CN117050760B - Intelligent coal charging and coke discharging system - Google Patents
Intelligent coal charging and coke discharging system Download PDFInfo
- Publication number
- CN117050760B CN117050760B CN202311327250.1A CN202311327250A CN117050760B CN 117050760 B CN117050760 B CN 117050760B CN 202311327250 A CN202311327250 A CN 202311327250A CN 117050760 B CN117050760 B CN 117050760B
- Authority
- CN
- China
- Prior art keywords
- coal
- charging
- quality score
- coke
- image
- 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
- 239000000571 coke Substances 0.000 title claims abstract description 121
- 239000003245 coal Substances 0.000 title claims abstract description 31
- 238000007599 discharging Methods 0.000 title claims abstract description 22
- 230000002159 abnormal effect Effects 0.000 claims description 37
- 230000005856 abnormality Effects 0.000 claims description 33
- 238000000034 method Methods 0.000 claims description 28
- 238000006243 chemical reaction Methods 0.000 claims description 17
- 238000006073 displacement reaction Methods 0.000 claims description 15
- 238000012795 verification Methods 0.000 claims description 9
- 238000001514 detection method Methods 0.000 claims description 8
- 230000001960 triggered effect Effects 0.000 claims description 8
- 238000004891 communication Methods 0.000 claims description 2
- 238000005235 decoking Methods 0.000 claims 2
- 238000004519 manufacturing process Methods 0.000 abstract description 6
- 238000004364 calculation method Methods 0.000 description 5
- 238000005259 measurement Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 230000002708 enhancing effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000004886 process control Methods 0.000 description 2
- 238000003723 Smelting Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003763 carbonization Methods 0.000 description 1
- 238000005266 casting Methods 0.000 description 1
- 238000004939 coking Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002309 gasification Methods 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000012265 solid product Substances 0.000 description 1
Classifications
-
- C—CHEMISTRY; METALLURGY
- C10—PETROLEUM, GAS OR COKE INDUSTRIES; TECHNICAL GASES CONTAINING CARBON MONOXIDE; FUELS; LUBRICANTS; PEAT
- C10B—DESTRUCTIVE DISTILLATION OF CARBONACEOUS MATERIALS FOR PRODUCTION OF GAS, COKE, TAR, OR SIMILAR MATERIALS
- C10B41/00—Safety devices, e.g. signalling or controlling devices for use in the discharge of coke
- C10B41/005—Safety devices, e.g. signalling or controlling devices for use in the discharge of coke for charging coal
-
- C—CHEMISTRY; METALLURGY
- C10—PETROLEUM, GAS OR COKE INDUSTRIES; TECHNICAL GASES CONTAINING CARBON MONOXIDE; FUELS; LUBRICANTS; PEAT
- C10B—DESTRUCTIVE DISTILLATION OF CARBONACEOUS MATERIALS FOR PRODUCTION OF GAS, COKE, TAR, OR SIMILAR MATERIALS
- C10B41/00—Safety devices, e.g. signalling or controlling devices for use in the discharge of coke
- C10B41/02—Safety devices, e.g. signalling or controlling devices for use in the discharge of coke for discharging coke
Landscapes
- Chemical & Material Sciences (AREA)
- Engineering & Computer Science (AREA)
- Materials Engineering (AREA)
- Oil, Petroleum & Natural Gas (AREA)
- Organic Chemistry (AREA)
- Coke Industry (AREA)
Abstract
The invention relates to the field of coke production, and discloses an intelligent coal charging and coke discharging system, which comprises the following steps: the image enhancement device is arranged on a door frame of the coke oven; the infrared positioning device comprises an infrared sensor and a plurality of positioning holes; the plurality of image acquisition devices comprise a vehicle-mounted image acquisition device and a fixed image acquisition device; the computing equipment is used for determining the quality score of the enhanced image, inputting the enhanced image into the positioning network to obtain position offset information if the quality score is greater than or equal to a quality score threshold value, and generating a running control instruction according to the position offset information; transmitting a running control instruction to a control unit of the coal-charging coke pusher so as to enable the coal-charging coke pusher to run to a target position; if the quality score is smaller than the quality score threshold, an infrared starting instruction is generated so as to start the infrared positioning device, control the coal-charging coke pushing vehicle to travel to a target position and execute corresponding coal-charging or coke pushing operation. Thus, the accurate alignment of the coal-charging coke pusher and the furnace door is realized.
Description
Technical Field
The invention relates to the field of coke production, in particular to an intelligent coal charging and coke discharging system.
Background
The coke is a solid product of high-temperature carbonization, and the coke obtained by high Wen Lianjiao is mainly used in the fields of blast furnace smelting, casting, gasification and the like, and the coke production comprises the procedures of coal preparation, coal blending, coking and the like.
The existing coke production process often has the following technical problems:
firstly, in the process of coal loading or coke discharging, the coal loading coke pushing trolley needs to be aligned with the furnace door of the coke furnace, so that a coke pushing rod on the coal loading coke pushing trolley can perform a series of operations such as furnace door picking and coke pushing, and in the process, the problem that the furnace door is failed to pick up or the coke pushing or the coal loading is failed due to the fact that the furnace door cannot be aligned accurately often occurs;
secondly, during the coal loading or coke discharging process, accurate process control measures are lacked, so that abnormal situations in the process cannot be found;
third, in the process of achieving alignment, a distance conversion coefficient needs to be predetermined, however, the distance conversion coefficient is affected by a series of external environments such as image capturing light, capturing equipment precision, and the like, and it is difficult to determine an accurate distance conversion coefficient, so that the coal-charging coke pusher often cannot be accurately moved to a corresponding position.
Disclosure of Invention
This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.
The invention provides an intelligent coal charging and coke discharging system which solves one or more of the technical problems mentioned in the background art section.
