CN111614784B - Edge computing box for heterogeneous data of a worksite - Google Patents
Edge computing box for heterogeneous data of a worksite Download PDFInfo
- Publication number
- CN111614784B CN111614784B CN202010485261.2A CN202010485261A CN111614784B CN 111614784 B CN111614784 B CN 111614784B CN 202010485261 A CN202010485261 A CN 202010485261A CN 111614784 B CN111614784 B CN 111614784B
- Authority
- CN
- China
- Prior art keywords
- data
- network
- unit
- interface unit
- module
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/51—Discovery or management thereof, e.g. service location protocol [SLP] or web services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/64—Protecting data integrity, e.g. using checksums, certificates or signatures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/04—Network architectures or network communication protocols for network security for providing a confidential data exchange among entities communicating through data packet networks
- H04L63/0428—Network architectures or network communication protocols for network security for providing a confidential data exchange among entities communicating through data packet networks wherein the data content is protected, e.g. by encrypting or encapsulating the payload
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/478—Supplemental services, e.g. displaying phone caller identification, shopping application
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45595—Network integration; Enabling network access in virtual machine instances
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
- G06F2216/03—Data mining
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Computer Security & Cryptography (AREA)
- Databases & Information Systems (AREA)
- Signal Processing (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Computer Hardware Design (AREA)
- Computing Systems (AREA)
- Multimedia (AREA)
- Computer Networks & Wireless Communication (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Fuzzy Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Bioethics (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention relates to an edge calculation box for heterogeneous data of a construction site, which is provided with a data processing module, a data interface module and a data application module, wherein the data processing module, the data interface module and the data application module are used for transmitting data step by step; the data processing module comprises a central processing unit with a plurality of CPUs (central processing units) and an image processing unit with a plurality of GPUs (graphic processing units), and the central processing unit and the image processing unit are respectively connected with the storage unit and the virtual machine unit; the data interface module comprises a screen interface unit, a network card interface unit, a USB interface unit and a network penetration analysis interface unit, and each interface unit is correspondingly connected with the data processing module; the data application module comprises a screen, a cloud server, an external data device and a connected network analysis and subscription server. The invention can effectively manage the data source and the data, carry out unified storage and secondary analysis on the historical video data, and greatly improve the edge computing capability and the environmental adaptability in the construction site application.
Description
Technical Field
The invention relates to a data processing device in the field of Internet of things, in particular to an edge computing box for heterogeneous data of a construction site.
Background
The NVR technology is that a network video camera or a video encoder (DVS) is installed at a front-end monitoring point, analog video, audio and other auxiliary signals are digitized by the video encoder and then uploaded to the NVR (network video recorder) in an IP (internet protocol) code stream mode; the back end has special video recording, storing and transmitting.
The visual identification box (edge counting box) corresponds to a small serving base station. The edge computing box can be accessed to various streaming media data, has certain storage and computing capabilities, and encapsulates AI visual recognition basic models, including mature face recognition models and license plate recognition models. The technology is that the current edge calculates the main flow direction of the box.
On the site of a construction site, due to the needs of various scenes, robots, positioning devices, various sensors and the like are usually deployed, and the devices generate a large amount of data, and need to perform edge calculation and transmit the calculation result back to a server.
The application of NVR technology to the current intelligent construction site has the following problems:
1. scattering of data sources:
a) the source of the video data is chaotic. In a construction site, various brands of video systems are commonly mixed, such as dahua, Haikangwei, Shang Tang and the like, remote access to videos needs to be achieved by respectively extracting video data from cloud systems of respective brands, and cloud ends lack unified management. Meanwhile, due to the fact that the construction site is in a non-professional network environment, management of account numbers and passwords is disordered, and therefore the situation that partial videos cannot be extracted exists in video data.
b) The intelligent device data source has no uniform specification. Acquiring a part of data from a cloud server; acquiring a part of data from the edge end box; and part of data is acquired from a sensor and a video monitoring and inspection robot equipment end. The data format is not uniform, the caliber is not uniform, and the standard is not uniform.
2. Edge data management confusion:
a) video data is chaotic to deposit. Different brands of devices, with their video data stored on different devices. The viewing of historical videos requires viewing on different devices and management sides.
b) There is no standard for data storage of different devices. Some of the storage is performed in a file manner, and some of the storage is database data. The core data is lost and data backup is easy to lose.
