemical Co Ltd Conductive adhesive composition for use in solar cells and uses thereof
CN102544203A (en) * 2011-12-26 2012-07-04 嘉兴优太太阳能有限公司 Monitoring operation system in solar photovoltaic component production line
CN103185721A (en) * 2011-12-31 2013-07-03 致茂电子股份有限公司 Optical detection system
CN103377094B (en) * 2012-04-12 2016-08-03 金蝶软件（中国）有限公司 Method for monitoring abnormality and device
CN102830336B (en) * 2012-08-23 2015-09-16 英利能源（中国）有限公司 For warning device and the method for photovoltaic cell unacceptable product
CN103676790A (en) * 2012-09-07 2014-03-26 苏州索力旺光伏设备有限公司 Photovoltaic pipeline ASI bus control system
CN103067230A (en) * 2013-01-23 2013-04-24 江苏天智互联科技有限公司 Method for achieving hyper text transport protocol (http) service monitoring through embedding monitoring code
KR101447052B1 (en) * 2013-03-04 2014-10-10 중앙대학교 산학협력단 Solar Cell, Mask for fabricating solar cell electrode and method for fabricating the solar cell
CN203178203U (en) * 2013-03-06 2013-09-04 江南大学 Automatic solar silicon wafer color detection device based on machine vision
CN105759748B (en) * 2014-12-18 2018-09-25 中芯国际集成电路制造(上海)有限公司 A kind of dynamic monitoring system and monitoring method of semiconductor production board hardware performance
CN106161122A (en) * 2015-03-27 2016-11-23 银联商务有限公司 A kind of automatization's Centralized Monitoring method for early warning and system
CN104753461B (en) * 2015-04-10 2017-04-12 福州大学 Method for diagnosing and classifying faults of photovoltaic power generation arrays on basis of particle swarm optimization support vector machines
KR101810659B1 (en) * 2015-04-24 2017-12-20 주식회사 엘지화학 Electrode Terminal Welding Device of Battery Cell Assembly Comprising Battery Cells and Welding Process Using the Same
CN105262441B (en) * 2015-09-08 2017-04-12 西安交通大学 Infrared image-based photovoltaic array fault grading method
CN105491144A (en) * 2015-12-16 2016-04-13 新奥光伏能源有限公司 Solar battery production line and remote control method and system thereof
CN105914257B (en) * 2016-04-26 2017-08-25 苏州阿特斯阳光电力科技有限公司 A kind of crystal silicon cell production process monitoring method based on data analysis
KR20170129535A (en) * 2016-05-17 2017-11-27 장관영 Plating Bumping omitted
CN106055932A (en) * 2016-05-26 2016-10-26 东莞博力威电池有限公司 MCU program anti-plagiarizing method and system with Boost loader function
CN106229381B (en) * 2016-08-30 2018-03-02 奥特斯维能源（太仓）有限公司 Solar-energy photo-voltaic cell piece series welding device and series welding method
JP6701038B2 (en) * 2016-09-08 2020-05-27 アズビル株式会社 Monitoring device
CN108204331B (en) * 2016-12-19 2020-02-14 北京金风科创风电设备有限公司 Fault processing method and device for wind generating set
US10622621B2 (en) * 2017-03-31 2020-04-14 GM Global Technology Operations LLC Methods for making patterned, thick, silicon-containing electrodes
CN107192759B (en) * 2017-06-09 2019-08-27 湖南大学 A kind of photovoltaic cell lossless detection method and system based on induction optical heat radiation
CN107153410B (en) * 2017-07-12 2019-03-05 上海云统创申智能科技有限公司 A kind of intelligence sandstone aggregate production line
CN108596903B (en) * 2017-09-27 2022-06-03 广东产品质量监督检验研究院（国家质量技术监督局广州电气安全检验所、广东省试验认证研究院、华安实验室） Method for detecting defects of black edges and fragments of solar cell
CN107818312A (en) * 2017-11-20 2018-03-20 湖南远钧科技有限公司 A kind of embedded system based on abnormal behaviour identification
CN108038624A (en) * 2017-12-26 2018-05-15 北京金风科创风电设备有限公司 Method and device for analyzing health state of wind turbine generator
CN108170801A (en) * 2017-12-28 2018-06-15 成都中建材光电材料有限公司 A kind of manufacture of solar cells equipment management system and its implementation
CN108427297A (en) * 2018-03-22 2018-08-21 浙江农林大学暨阳学院 A kind of intelligent domestic system
CN108537347A (en) * 2018-04-17 2018-09-14 成都致云科技有限公司 Information technoloy equipment monitoring system and method
CN108878582A (en) * 2018-06-19 2018-11-23 成都科微科技有限公司 Improve the production method of solar panel precision
CN108803517A (en) * 2018-06-22 2018-11-13 江苏高远智能科技有限公司 A kind of Intelligentized regulation system and its method of drinks bottle placer production speed
CN109086827A (en) * 2018-08-10 2018-12-25 北京百度网讯科技有限公司 Method and apparatus for detecting monocrystaline silicon solar cell defect
CN109084957B (en) * 2018-08-31 2024-03-19 华南理工大学 Defect detection and color sorting method and system for photovoltaic solar crystalline silicon cell
CN109103288A (en) * 2018-10-09 2018-12-28 黄石金能光伏有限公司 A kind of photovoltaic module manufacturing method and photovoltaic module
CN109175581A (en) * 2018-10-16 2019-01-11 苏州辰正太阳能设备有限公司 Photovoltaic cell efficient welding device and welding procedure
CN109244192B (en) * 2018-10-25 2024-03-01 宁夏小牛自动化设备股份有限公司 Scribing and welding integrated device with half-piece battery piece standby feeding device