The invention provides an intelligent coal-charging coke-discharging system, which comprises: the image enhancement device is arranged on a furnace door frame of the coke furnace and is used for marking the position of the furnace door frame; the infrared positioning device comprises a plurality of infrared sensors arranged on the coal-charging coke pusher and a plurality of positioning holes arranged on the furnace door frame, and the infrared sensors and the positioning holes are in one-to-one correspondence; the system comprises a plurality of image acquisition devices, a plurality of image acquisition devices and a plurality of image acquisition devices, wherein the plurality of image acquisition devices comprise a vehicle-mounted image acquisition device and a fixed image acquisition device, the vehicle-mounted image acquisition device is arranged on a coal-charging coke pusher, and the vehicle-mounted image acquisition device is used for acquiring images of an image enhancement device when the distance between the coal-charging coke pusher and a furnace door frame is smaller than a preset threshold value, so as to obtain enhanced images; the computing equipment is in communication connection with the vehicle-mounted image acquisition equipment and is used for acquiring the enhanced image from the vehicle-mounted image acquisition equipment and determining the quality score of the enhanced image, if the quality score is greater than or equal to a preset quality score threshold value, the enhanced image is input into a pre-trained positioning network to obtain the position deviation information of the image enhancement device in the enhanced image, and a running control instruction for the coal-charging coke pusher is generated according to the position deviation information; transmitting a running control instruction to a control unit of the coal-charging coke pusher so that the control unit controls the coal-charging coke pusher to run to a target position and execute corresponding coal-charging or coke pushing operation; if the quality score is smaller than a preset quality score threshold, an infrared starting instruction is generated, and the infrared starting instruction is sent to the control unit, so that the control unit starts the infrared positioning device and receives a feedback signal of the infrared positioning device, and controls the coal-charging coke pushing vehicle to travel to a target position and execute corresponding coal-charging or coke pushing operation based on the feedback signal.
Optionally, the intelligent coal charging and coke discharging system further comprises: the sensing device is buried below the track of the coal-charging coke pusher and is arranged corresponding to the coke oven, when the coal-charging coke pusher reaches a target position, the sensing device detects that the coal-charging coke pusher reaches and records the starting time, when the coal-charging coke pusher leaves the target position, the sensing device detects that the coal-charging coke pusher leaves and records the leaving time, the starting time, the leaving time and the current operation type are determined as operation record information, and the operation record information is sent to the computing equipment; the computing equipment is also used for matching a corresponding abnormal recognition model from a pre-trained abnormal recognition model pool according to the operation type included in the operation record information, determining the matched abnormal recognition model as a target abnormal recognition model, inputting the starting time and the leaving time included in the operation record information into the target abnormal recognition model, and generating abnormal recognition information, wherein the abnormal recognition information characterizes whether the operation record information has an abnormality; if the abnormality identification information characterizes that the operation record information is abnormal, determining a corresponding historical operation video from a historical operation video library according to the starting time and the leaving time; analyzing the historical operation video to obtain a key frame group in the historical operation video, and generating abnormal verification information according to the key frame group.
Optionally, the target anomaly identification model is a comparator, and the comparator is used for determining a residence time according to the start time and the departure time and comparing the residence time with a historical residence mean value; outputting abnormal identification information representing that the abnormality exists if the stay time length is longer than the historical stay mean value, and outputting abnormal identification information representing that the abnormality does not exist if the stay time length is shorter than or equal to the historical stay mean value; if the abnormal identification information indicates that no abnormality exists, the corresponding stay time length is added into the historical stay time length database, an updated historical stay time length database is obtained, and the historical stay mean value is updated according to the updated historical stay time length database.
Optionally, the quality score threshold comprises a first quality score threshold and a second quality score threshold, the first quality score threshold being greater than the second quality score threshold; and if the quality score is greater than or equal to a preset quality score threshold, inputting the enhanced image into a pre-trained positioning network to obtain the position deviation information of the image enhancement device in the enhanced image, wherein the method comprises the following steps: if the quality score is greater than or equal to a first quality score threshold, inputting the enhanced image into a pre-trained positioning network to obtain the position offset information of the image enhancement device in the enhanced image; if the quality score is greater than or equal to the second quality score threshold and smaller than the first quality score threshold, defogging the enhanced image to obtain a defogging image, and inputting the defogging image into a positioning network to obtain the position deviation information of the image enhancement device in the enhanced image; and if the quality score is smaller than a preset quality score threshold, generating an infrared starting instruction, which comprises the following steps: and if the quality score is smaller than the second quality score threshold value, generating an infrared starting instruction.
Optionally, generating the anomaly verification information according to the keyframe group includes: extracting features of each key frame in the key frame group to obtain key frame features corresponding to each key frame; splicing key frame characteristics corresponding to each key frame in the key frame group respectively to obtain frame group characteristics corresponding to the key frame group; word embedding is carried out on the operation type to obtain operation type characteristics; and after the frame group characteristics and the operation type characteristics are spliced, inputting the spliced frame group characteristics and the operation type characteristics into a classifier, and outputting abnormal verification information.
Optionally, the position offset information includes a position offset value and an offset direction; and generating a travel control command acting on the coal-charging coke pusher according to the positional deviation information, comprising: determining a moving distance according to the position offset value and a predetermined distance conversion coefficient, and determining a moving direction according to the offset direction, wherein the moving distance and the moving direction form a running control instruction; and updating the distance conversion coefficient by the following steps: acquiring an updated enhanced image corresponding to the target position; the method comprises the steps of respectively carrying out feature detection on an updated enhanced image and an enhanced image by utilizing a Gaussian differential function so as to identify feature points with scale and rotation invariance, and obtaining a first key point set corresponding to the enhanced image and a second key point set corresponding to the updated enhanced image; for any first key point in the first key point set, determining the key point closest to the Euclidean distance of the first key point in the second key point set as a matching key point of the first key point; determining whether at least one first key point group with the same corresponding matching key point exists in the first key point set; if at least one first key point group with the same corresponding matching key point exists, reserving first key points closest to the matching key point for a plurality of first key points in each first key point group in the at least one first key point group, and deleting the rest first key points to obtain an updated first key point set; determining a displacement between the updated enhanced image and the enhanced image according to the updated first key point set and the corresponding matched key point set; and determining whether the difference value of the displacement and the moving distance is larger than a preset threshold value, and if the difference value of the displacement and the moving distance is larger than the preset threshold value, adjusting the distance conversion coefficient according to the displacement.