3. The site environment adaptability is poor:
a) generally, the networks of construction sites are based on household broadband, the bandwidth is small, the large data volume transmitted is unsmooth and unstable, and the data loss is easy to occur because the IP is dynamically acquired.
b) There are multiple network segments, and the network between the segments is not through, but the data needs to be interacted cooperatively.
c) The existing boxes at the edge end do not have ultra-strong intelligent analysis capability.
d) Different devices have different edge computing environments, some are based on Linux, some are based on Windows, and the consumption of computing resources is different. For this reason, a large number of calculation servers are used in a construction site, but the actual site cannot effectively provide support regardless of space and environment, and therefore integration of edge calculation is required.
Disclosure of Invention
The invention provides an edge calculation box for heterogeneous data of a construction site, which is used for solving the problems of disordered data source and data management and unified storage and secondary analysis of historical video data, and simultaneously improving the edge calculation capability and the environment adaptability.
The invention relates to an edge calculation box for heterogeneous data of a construction site, which is provided with a data processing module, a data interface module and a data application module, wherein the data processing module, the data interface module and the data application module are used for transmitting data step by step;
the data processing module comprises a central processing unit with a plurality of CPUs and an image processing unit with a plurality of GPUs, and the central processing unit and the image processing unit are respectively connected with a storage unit and a virtual machine unit;
the data interface module comprises a screen interface unit, a network card interface unit, a USB interface unit and a network penetration analysis interface unit, and each interface unit is correspondingly connected with the unit in the data processing module as required through a data bus;
the data application module comprises a screen connected with the screen interface unit, a cloud server connected with the network card interface unit, an external data device connected with the USB interface unit, and a network analysis and subscription server connected with the network penetration analysis interface unit.
The hardware part of the invention adopts the design of fanless heat dissipation, is dustproof and moistureproof, is stable and compatible, can support the high-temperature operation of more than 40 ℃, and is provided with a CPU (central processing unit), a GPU (graphic processing unit), a PMIC (power management integrated circuit), a DRAM (random access memory) and a flash memory, and has expandability. The system can be deployed in the positions of garages, bayonets, gas stations, factories, remote mountainous areas, severe unmanned areas and the like, and meets the requirements of industrial environment. In addition, the edge computing box adopts an NVIDIA GPU + ARM CPU architecture, can realize deep learning memory zero copy through a CPU and GPU memory multiplexing technology, provides rich standard hardware interfaces, perfectly adapts to various products and appearance specifications, and realizes an intelligent computing terminal in the true sense.
Meanwhile, the software setting of the edge computing box has the connection function of the existing equipment, and can be connected with the timing equipment such as a sensor, a PLC and the like commonly used in a construction site and can also be connected with the streaming media equipment such as a camera and the like. The edge computing box can integrate structured data, streaming media data, sensor data, time sequence data and the like, and carry out unified nanotube according to safety and data standards. For time sequence data, common operators such as common time window functions and aggregation functions can be called, and complex logic processing can be carried out by calling non-service functions and related existing models. For the streaming media data such as video, audio and the like, a codec, a shunt and a recombiner which are specific to the streaming media data are also provided, so that a user can be helped to quickly analyze the data from various streaming media sources, and simultaneously, the functions of video detection, video recognition, voice recognition and the like can be quickly realized through a tracker model and a voice model.
A data BUS (BUS) is arranged in the edge computing box, and a user can extract structured data from streaming media and act on processing of time series data, so that the processing boundary of the structured data and the unstructured data is broken.
The reverse proxy application is realized through a network penetration analysis interface unit, protocols such as TCP, HTTP, HTTPS and the like are supported, end-to-end TLS encrypted communication, black/white list anti-black verification is realized, and equipment for an external network device to penetrate through various complex routes and firewalls to access an internal network is supported.
The edge computing box is also provided with various existing models, so that the models have intelligent analysis capability, and the models include but are not limited to analysis models such as vehicle identification, license plate identification, face identification, motion identification, foreign matter intrusion identification and the like.
The edge computing box of the invention has the technical characteristics that:
1) virtualization
A single edge computing box can simulate a plurality of hosts through a virtual machine unit, each host can be a different operating system, mutual independence is achieved, an independent running environment is achieved, and mutual influence of edge computing is avoided.