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009086907A (en) * 2007-09-28 2009-04-23 Sharp Corp Defective factor estimation method, defective factor estimation device, program and recording medium
WO2010110365A1 (en) * 2009-03-26 2010-09-30 シャープ株式会社 Factor analysis apparatus and factor analysis method
CN105302104A (en) * 2015-11-24 2016-02-03 新奥光伏能源有限公司 Solar cell production line and control system and method thereof
CN107507885A (en) * 2017-07-17 2017-12-22 北京大学 Manufacture of solar cells process monitoring method based on multichannel sensor data
CN207586732U (en) * 2017-08-22 2018-07-06 正泰集团股份有限公司 Solar energy electroplax intelligent production system
CN109759755A (en) * 2019-02-28 2019-05-17 武汉三工智能装备制造有限公司 AI intelligent process anomalous identification processing system and solar cell chip bonding machine

Also Published As

Publication number Publication date
CN109828545A (en) 2019-05-31
CN109828545B (en) 2020-09-11

Similar Documents

Publication Publication Date Title
WO2021003930A1 (en) Quality inspection method, apparatus, and device for customer service audio, and computer readable storage medium
WO2020172919A1 (en) Ai intelligent process abnormality recognition closed-loop control method, host and device system
WO2020073688A1 (en) Method, apparatus and device for predicting network device anomaly, and storage medium
WO2014204179A1 (en) Method for verifying bad pattern in time series sensing data and apparatus thereof
WO2020119369A1 (en) Intelligent it operation and maintenance fault positioning method, apparatus and device, and readable storage medium
WO2018201638A1 (en) Image recognition based information collection method, mobile terminal and storage medium
WO2018143607A1 (en) Apparatus and method for skill-based robot programming
WO2019112379A1 (en) Method and system for designing layout of smart factory
WO2015186862A1 (en) Display device and method of controlling therefor
WO2021101069A1 (en) Apparatus and method for testing semiconductor device by using machine learning model
WO2022114452A1 (en) High-accuracy optical particle measurement device and particle measurement method using laser power scanning
WO2022203252A1 (en) Roll map for electrode coating process, roll map creation method, and roll map creation system
WO2019169957A1 (en) Method and device for detecting performance of high-frequency cables, and readable storage medium
WO2020013619A1 (en) Substation asset management method and apparatus based on power system reliability index
WO2024071607A1 (en) Smart platform for explosion-proof inspection in industrial site
WO2020062615A1 (en) Gamma value adjustment method and apparatus for display panel, and display device
WO2020054926A1 (en) Server and controlling method thereof
WO2023214618A2 (en) Injection-molded product quality analysis and monitoring system for injection-molded product quality management
WO2023149765A1 (en) Machine tool thermal displacement compensation device and compensation method therefor
WO2019041319A1 (en) Wifi function testing method, terminal apparatus, system, and storage medium
WO2020082432A1 (en) Attaching control method for object, attaching machine and storage medium
WO2022250190A1 (en) Defect determination system for image inspection object by using deep learning model
WO2022039575A1 (en) Deep-learning-based real-time process monitoring system, and method therefor
WO2019041318A1 (en) Gps function testing method, terminal apparatus, system, and storage medium
WO2020082549A1 (en) Method for detecting gamma voltage value, gamma chip, and computer readable storage medium

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19917464

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19917464

Country of ref document: EP

Kind code of ref document: A1