Optionally, the feedback signal includes a measured distance detected by each of the plurality of infrared sensors; and controlling the coal-charging coke pusher to travel to a target position and performing a corresponding coal-charging or coke pushing operation based on the feedback signal, comprising: comparing the measured distance with a preset distance for each of the plurality of infrared sensors to determine whether each of the infrared sensors is triggered; when the number of the triggered infrared sensors is larger than the preset number, the coal-charging coke pushing vehicle is controlled to travel to the target position and corresponding coal-charging or coke pushing operation is executed.
The invention has the following beneficial effects:
1. the alignment is realized through two modes of infrared positioning and image positioning, and finally, the accurate alignment of the coal-charging coke pusher and the furnace door is realized, so that the situation that the furnace door is not picked or the coke is pushed or the coal is not charged is avoided to the greatest extent. In practice, the inventor finds that because the coke oven frame is connected with other structures such as the oven body, the problem of false identification often occurs in the image positioning process, and the oven door cannot be accurately aligned. Based on the method, the image enhancement device is arranged, and the furnace door frame is coated with the specified color or the locating plate with the specified color is arranged to form clear contrast with the surrounding environment, so that the accuracy of image locating is improved, the accuracy of locating is further improved, and the accurate alignment with the furnace door is realized. In addition, the inventor also finds that the accuracy of image positioning is also affected when special weather such as rainy days or heavy fog is met. Based on the method, the quality scoring is carried out on the shot enhanced image, and when the quality scoring is smaller than a preset quality scoring threshold value, the alignment is realized by utilizing a plurality of infrared sensors and a plurality of positioning holes matched with the infrared sensors, so that the accurate alignment under special weather conditions is further realized;
2. by arranging the sensing device, accurate measurement of the residence time is realized, and the accurate measurement of the residence time can reflect abnormal situations in the coal loading or coke discharging process. Furthermore, by combining the historical operation video, the abnormal situation can be verified, the abnormal situation is prevented from being wrongly identified, and the identification accuracy of the abnormal situation is improved, so that the abnormal situation in the coal charging or coke discharging process can be timely and accurately found, the abnormality can be eliminated, and the normal production operation is ensured.
3. The dynamic adjustment of the distance conversion coefficient is realized, so that the applicability to different external environments is improved, the coal-charging coke pusher can be ensured to accurately move to the corresponding position, and the accurate alignment of the furnace door is realized. Specifically, by performing feature detection on the updated enhanced image and the enhanced image, corresponding key point sets are respectively obtained. On the basis, the key point matching and the de-duplication are carried out, so that the displacement between the two images is obtained, the displacement is compared with the moving distance, and further, the distance conversion coefficient is checked and adjusted, so that the distance conversion coefficient is more accurate.
Drawings
The above and other features, advantages and aspects of embodiments of the present invention will become more apparent by reference to the following detailed description when taken in conjunction with the accompanying drawings. The same or similar reference numbers will be used throughout the drawings to refer to the same or like elements. It should be understood that the figures are schematic and that elements and components are not necessarily drawn to scale.
FIG. 1 is an exemplary schematic diagram of an intelligent coal charging and discharging system of the present invention.
Detailed Description
The invention will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the invention have been illustrated in the accompanying drawings, it is to be understood that the invention may be embodied in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete. It should be understood that the drawings and embodiments of the invention are for illustration purposes only and are not intended to limit the scope of the present invention.
It should be noted that, for convenience of description, only the portions related to the present invention are shown in the drawings. Embodiments of the invention and features of the embodiments may be combined with each other without conflict.
It should be noted that the terms "first," "second," and the like herein are merely used for distinguishing between different devices, modules, or units and not for limiting the order or interdependence of the functions performed by such devices, modules, or units.
It should be noted that references to "one", "a plurality" and "a plurality" in this disclosure are intended to be illustrative rather than limiting, and those skilled in the art will appreciate that "one or more" is intended to be construed as "one or more" unless the context clearly indicates otherwise.
The names of messages or information interacted between the devices of the present invention are for illustrative purposes only and are not intended to limit the scope of such messages or information.
The invention will be described in detail below with reference to the drawings in connection with embodiments.
As shown in fig. 1, an exemplary architecture diagram of the intelligent coal charging and discharging system of the present invention is shown. The intelligent coal charging and coke discharging system comprises an image enhancement device 101, an infrared positioning device 102, a plurality of image acquisition devices (comprising a vehicle-mounted image acquisition device 103 and a fixed image acquisition device 105) and a computing device 104. The infrared positioning device 102 comprises a plurality of infrared sensors, the infrared sensors are longitudinally arranged on one side of the coal-charging coke pusher, and meanwhile, the infrared sensors are in one-to-one correspondence with a plurality of positioning holes formed in the oven door frame. That is, in an ideal case, the light of each infrared sensor is irradiated in the positioning hole and the measured distance between the sensor and the positioning hole is equal to the preset distance. In practice, for any one infrared sensor, when the reflected light is detected, and the measured distance between the measured sensor and the positioning hole is equal to the preset distance, the infrared sensor can be considered to be triggered. When the number of triggered infrared sensors is greater than the preset number, it can be considered that the oven door is aligned at this time.
As shown in fig. 1, the image enhancing device 101 is mounted on a door frame of a coke oven, for example, the image enhancing device 101 may be a positioning plate of a specified color. Wherein the designated color (e.g., blue) is high in contrast to the color of the coke oven. If necessary, the door frame may be painted with a predetermined color, and the image intensifier 101 may be a door frame body.