2) Quick installation
The virtual host machine is provided with clean Linux and windows environment images, and the operating system can be quickly rebuilt.
3) Fast access
The user can connect devices with different protocols and different data formats through the familiar language; communication among different devices is realized, and more valuable insights are mined from the device data which are mutually fused.
4) High intelligence
The edge end packaging visual analysis related excellent algorithm supports deep learning frameworks such as Tensorflow, caffe, MXNet, Pythrch, Open-CV and the like. The intelligent learning system can provide AI intelligent learning, voice recognition and video recognition capabilities, and is combined with cloud capabilities to improve local intelligence.
5) Intranet penetration and data subscription
Under the dynamic IP environment, the intranet penetration can be realized through the network penetration analysis interface unit, and data service is provided for cloud users. The user can subscribe the data in the edge computing box through the cloud.
6) Low cost
For some mass data processing, the edge computing box end is used for preprocessing data to generate valuable and structured data, and the valuable and structured data are uploaded to the cloud end, so that the bandwidth and the cost can be greatly reduced.
7) Low delay
The data acquisition of the equipment can be completed in the local network where the edge computing box is located, a control strategy is realized, and the equipment data is cleaned, computed and analyzed locally in a more real-time and more reliable manner.
8) High reliability
Local calculation, storage and analysis can be operated under the condition of network disconnection or weak network environment, and no data loss or operation failure is ensured. The core data has certain local disaster recovery capability. When the network is recovered, the network can be continuously transmitted to the cloud end at a breakpoint.
9) High safety
Aiming at the data privacy problem, the risk of privacy data leakage can be relieved through data processing and storage of the edge end, and valuable desensitization data (certain sensitive information is subjected to data deformation through desensitization rules, and sensitive and privacy-based data are achieved) after processing are uploaded to the cloud.
Furthermore, the storage unit in the data processing module is provided with a disk storage device with an array structure. The multiple disk storage devices are arranged in an array structure and used as a single disk, the multiple disk storage devices store data in different disk storage devices in a segmented mode, and when the data are accessed, the related disk storage devices in the array act together, so that the data access time is greatly reduced, and meanwhile, the space utilization rate is better. And also supports the uniform format and uniform storage of a large number of video hard disks.
Furthermore, a multi-screen display interface is arranged in the screen interface unit in the data interface module; and the screen in the data application module is a multi-screen separate display screen.
Furthermore, at least two network card interfaces corresponding to different network segments are arranged in the network card interface unit in the data interface module; and the cloud server in the data application module is a cascade structure server corresponding to the network card interface of each network segment. Therefore, the access capability of different network segments can be provided, and the cascade function can be provided under the condition of not changing the network segments.
Furthermore, at least two USB interfaces are arranged in the USB interface unit in the data interface module; and the external data equipment in the data application module is respectively connected with the corresponding USB interfaces according to the requirements.
The edge computing box can perform virtual resource management and resource scheduling, can virtualize resources into a plurality of nodes, performs distributed computing, storage and data analysis, and performs respective standardized management (such as robots, common cameras, Network Video Recorders (NVRs) and the like) aiming at different equipment data.
The beneficial effects of the invention include:
(1) the data can be directly collected by cross-network segment connection without being influenced by network modification limitation.
(2) Through virtualization, different environments can be deployed to provide powerful edge computing power, saving space.
(3) The video analysis service capability is provided, and the visual recognition can be carried out in real time.
(4) And performing edge processing, unified storage, core data disaster recovery capability and breakpoint continuous transmission capability on heterogeneous data.
(5) The video data can provide uniform storage and uniform video service regardless of manufacturer brands.
(6) The cloud terminal data subscription system has network penetration capability and provides data subscription service for cloud terminal users in a secure environment.
The present invention will be described in further detail with reference to the following examples. This should not be understood as limiting the scope of the above-described subject matter of the present invention to the following examples. Various substitutions and alterations according to the general knowledge and conventional practice in the art are intended to be included within the scope of the present invention without departing from the technical spirit of the present invention as described above.
Drawings
FIG. 1 is a block diagram of an edge computing box for heterogeneous data for a worksite according to the present invention.