The vehicle-mounted image acquisition 103 in the plurality of image acquisition devices is arranged on the coal-charging coke pusher and faces one side of the furnace door frame, the coal-charging coke pusher can run on a track, and the distance between the vehicle-mounted image acquisition 103 and the furnace door frame is detected through a distance sensor on the coal-charging coke pusher. The stationary image acquisition device 105 is mounted near the coke oven.
When the distance is smaller than the preset threshold value, triggering the vehicle-mounted image acquisition 103 to acquire an image of the image enhancement device, and obtaining an enhanced image. The preset threshold value can be preset by a technician, and the coal-charging coke pusher can be controlled to reach a certain range near the coke oven through the preset threshold value. Specifically, the position information of the coke oven can be stored in the control unit of the coal-charging coke pusher in advance, so that the vehicle can travel to a certain range near the coke oven. In addition, whether the distance between the coal-charging coke pusher and the furnace door frame is smaller than a preset threshold can be identified through the driving process image shot by the vehicle-mounted image acquisition 103, and when the local image enhancement device starts to appear in the driving process image or the local image proportion exceeds a certain proportion, the distance between the coal-charging coke pusher and the furnace door frame can be considered to be smaller than the preset threshold.
On this basis, the computing device 104 acquires an enhanced image from the in-vehicle image acquisition 103. And then, determining the quality score of the enhanced image, and if the quality score is greater than or equal to a preset quality score threshold, inputting the enhanced image into a pre-trained positioning network to obtain the position information of the image enhancement device in the enhanced image. The image quality index values of contrast, brightness and the like of the enhanced image can be detected respectively, and the image quality index values are input into an image scoring calculation formula to obtain the quality score of the enhanced image. Wherein the positioning network comprises a target detection network and an offset calculation network. The location network may be any of a variety of object detection networks, such as Yolo (You Only Look Once). The positioning network takes the enhanced image as input, thereby outputting a bounding box of the image enhancing device in the enhanced image, the bounding box being a rectangular box, the position of the rectangular box being determinable by the coordinates of the upper left and lower right corners of the rectangular box. The offset calculation network is used to calculate coordinates of the enhanced center point based on the coordinates of the upper left and lower right corners of the bounding box. And then comparing the center point coordinates with preset reference center point coordinates, and calculating the offset between the center point coordinates and the reference center point coordinates to obtain position offset information. The offset calculation network may employ various networks, such as BP (Back Propagation) networks. In practice, the Yolo network and the BP network can be pre-trained by using corresponding training sample sets respectively according to the needs, so that the prediction accuracy of the network meets the requirements, and the target detection network and the offset calculation network are obtained respectively. For example, the training samples in the corresponding training sample set of the target detection network include sample enhancement images and sample location information.
Further, the computing device 104 may also generate travel control instructions for the coal-charging cart based on the positional offset information. Wherein the positional offset information includes a positional offset value and an offset direction. Thus, the moving distance can be obtained by multiplying the position offset value by a predetermined distance conversion coefficient, and then the offset direction is inverted to obtain the moving direction, for example, the moving direction is offset to the right, and then the moving distance is moved to the left. The moving distance and the moving direction form a running control instruction according to a certain combination format. And then, sending a running control instruction to a control unit of the coal-charging coke pusher so that the control unit controls the coal-charging coke pusher to run to a target position. The control unit of the coal-charging coke pushing vehicle can control the coke pushing rod of the coal-charging coke pushing vehicle and related components to execute a series of operations such as furnace door picking, coke pushing, coal charging, running and the like. The target position is the corresponding position after traveling according to the moving distance and the moving direction. Distance sensors, speed sensors, etc. can be arranged on the coal-charging coke pusher as required to assist the coal-charging coke pusher in traveling to the target position.
If the quality score is smaller than a preset quality score threshold, an infrared starting instruction is generated, and the infrared starting instruction is sent to the control unit, so that the control unit starts the infrared positioning device and receives a feedback signal of the infrared positioning device, and controls the coal-charging coke pushing vehicle to travel to a target position and execute corresponding coal-charging or coke pushing operation based on the feedback signal. Wherein the feedback signal includes a measured distance detected by each of the infrared sensors. Based on this, it can be determined whether each infrared sensor is triggered or not, based on the comparison of the measurement with the preset distance. When the number of the sensors in the triggered state is greater than the preset number, the furnace door can be considered to be aligned at the moment, so that the coal-charging coke pusher can be controlled to travel to the target position and corresponding coal-charging or coke pushing operation can be performed.
In some embodiments, the alignment is realized through two modes of infrared positioning and image positioning, and finally, the accurate alignment of the coal-charging coke pusher and the furnace door is realized, so that the situation of failure in picking the furnace door or failure in pushing coke or charging coal is avoided to the greatest extent. In practice, the inventor finds that because the coke oven frame is connected with other structures such as the oven body, the problem of false identification often occurs in the image positioning process, and the oven door cannot be accurately aligned. Based on the method, the image enhancement device is arranged, and the furnace door frame is coated with the specified color or the locating plate with the specified color is arranged to form clear contrast with the surrounding environment, so that the accuracy of image locating is improved, the accuracy of locating is further improved, and the accurate alignment with the furnace door is realized. In addition, the inventor also finds that the accuracy of image positioning is also affected when special weather such as rainy days or heavy fog is met. Based on the method, the quality scoring is carried out on the shot enhanced image, and when the quality scoring is smaller than a preset quality scoring threshold value, the alignment is realized by utilizing a plurality of infrared sensors and a plurality of positioning holes matched with the infrared sensors, so that the accurate alignment under special weather conditions is further realized.
In some embodiments, the quality score threshold comprises a first quality score threshold and a second quality score threshold, the first quality score threshold being greater than the second quality score threshold. At this time, if the quality score is greater than or equal to a preset quality score threshold, inputting the enhanced image into a pre-trained positioning network to obtain position offset information of the image enhancement device in the enhanced image, including the following steps:
step one, if the quality score is greater than or equal to a first quality score threshold, inputting the enhanced image into a pre-trained positioning network to obtain the position deviation information of the image enhancement device in the enhanced image;
and step two, if the quality score is larger than or equal to the second quality score threshold and smaller than the first quality score threshold, defogging the enhanced image to obtain a defogging image, and inputting the defogging image into a positioning network to obtain the position offset information of the image enhancement device in the enhanced image.
At this time, if the quality score is smaller than a preset quality score threshold, generating an infrared start instruction includes: and if the quality score is smaller than the second quality score threshold value, generating an infrared starting instruction.
In the process, by scoring the enhanced images and setting different quality scoring thresholds, targeted processing of the enhanced images corresponding to the different quality scoring thresholds can be realized. Specifically, for the enhanced image which is greater than or equal to the first quality score threshold, the quality can be considered to be better, and the image recognition can be directly performed, so that the position offset information is obtained; for the quality which is larger than or equal to the second quality grading threshold value and smaller than the first quality grading threshold value, the quality is considered to be acceptable, and image recognition can be performed after defogging treatment, so that position offset information is obtained; for being smaller than the second quality scoring threshold value, the image quality can be considered to be poor, the requirement of image recognition cannot be met, the infrared positioning device can be started, and then alignment is carried out through the infrared positioning device, so that automatic switching and organic combination of two modes of infrared positioning and image positioning are realized, different scene requirements are met, and accurate alignment of the coal-charging coke pusher and the furnace door is further realized.
In some embodiments, in order to further solve the second technical problem described in the background section, that is, "lack of accurate process control measures during the process of coal charging or coke discharging, and thus, abnormal situations in the process cannot be found", the intelligent coal charging and coke discharging system according to some embodiments of the present invention further includes a sensing device. The induction device is buried under the track of the coal-charging coke pusher and is arranged corresponding to the coke oven. In practice, a coke oven may be provided with a corresponding sensing device. The sensing means may be a geomagnetic sensor. When the coal-charging coke pusher arrives at the target position, the sensing device detects that the coal-charging coke pusher arrives and records the starting time, when the coal-charging coke pusher leaves the target position, the sensing device detects that the coal-charging coke pusher leaves and records the leaving time, the starting time, the leaving time and the currently executed operation type are determined as operation record information, and the operation record information is sent to the computing equipment.
On this basis, the computing device matches a corresponding abnormality recognition model from a pool of pre-trained abnormality recognition models according to the operation type included in the operation record information and determines the matched abnormality recognition model as a target abnormality recognition model. When the operation types are different, the mapping relationship between the operation record information and the anomaly identification information is different, so that one anomaly identification model can be trained in advance for each operation type, and one anomaly identification model in the anomaly identification model pool corresponds to one operation type. According to the requirement, the abnormal recognition model in the abnormal recognition model pool can be obtained by training the training sample sets corresponding to different operation types by adopting the same model structure. Of course, a different model structure may be adopted, for example, the model structure may be a comparator, the comparator may determine the residence time according to the input start time and the departure time, compare the residence time with the historical residence mean value, output the abnormality identification information indicating that the abnormality exists if the residence time is greater than the historical residence mean value, and output the abnormality identification information indicating that the abnormality does not exist if the residence time is less than or equal to the historical residence mean value. If the abnormal identification information indicates that no abnormality exists, the residence time corresponding to the abnormal identification information is added into a historical residence time database, an updated historical residence time database is obtained, and the historical residence mean value is updated according to the updated historical residence time database.
Thereafter, the computing device may input the start time and the departure time included in the operation record information into the target abnormality recognition model, generating abnormality recognition information that characterizes whether or not the operation record information is abnormal.
If the abnormality identification information characterizes that the operation record information is abnormal, determining a corresponding historical operation video from a historical operation video library according to the starting time and the leaving time. Wherein, the historical operation video in the historical operation video library is shot by fixed image acquisition equipment installed around the coke oven.
On the basis, analyzing the historical operation video to obtain a key frame group in the historical operation video, and generating abnormal verification information according to the key frame group. Specifically, the method comprises the following steps:
step one, extracting features of each key frame in a key frame group to obtain key frame features corresponding to each key frame;
splicing key frame features corresponding to the key frames respectively to obtain frame group features corresponding to the key frame groups;
step three, word embedding is carried out on the operation type to obtain operation type characteristics;
and fourthly, inputting the spliced frame group characteristics and operation type characteristics into a classifier, and outputting abnormal verification information. The anomaly verification information characterizes whether anomalies exist.
In these embodiments, the start time and the departure time are obtained by setting the sensing device, and thus the operation record information is generated. Thus realizing accurate measurement of the residence time, and the accurate measurement of the residence time can reflect abnormal situations in the coal loading or coke discharging process. Therefore, the computing equipment further carries out abnormality recognition on the operation record information and generates abnormality recognition information, so that abnormal situations in the coal charging or coke discharging process are found in time, the abnormality is eliminated, and normal production operation is ensured.
In some embodiments, in order to further solve the third technical problem described in the background section, that is, "in the process of achieving alignment, a distance conversion coefficient needs to be predetermined, however, the distance conversion coefficient is affected by a series of external environments such as image capturing light, capturing equipment precision, and the like, it is difficult to determine an accurate distance conversion coefficient, so that the coal-charging coke pusher often cannot be moved to a corresponding position accurately. In order to further achieve accurate alignment of the coal-charging cart with the oven door and solve the technical problems described in the background section, in some embodiments of the present invention, the computing device is further configured to: when the indication information sent by the control unit is received, the image acquisition equipment is controlled to acquire the updated enhanced image corresponding to the target position. And determining the moving distance of the coal-charging coke pusher based on the updated enhanced image and the enhanced image. Specifically, the method comprises the following steps:
the method comprises the steps that firstly, a Gaussian differential function is used for carrying out feature detection on an updated enhanced image and an enhanced image respectively to identify feature points with scale and rotation invariance, and a first key point set corresponding to the enhanced image and a second key point set corresponding to the updated enhanced image are obtained;
step two, for a certain first key point in the first key point set, determining the key point closest to the Euclidean distance of the first key point in the second key point set as a matching key point of the first key point, so as to determine a corresponding matching key point for any first key point in the first key point set;
step three, determining whether at least one first key point group with the same corresponding matching key point exists in the first key point set;
step four, if at least one first key point group with the same corresponding matching key point exists, reserving the first key point closest to the matching key point for a plurality of first key points in each first key point group, and deleting the rest first key points to obtain an updated first key point set;
fifthly, determining displacement between the updated enhanced image and the enhanced image according to the updated first key point set and the corresponding matched key point set; specifically, the matched key points can be determined as key point groups, and a pre-trained convolutional neural network is input to obtain the displacement between the updated enhanced image and the enhanced image.