Detailed Description
The edge computing box for heterogeneous data of a construction site, disclosed by the invention, is provided with a data processing module, a data interface module and a data application module, wherein the data processing module, the data interface module and the data application module are used for transmitting data step by step;
the data processing module comprises a central processing unit with a plurality of CPUs and an image processing unit with a plurality of GPUs, wherein the central processing unit and the image processing unit are respectively connected with a storage unit and a virtual machine unit with a multi-screen display interface. Wherein, the storage unit is provided with a magnetic disk storage device with an array structure. The multiple disk storage devices are arranged in an array structure and used as a single disk, the multiple disk storage devices store data in different disk storage devices in a segmented mode, and when the data are accessed, the related disk storage devices in the array act together, so that the data access time is greatly reduced, and meanwhile, the space utilization rate is better. And also supports the uniform format and uniform storage of a large number of video hard disks.
The data interface module comprises a screen interface unit, a network card interface unit, a USB interface unit and a network penetration analysis interface unit, and each interface unit is correspondingly connected with the unit in the data processing module as required through a data BUS BUS. The network card interface unit is provided with a plurality of network card interfaces corresponding to different network segments, and the USB interface unit is also provided with a plurality of USB interfaces.
The data application module comprises a screen of a multi-screen sub-display structure connected with the screen interface unit, a cloud server of a cascade structure correspondingly connected with the network card interfaces of all network segments, external data equipment correspondingly connected with the USB interfaces, and a network analysis and subscription server connected with the network penetration analysis interface unit.
The edge computing box can perform virtual resource management and resource scheduling, can virtualize resources into a plurality of nodes, performs distributed computing, storage and data analysis, and performs respective standardized management (such as robots, common cameras, Network Video Recorders (NVRs) and the like) aiming at different equipment data.
In addition, the hardware part of the invention adopts a fanless heat dissipation design, can prevent dust and moisture, is stable and compatible, can support high-temperature operation above 40 ℃, and is provided with a CPU (central processing unit), a GPU (graphic processing unit), a PMIC (power management integrated circuit), a DRAM (random access memory) and a flash memory, and has expandability. The system can be deployed in the positions of garages, bayonets, gas stations, factories, remote mountainous areas, severe unmanned areas and the like, and meets the requirements of industrial environment. In addition, the edge computing box adopts an NVIDIA GPU + ARM CPU architecture, can realize deep learning memory zero copy through a CPU and GPU memory multiplexing technology, provides rich standard hardware interfaces, perfectly adapts to various products and appearance specifications, and realizes an intelligent computing terminal in the true sense.
Meanwhile, the software setting of the edge computing box has the connection function of the existing equipment, and can be connected with the timing equipment such as a sensor, a PLC and the like commonly used in a construction site and can also be connected with the streaming media equipment such as a camera and the like. The edge computing box can integrate structured data, streaming media data, sensor data, time sequence data and the like, and carry out unified nanotube according to safety and data standards. For time sequence data, common operators such as common time window functions and aggregation functions can be called, and complex logic processing can be carried out by calling non-service functions and related existing models. For the streaming media data such as video, audio and the like, a codec, a shunt and a recombiner which are specific to the streaming media data are also provided, so that a user can be helped to quickly analyze the data from various streaming media sources, and simultaneously, the functions of video detection, video recognition, voice recognition and the like can be quickly realized through a tracker model and a voice model.
A data BUS (BUS) is arranged in the edge computing box, and a user can extract structured data from streaming media and act on processing of time series data, so that the processing boundary of the structured data and the unstructured data is broken.
The reverse proxy application is realized through a network penetration analysis interface unit, protocols such as TCP, HTTP, HTTPS and the like are supported, end-to-end TLS encrypted communication, black/white list anti-black verification is realized, and equipment for an external network device to penetrate through various complex routes and firewalls to access an internal network is supported.
The edge computing box is also provided with various existing models, so that the models have intelligent analysis capability, and the models include but are not limited to analysis models such as vehicle identification, license plate identification, face identification, motion identification, foreign matter intrusion identification and the like.
The edge computing box of the present invention can implement:
1) virtualization
A single edge computing box can simulate a plurality of hosts through a virtual machine unit, each host can be a different operating system, mutual independence is achieved, an independent running environment is achieved, and mutual influence of edge computing is avoided.