Step six, determining whether the difference value between the displacement and the moving distance is larger than a preset threshold value, and if so, adjusting the distance conversion coefficient according to the displacement. The specific adjustment mode can be determined according to actual needs. For example, a fixed value may be incremented or decremented stepwise.
In some embodiments, the distance scaling factor is checksum adjusted by updating the enhanced image and the enhanced image to determine the displacement and comparing it to the distance moved so that the distance scaling factor is more accurate.
The above description is only illustrative of the few preferred embodiments of the present invention and of the principles of the technology employed. It will be appreciated by persons skilled in the art that the scope of the invention referred to in the present invention is not limited to the specific combinations of the technical features described above, but also covers other technical features formed by any combination of the technical features described above or their equivalents without departing from the inventive concept described above. Such as the above-mentioned features and the technical features disclosed in the present invention (but not limited to) having similar functions are replaced with each other.
Claims (6)
1. An intelligent coal charging and coke discharging system, which is characterized by comprising:
an image enhancement device mounted on a door frame of a coke oven, the image enhancement device for marking a position of the door frame;
the infrared positioning device comprises a plurality of infrared sensors arranged on the coal-charging coke pusher and a plurality of positioning holes arranged on the oven door frame, and the infrared sensors and the positioning holes are in one-to-one correspondence;
the system comprises a plurality of image acquisition devices, a plurality of image acquisition devices and a plurality of image acquisition devices, wherein the plurality of image acquisition devices comprise a vehicle-mounted image acquisition device and a fixed image acquisition device, the vehicle-mounted image acquisition device is arranged on the coal-charging coke pusher, and the vehicle-mounted image acquisition device is used for acquiring images of the image enhancement device when the distance between the coal-charging coke pusher and the furnace door frame is smaller than a preset threshold value, so as to obtain enhanced images;
the computing equipment is in communication connection with the vehicle-mounted image acquisition equipment and is used for acquiring the enhanced image from the vehicle-mounted image acquisition equipment and determining the quality score of the enhanced image, if the quality score is greater than or equal to a preset quality score threshold value, the enhanced image is input into a pre-trained positioning network to obtain the position deviation information of the image enhancement device in the enhanced image, and a running control instruction acting on the coal-charging coke pusher is generated according to the position deviation information; the running control instruction is sent to a control unit of the coal-charging coke pusher so that the control unit controls the coal-charging coke pusher to run to a target position and executes corresponding coal-charging or coke-pushing operation; if the quality score is smaller than a preset quality score threshold, an infrared starting instruction is generated, the infrared starting instruction is sent to the control unit, so that the control unit starts the infrared positioning device and receives a feedback signal of the infrared positioning device, and the coal-charging coke pusher is controlled to travel to a target position and corresponding coal-charging or coke pushing operation is executed based on the feedback signal;
the sensing device is buried below a track of the coal-charging coke pusher and is arranged corresponding to the coke oven, when the coal-charging coke pusher reaches the target position, the sensing device detects that the coal-charging coke pusher reaches and records a starting time, when the coal-charging coke pusher leaves the target position, the sensing device detects that the coal-charging coke pusher leaves and records a leaving time, the starting time, the leaving time and the current operation type are determined to be operation record information, and the operation record information is sent to the computing equipment;
the computing device is further used for matching a corresponding abnormality recognition model from a pre-trained abnormality recognition model pool according to the operation type included in the operation record information, determining the matched abnormality recognition model as a target abnormality recognition model, inputting the start time and the departure time included in the operation record information into the target abnormality recognition model, and generating abnormality recognition information, wherein the abnormality recognition information represents whether the operation record information has an abnormality; if the abnormality identification information characterizes that the operation record information is abnormal, determining a corresponding historical operation video from a historical operation video library according to the starting time and the leaving time; analyzing the historical operation video to obtain a key frame group in the historical operation video, and generating abnormal verification information according to the key frame group.
2. The intelligent coal charging and coke discharging system according to claim 1, wherein the target anomaly identification model is a comparator for determining a residence time based on the start time and the exit time and comparing the residence time with a historical residence mean; outputting abnormal identification information representing that the abnormality exists if the stay time length is larger than the historical stay mean value, and outputting abnormal identification information representing that the abnormality does not exist if the stay time length is smaller than or equal to the historical stay mean value; and if the abnormal identification information indicates that no abnormality exists, adding the corresponding stay time into a historical stay time database to obtain an updated historical stay time database, and updating the historical stay mean value according to the updated historical stay time database.
3. The intelligent coal charging and decoking system of claim 2, wherein the quality score threshold comprises a first quality score threshold and a second quality score threshold, the first quality score threshold being greater than the second quality score threshold; and
if the quality score is greater than or equal to a preset quality score threshold, inputting the enhanced image into a pre-trained positioning network to obtain position offset information of the image enhancement device in the enhanced image, wherein the method comprises the following steps:
if the quality score is greater than or equal to the first quality score threshold, inputting the enhanced image into a pre-trained positioning network to obtain position deviation information of the image enhancement device in the enhanced image;
if the quality score is greater than or equal to the second quality score threshold and smaller than the first quality score threshold, defogging the enhanced image to obtain a defogging image, and inputting the defogging image into the positioning network to obtain position offset information of the image enhancement device in the enhanced image; and
and if the quality score is smaller than a preset quality score threshold, generating an infrared starting instruction, which comprises the following steps:
and if the quality score is smaller than the second quality score threshold value, generating an infrared starting instruction.
4. The intelligent coal charging and coke discharging system according to claim 3, wherein the generating abnormal verification information according to the key frame group comprises:
extracting features of each key frame in the key frame group to obtain key frame features corresponding to each key frame;
splicing key frame characteristics corresponding to each key frame in the key frame group respectively to obtain frame group characteristics corresponding to the key frame group;
word embedding is carried out on the operation type to obtain operation type characteristics;
and after the frame group characteristics and the operation type characteristics are spliced, inputting the spliced frame group characteristics and the operation type characteristics into a classifier, and outputting the abnormal verification information.
5. The intelligent coal charging and discharging system according to claim 4, wherein the positional offset information includes a positional offset value and an offset direction; and
the step of generating a running control instruction acting on the coal-charging coke pusher according to the position offset information comprises the following steps:
determining a moving distance according to the position offset value and a predetermined distance conversion coefficient, and determining a moving direction according to the offset direction, wherein the moving distance and the moving direction form the running control instruction; and
the distance conversion coefficient is updated by the following steps:
acquiring an updated enhanced image corresponding to the target position;
performing feature detection on the updated enhanced image and the enhanced image by using a Gaussian differential function respectively to identify feature points with scale and rotation invariance, so as to obtain a first key point set corresponding to the enhanced image and a second key point set corresponding to the updated enhanced image;
for any first key point in the first key point set, determining a key point with the nearest Euclidean distance with the first key point in the second key point set as a matching key point of the first key point;
determining whether at least one first key point group with the same corresponding matching key point exists in the first key point set;
if at least one first key point group with the same corresponding matching key point exists, reserving first key points closest to the matching key point for a plurality of first key points in each first key point group in the at least one first key point group, and deleting the rest first key points to obtain an updated first key point set;
determining the displacement between the updated enhanced image and the enhanced image according to the updated first key point set and the corresponding matched key point set;
and determining whether the difference value between the displacement and the moving distance is larger than a preset threshold value, and if the difference value between the displacement and the moving distance is larger than the preset threshold value, adjusting the distance conversion coefficient according to the displacement.
6. The intelligent coal charging and decoking system of claim 5, wherein the feedback signal includes a measured distance detected by each of the plurality of infrared sensors; and
the control of the coal-charging coke pusher to travel to a target position and to execute corresponding coal-charging or coke-pushing operations based on the feedback signal includes:
comparing a measured distance with a preset distance for each infrared sensor of the plurality of infrared sensors to determine whether each infrared sensor is triggered; and when the number of the triggered infrared sensors is larger than the preset number, controlling the coal-charging coke pushing vehicle to travel to a target position and executing corresponding coal-charging or coke pushing operation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311327250.1A CN117050760B (en) | 2023-10-13 | 2023-10-13 | Intelligent coal charging and coke discharging system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311327250.1A CN117050760B (en) | 2023-10-13 | 2023-10-13 | Intelligent coal charging and coke discharging system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117050760A CN117050760A (en) | 2023-11-14 |
| CN117050760B true CN117050760B (en) | 2023-12-15 |
Family
ID=88669657
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311327250.1A Active CN117050760B (en) | 2023-10-13 | 2023-10-13 | Intelligent coal charging and coke discharging system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117050760B (en) |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008024869A (en) * | 2006-07-24 | 2008-02-07 | Jfe Steel Kk | Method for controlling movement of coke oven and apparatus for the same |
| KR20100009310A (en) * | 2008-07-18 | 2010-01-27 | 주식회사 포스코 | System and method for controlling travelling position of charging car in coke plant |
| JP2014218615A (en) * | 2013-05-10 | 2014-11-20 | 新日鐵住金株式会社 | Device and method for controlling stopping of moving machine of coke oven |
| CN107987852A (en) * | 2017-12-13 | 2018-05-04 | 天津新港船舶重工有限责任公司 | Radio frequency is combined accurate aligning device with grating technology |
| CN109146963A (en) * | 2017-06-13 | 2019-01-04 | 南京鑫和汇通电子科技有限公司 | One kind being based on the matched image position offsets detection method of swift nature |
| CN111323009A (en) * | 2020-03-09 | 2020-06-23 | 西南交通大学 | Magnetic suspension train positioning method and system |
| WO2020196527A1 (en) * | 2019-03-26 | 2020-10-01 | Ｊｆｅスチール株式会社 | Inspection device and inspection method upon construction of coke oven, and coke oven construction method |
| CN112528786A (en) * | 2020-11-30 | 2021-03-19 | 北京百度网讯科技有限公司 | Vehicle tracking method and device and electronic equipment |
| CN113563905A (en) * | 2021-09-22 | 2021-10-29 | 深圳市信润富联数字科技有限公司 | Coke pusher positioning method and device, coke pusher system, coke pusher and storage medium |
| JP2021187957A (en) * | 2020-05-29 | 2021-12-13 | 住友重機械プロセス機器株式会社 | Coke oven imaging device, coke oven inspection device, image processing device, and industrial facility imaging device |
| CN114854435A (en) * | 2022-06-10 | 2022-08-05 | 山西深蓝海拓智能机电设备有限公司 | Method and device for checking whether furnace door is accurately aligned in coking operation process |