2) Quick installation
The virtual host machine is provided with clean Linux and windows environment images, and the operating system can be quickly rebuilt.
3) Fast access
The user can connect devices with different protocols and different data formats through the familiar language; communication among different devices is realized, and more valuable insights are mined from the device data which are mutually fused.
4) High intelligence
The edge end packaging visual analysis related excellent algorithm supports deep learning frameworks such as Tensorflow, caffe, MXNet, Pythrch, Open-CV and the like. The intelligent learning system can provide AI intelligent learning, voice recognition and video recognition capabilities, and is combined with cloud capabilities to improve local intelligence.
5) Intranet penetration and data subscription
Under the dynamic IP environment, the intranet penetration can be realized through the network penetration analysis interface unit, and data service is provided for cloud users. The user can subscribe the data in the edge computing box through the cloud.
6) Low cost
For some mass data processing, the edge computing box end is used for preprocessing data to generate valuable and structured data, and the valuable and structured data are uploaded to the cloud end, so that the bandwidth and the cost can be greatly reduced.
7) Low delay
The data acquisition of the equipment can be completed in the local network where the edge computing box is located, a control strategy is realized, and the equipment data is cleaned, computed and analyzed locally in a more real-time and more reliable manner.
8) High reliability
Local calculation, storage and analysis can be operated under the condition of network disconnection or weak network environment, and no data loss or operation failure is ensured. The core data has certain local disaster recovery capability. When the network is recovered, the network can be continuously transmitted to the cloud end at a breakpoint.
9) High safety
Aiming at the data privacy problem, the risk of privacy data leakage can be relieved through data processing and storage of the edge end, and valuable desensitization data (certain sensitive information is subjected to data deformation through desensitization rules, and sensitive and privacy-based data are achieved) after processing are uploaded to the cloud.
Meanwhile, the invention can also support the communication protocols of main stream equipment such as Modbus, CoAP, OPC-UA, MQTT, RTSP, RTMP, UVC and the like, and provides an interface to support the butt joint of a third party private protocol. The streaming media device can access the platform through the ONVIF standard protocol without providing an SDK (software development kit) by a camera manufacturer. The cloud defines the product type and the function model, and the edge end is accessed into the equipment and managed.
In the edge computing box of the present invention, existing software functions are provided including:
a rule engine:
dragging the visual assembly, designing a unified data transfer rule, and flexibly defining data processing, data storage and data forwarding. For time sequence data, common operators such as a time window function, an aggregation function and the like are provided, and a user is helped to quickly set up a time sequence data processing flow; for streaming media data, codecs, splitters, combiners specific to the streaming media data are provided to help users quickly parse data from various streaming media sources.
And function calculation:
in addition to the predefined components provided by the rule engine, the user can process the data by calling the non-service function and call the non-service function in the rule engine to meet the implementation of complex business logic.
Intelligent reasoning:
by deploying inference service locally, more intelligent data processing logic is completed, and an AI intelligent inference function with low delay is realized.
Rule application:
based on visual rule definition, a rule example is established, the data flow definition of the multiple devices can be realized by mapping the entity device, the AI model and the rule with each other, and business personnel only face the devices without programming.
And (3) continuous transmission in a broken network:
under the condition of network disconnection, the edge computing box end can independently operate, and real-time data processing is not influenced. After the network is recovered, the edge data and the cloud data can be continuously transmitted and synchronized.
Data visualization:
the user can define the instrument panel by himself, and the user can view the original real-time data of the access device and also view the data output after the access rule calculation, so that the long-term monitoring of the device data is realized.
Cloud and edge integration:
the user can carry out version management, gray scale upgrading and other operations on the application, the function, the rule and the model of the edge computing box end through the cloud end, and the operation and maintenance cost of the edge end is reduced. The cloud provides a uniform visual angle of the edge end, all registered edge nodes and states can be checked, and applications, models, functions and rules of the cloud are distributed to the edge end, so that one-time development and multiple deployments are realized.