| CN114940913A (en) * | 2022-05-11 | 2022-08-26 | 太原重工股份有限公司 | Remote one-key coking system of coke oven mechanical equipment |
| CN115731380A (en) * | 2022-11-17 | 2023-03-03 | 北京中冶设备研究设计总院有限公司 | Computer vision-based positioning method, control method and device for four coke oven trucks |
| CN115873613A (en) * | 2022-10-08 | 2023-03-31 | 唐山首钢京唐西山焦化有限责任公司 | Coke oven coking process management method and system and electronic equipment |
| CN115984837A (en) * | 2021-10-12 | 2023-04-18 | 安徽工业大学 | Coke oven number online identification system based on deep neural network |
| CN116651306A (en) * | 2023-08-01 | 2023-08-29 | 山西中科冶金建设有限公司 | Intelligent coking coal proportioning system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2016382975A1 (en) * | 2015-12-28 | 2018-07-19 | Suncoke Technology And Development Llc | Method and system for dynamically charging a coke oven |
| CN112528850B (en) * | 2020-12-11 | 2024-06-04 | 北京百度网讯科技有限公司 | Human body identification method, device, equipment and storage medium |
-
2023
- 2023-10-13 CN CN202311327250.1A patent/CN117050760B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008024869A (en) * | 2006-07-24 | 2008-02-07 | Jfe Steel Kk | Method for controlling movement of coke oven and apparatus for the same |
| KR20100009310A (en) * | 2008-07-18 | 2010-01-27 | 주식회사 포스코 | System and method for controlling travelling position of charging car in coke plant |
| JP2014218615A (en) * | 2013-05-10 | 2014-11-20 | 新日鐵住金株式会社 | Device and method for controlling stopping of moving machine of coke oven |
| CN109146963A (en) * | 2017-06-13 | 2019-01-04 | 南京鑫和汇通电子科技有限公司 | One kind being based on the matched image position offsets detection method of swift nature |
| CN107987852A (en) * | 2017-12-13 | 2018-05-04 | 天津新港船舶重工有限责任公司 | Radio frequency is combined accurate aligning device with grating technology |
| WO2020196527A1 (en) * | 2019-03-26 | 2020-10-01 | Ｊｆｅスチール株式会社 | Inspection device and inspection method upon construction of coke oven, and coke oven construction method |
| CN111323009A (en) * | 2020-03-09 | 2020-06-23 | 西南交通大学 | Magnetic suspension train positioning method and system |
| JP2021187957A (en) * | 2020-05-29 | 2021-12-13 | 住友重機械プロセス機器株式会社 | Coke oven imaging device, coke oven inspection device, image processing device, and industrial facility imaging device |
| CN112528786A (en) * | 2020-11-30 | 2021-03-19 | 北京百度网讯科技有限公司 | Vehicle tracking method and device and electronic equipment |
| CN113563905A (en) * | 2021-09-22 | 2021-10-29 | 深圳市信润富联数字科技有限公司 | Coke pusher positioning method and device, coke pusher system, coke pusher and storage medium |
| CN115984837A (en) * | 2021-10-12 | 2023-04-18 | 安徽工业大学 | Coke oven number online identification system based on deep neural network |
| CN114940913A (en) * | 2022-05-11 | 2022-08-26 | 太原重工股份有限公司 | Remote one-key coking system of coke oven mechanical equipment |
| CN114854435A (en) * | 2022-06-10 | 2022-08-05 | 山西深蓝海拓智能机电设备有限公司 | Method and device for checking whether furnace door is accurately aligned in coking operation process |
| CN115873613A (en) * | 2022-10-08 | 2023-03-31 | 唐山首钢京唐西山焦化有限责任公司 | Coke oven coking process management method and system and electronic equipment |
| CN115731380A (en) * | 2022-11-17 | 2023-03-03 | 北京中冶设备研究设计总院有限公司 | Computer vision-based positioning method, control method and device for four coke oven trucks |
| CN116651306A (en) * | 2023-08-01 | 2023-08-29 | 山西中科冶金建设有限公司 | Intelligent coking coal proportioning system |
Non-Patent Citations (2)
| Title |
|---|
| 优化卷积网络及低分辨率热成像的夜间人车检测与识别;于龙姣;于博;李春庚;安居白;;红外技术(07);35-43 * |
| 焦炉车辆连锁***在生产中的应用和改进;李雕;;电子世界(08);32-38 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117050760A (en) | 2023-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107292870B (en) | Track plug pin fault detection method and system based on image alignment and detection network model | |
| CN105185118B (en) | A kind of vehicles peccancy detection means and its method based on video detection | |
| JP2020204804A (en) | Image recognition device | |
| CN111524350B (en) | Method, system, terminal device and medium for detecting abnormal driving condition of vehicle and road cooperation | |
| SE510543C2 (en) | Procedure for associating between target sequences and using the procedure in a radar | |
| CN109447090A (en) | A kind of shield door obstacle detection method and detection system | |
| EP3951741A3 (en) | Method for acquiring traffic state, relevant apparatus, roadside device and cloud control platform | |
| CN117050760B (en) | Intelligent coal charging and coke discharging system | |
| CN114926422B (en) | Method and system for detecting passenger flow of getting on and off vehicles | |
| CN116740833A (en) | Line inspection and card punching method based on unmanned aerial vehicle | |
| CN110660084A (en) | Multi-target tracking method and device | |
| CN113420810A (en) | Cable trench intelligent inspection system and method based on infrared and visible light | |
| CN115683089A (en) | Radar and visual track prediction and correction method | |
| CN113313058B (en) | Unmanned aerial vehicle aerial image target detection method based on self-adaptive model integration | |
| Li et al. | Enhancing feature fusion using attention for small object detection | |
| JPH06199297A (en) | Parking position measuring method | |
| CN111583670B (en) | Method for monitoring overspeed driving by using unmanned aerial vehicle, monitoring system and unmanned aerial vehicle | |
| CN112364854B (en) | Airborne target approaching guidance system and method based on detection, tracking and fusion | |
| CN106874928A (en) | Tracking target the burst automatic decision method of critical event and system | |
| Nakashima et al. | Proposal of a method for estimating the number of passengers with using drive recorder and sensors equipped in buses | |
| CN117315574B (en) | Blind area track completion method, blind area track completion system, computer equipment and storage medium | |
| CN117494029B (en) | Road casting event identification method and device | |
| CN117612047B (en) | Unmanned aerial vehicle inspection image recognition method for power grid based on AI large model | |
| CN113723529B (en) | Traffic information credible identification method based on speed prediction algorithm | |
| CN110991337B (en) | Vehicle detection method based on self-adaptive two-way detection network |
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 |