Claims (5)
1. An edge computing box for heterogeneous data of a worksite, characterized by: the data processing module, the data interface module and the data application module are used for transmitting data step by step;
the data processing module comprises a central processing unit with a plurality of CPUs and an image processing unit with a plurality of GPUs, and the central processing unit and the image processing unit are respectively connected with a storage unit and a virtual machine unit;
the data interface module comprises a screen interface unit, a network card interface unit, a USB interface unit and a network penetration analysis interface unit, and each interface unit is correspondingly connected with the unit in the data processing module as required through a data bus;
the data application module comprises a screen connected with the screen interface unit, a cloud server connected with the network card interface unit, an external data device connected with the USB interface unit, and a network analysis and subscription server connected with the network penetration analysis interface unit;
further comprising:
a rule engine: providing a dragging visualization component, and designing a uniform data flow rule for defining data processing, data storage and data forwarding; specifically, for time series data, a common operator comprising a time window function and an aggregation function is provided, and a user is helped to quickly set up a time series data processing flow; for the streaming media data, a codec, a splitter and a compositor which are specific to the streaming media data are provided, so that a user can be helped to rapidly analyze the data from various streaming media sources;
and function calculation: the user processes the data by calling the non-service function and calls the non-service function in the rule engine to meet the realization of complex service logic;
intelligent reasoning: reasoning service is deployed locally to complete more intelligent data processing logic;
rule application: establishing a rule instance based on the visualized rule definition, and mapping the entity equipment, the AI model and the rule with each other;
and (3) continuous transmission in a broken network: independently operating under the condition of network interruption, performing real-time data processing, and continuously transmitting and synchronizing the edge data and the cloud data after network recovery;
data visualization: a user obtains original real-time data of access equipment and data output after access rule calculation through a user-defined instrument panel, and long-term monitoring of the equipment data is kept;
cloud and edge integration: the method comprises the steps that a user conducts operations including version management and gray scale upgrading on an application, a function, a rule and a model of an edge computing box end through a cloud end, a unified visual angle of the edge computing box end is provided at the cloud end, all registered edge nodes and states are checked and obtained, and then the application, the model, the function and the rule which are put on the shelf at the cloud end are distributed to the edge computing box end for realizing one-time development and multiple deployment.
2. An edge computing box for heterogeneous data of a worksite according to claim 1, wherein: the storage unit in the data processing module is provided with a disk storage device with an array structure.
3. An edge computing box for heterogeneous data of a worksite according to claim 1, wherein: a multi-screen display interface is arranged in the screen interface unit in the data interface module; and the screen in the data application module is a multi-screen separate display screen.
4. An edge computing box for heterogeneous data of a worksite according to claim 1, wherein: at least two network card interfaces corresponding to different network segments are arranged in the network card interface unit in the data interface module; and the cloud server in the data application module is a cascade structure server corresponding to the network card interface of each network segment.
5. An edge computing box for heterogeneous data of a worksite according to claim 1, wherein: at least two USB interfaces are arranged in the USB interface unit in the data interface module; and the external data equipment in the data application module is respectively connected with the corresponding USB interfaces according to the requirements.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010485261.2A CN111614784B (en) | 2020-06-01 | 2020-06-01 | Edge computing box for heterogeneous data of a worksite |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010485261.2A CN111614784B (en) | 2020-06-01 | 2020-06-01 | Edge computing box for heterogeneous data of a worksite |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111614784A CN111614784A (en) | 2020-09-01 |
CN111614784B true CN111614784B (en) | 2021-05-11 |
Family
ID=72201967
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010485261.2A Active CN111614784B (en) | 2020-06-01 | 2020-06-01 | Edge computing box for heterogeneous data of a worksite |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111614784B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113191455B (en) * | 2021-05-26 | 2024-06-07 | 平安国际智慧城市科技股份有限公司 | Edge computing box election method and device, electronic equipment and medium |
CN113282376B (en) * | 2021-07-22 | 2021-11-12 | 北京关键科技股份有限公司 | UKey virtual machine penetration method applied to cloud platform architecture |
CN115311796A (en) * | 2022-07-11 | 2022-11-08 | 西安电子科技大学广州研究院 | Edge intelligent security alarm system |
CN117636264A (en) * | 2024-01-25 | 2024-03-01 | 泉州装备制造研究所 | Intelligent monitoring method and system for factory safety detection based on edge computing box |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110505073A (en) * | 2018-05-16 | 2019-11-26 | 华为技术有限公司 | A kind of mobile edge calculations method and device |
CN110545307A (en) * | 2019-07-19 | 2019-12-06 | 中移(杭州)信息技术有限公司 | Edge computing platform, calling method and computer readable storage medium |
CN209895339U (en) * | 2019-04-11 | 2020-01-03 | 北京宙心科技有限公司 | Artificial intelligence edge computing equipment |
CN111031034A (en) * | 2019-12-11 | 2020-04-17 | 研祥智能科技股份有限公司 | Multi-protocol convergence edge computing gateway |
CN111106998A (en) * | 2019-12-30 | 2020-05-05 | 江苏欧联智能科技有限公司 | Industrial edge computing service gateway |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060085573A1 (en) * | 2004-10-15 | 2006-04-20 | Dell Products L.P. | Multi-context selection with PCI express to support hardware partitioning |
US9407542B2 (en) * | 2011-12-20 | 2016-08-02 | Cisco Technology, Inc. | Network architecture for minimalistic connected objects |
CN108355350B (en) * | 2018-02-11 | 2020-04-10 | 清华大学 | Application service access method and device based on mobile edge computing |
CN110032540A (en) * | 2019-04-11 | 2019-07-19 | 北京宙心科技有限公司 | A kind of artificial intelligence edge calculations equipment |
CN110572448B (en) * | 2019-08-30 | 2020-11-06 | 烽火通信科技股份有限公司 | Distributed edge cloud system architecture |
-
2020
- 2020-06-01 CN CN202010485261.2A patent/CN111614784B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110505073A (en) * | 2018-05-16 | 2019-11-26 | 华为技术有限公司 | A kind of mobile edge calculations method and device |
CN209895339U (en) * | 2019-04-11 | 2020-01-03 | 北京宙心科技有限公司 | Artificial intelligence edge computing equipment |
CN110545307A (en) * | 2019-07-19 | 2019-12-06 | 中移(杭州)信息技术有限公司 | Edge computing platform, calling method and computer readable storage medium |
CN111031034A (en) * | 2019-12-11 | 2020-04-17 | 研祥智能科技股份有限公司 | Multi-protocol convergence edge computing gateway |
CN111106998A (en) * | 2019-12-30 | 2020-05-05 | 江苏欧联智能科技有限公司 | Industrial edge computing service gateway |
Also Published As
Publication number | Publication date |
---|---|
CN111614784A (en) | 2020-09-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111614784B (en) | Edge computing box for heterogeneous data of a worksite | |
CN110351257B (en) | Distributed Internet of things security access system | |
Khare et al. | Big data in IoT | |
CN102752574B (en) | Video monitoring system and method | |
CN104079885A (en) | Nobody-monitored and linkage-tracked network camera shooting method and device | |
US11575659B2 (en) | Systems and methods for establishing secure remote connections to media devices | |
CN112596928A (en) | Industrial robot data management method, device, equipment and computer storage medium | |
CN117540321A (en) | Data service monitoring system | |
CN115409643A (en) | Digital twin operation monitoring system for electric power facilities | |
US10921792B2 (en) | Edge cloud-based resin material drying system and method | |
CN113596104B (en) | Data transmission method and device for mobile robot and cloud server | |
US10693736B2 (en) | Real time simulation monitoring | |
CN112486706B (en) | Internet of things local equipment linkage method based on MQTT message driving mechanism | |
CN116580362B (en) | Transmission operation cross-system fusion data acquisition method and digital asset processing system | |
CN116389492A (en) | Video analysis system, method, apparatus, and computer-readable storage medium | |
Blok et al. | The System of the Supervision and the Visualization of Multimedia Data for BG | |
CN112416532A (en) | Industrial data processing system and method | |
CN202652404U (en) | High definition video processing and conference platform system based on cloud computing | |
Muzelak et al. | Edge computing implementation of safety monitoring system in frame of IIoT | |
US11621881B2 (en) | Error detection and broadcasting using partner sensors | |
CN115934327B (en) | Distributed interactive collaboration method for industrial edge server application | |
KR102391898B1 (en) | Remote access system for transmitting video data and method performing there of | |
CN117234739B (en) | Method, apparatus, system, and storage medium for industrial data analysis | |
CN203181063U (en) | Network camera of safety production cloud service platform in industrial and mining enterprise | |
US20230142107A1 (en) | Data pipeline management in operational technology hardware and networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |