A method and device for optimizing electric power material detection trajectory

Abstract

The present invention discloses a method and device for optimizing the detection trajectory of electric power materials, the method comprising: collecting mobile video data of materials entering and circulating in a target area; determining the start and end time of material detection in the detection area, and collecting video data after the detection equipment in the detection area is started; optimizing the detection trajectory of electric power materials based on the video data of each detection equipment and the mobile video data of all materials within a preset time period. The present invention accurately records the circulation path and position of materials in the target area, reduces human recording errors, and improves the management efficiency of the target area; and dynamically adjusts the use sequence of the detection equipment, optimizes the material circulation path, improves the detection efficiency, and reduces the detection queue time.

Description

Technical Field

[0001] The present invention belongs to the field of computer technology, and in particular, relates to a method and device for optimizing electric power material detection trajectory. Background Art

[0002] In the power industry, the inspection of power materials is an important part of ensuring the quality of equipment and materials. However, there are many problems with the current management of power material inspection, including inaccurate tracking of material locations, incomplete inspection process records, inefficient use of inspection equipment, and safety hazards during the inspection process.

[0003] Patent CN115271594A discloses a material storage management optimization system including: a collection module: collecting panoramic images in the warehouse, an analysis module: analyzing the panoramic images, analyzing whether the current storage position corresponding to each material is consistent with the preset storage position, if not, obtaining materials inconsistent with the preset storage position as materials to be adjusted, obtaining the current storage position to be adjusted corresponding to the materials to be adjusted, and generating adjustment instructions, a distribution module: based on the adjustment instructions, obtaining the attributes corresponding to the materials to be adjusted, and querying the preset attribute-material storage position data table to determine the target storage position corresponding to the materials to be adjusted, and a reminder module: establishing a material adjustment path according to the current storage position to be adjusted and the target storage position corresponding to the materials to be adjusted, and generating reminder instructions according to the material adjustment path, so as to analyze the classification of materials in the warehouse through image technology to avoid material loss or damage. It only realizes classification and induction, which is convenient for counting the number of materials, but because the power material detection management method not only involves classification and induction, but also mainly lies in the management of detection and detection trajectory, therefore, the existing material storage management optimization system cannot adapt to the current power material detection management. In addition, the existing material inspection management relies on manual records and manual operations, which are prone to recording errors and omissions, making it difficult to accurately track the flow path of materials in the warehouse. At the same time, the manual recording of the start and end time of the inspection is not only time-consuming and labor-intensive, but also prone to errors, affecting the efficiency of the overall inspection process. The use of inspection equipment has not been scientifically optimized, the equipment utilization rate is not high, the inspection queue is long, the waiting time is increased, and the inspection efficiency is reduced. In addition, long operation time or improper operation may cause inspection accidents, and there is a lack of effective monitoring and prevention measures.

[0004] With the increase in the types and quantities of power materials, how to improve detection efficiency and accuracy, reduce human errors, optimize the use of detection equipment, and prevent detection accidents have become urgent issues to be solved. A new method is urgently needed to comprehensively optimize the detection trajectory of power materials through advanced monitoring and optimization technology. Summary of the invention

[0005] In view of the defects existing in the above-mentioned prior art, the present invention provides a method for optimizing the detection trajectory of electric power materials, the method comprising:

[0006] Collect mobile video data of materials entering and circulating in the target area;

[0007] Determine the start and end time of material inspection in the inspection area, and collect video data after the inspection equipment in the inspection area is started;

[0008] Based on the video data of each detection device and the mobile video data of all materials within the preset time period, it is determined whether the preset conditions are triggered, and different optimization models are selected to optimize the power material detection trajectory according to different triggering preset conditions.

[0009] The material has multiple detection steps in the target area, and each detection step corresponds to multiple detection equipment that can be selected to complete the detection in the detection area.

[0010] Among them, the detection trajectory of power materials is optimized, including the adjustment of the usage sequence of detection equipment.

[0011] Among them, before optimizing the detection trajectory of power materials, the manual operation time is corrected in real time based on the machine operation time, and the corrected manual operation time is given.

[0012] Among them, the key node of obtaining the machine operation time reported by the detection equipment;

[0013] According to the key nodes of the machine operation time, determine the start time point and the end time point of the machine operation, and provide the video data of the detection area corresponding to the detection equipment between the start time point and the end time point;

[0014] Perform action recognition analysis on the video data between the start time point and the end time point, and give the recognized manual operation time period;

[0015] Match and calibrate the key nodes of machine operation time with the manual operation time period identified by video;

[0016] Correct the manual operation time of video recognition based on the matching and calibration results.

[0017] The preset condition includes that the corrected manual operation time in any detection area exceeds a first preset threshold, and the preset condition also includes that the sample detection time exceeds a second preset threshold.

[0018] The sample detection time is the time from when the material enters the target area to when the detection is completed. The sample detection time is calculated by recording the time when the material enters the target area, that is, the time when the material is identified, and the end time of the detection.

[0019] Among them, real-time video data of materials entering the target area is obtained to identify the change in the location of the materials and determine the sample detection time;

[0020] After determining that the material has been moved to the detection area and the detection has begun, the time when the key behavior occurs is recorded, and the key behavior at least includes the end of the detection.

[0021] Among them, according to different trigger preset conditions, the system will select different optimization models to optimize the power material detection trajectory, including:

[0022] If the corrected manual operation time in any detection area exceeds a first preset threshold, a first optimization model of the model that optimizes the operation time is selected;

[0023] And, if the sample detection time exceeds a second preset threshold, a second optimization model of the model that optimizes the sample detection time is selected.

[0024] Among them, the power material detection trajectory is optimized, including:

[0025] Predict the time it takes for materials to move from the previous testing device to the next testing device based on historical data;

[0026] According to the current usage of testing equipment, predict the waiting time of materials on each testing equipment;

[0027] Determine the initial solution based on the material movement time and detection waiting time;

[0028] The optimization function of the optimization model determines the final solution based on the initial solution.

[0029] The first optimization model or the second optimization model optimizes the power material detection trajectory, including:

[0030] Predict the time it takes for materials to move from the previous testing device to the next testing device based on historical data;

[0031] According to the current usage of testing equipment, predict the waiting time of materials on each testing equipment;

[0032] Determine the initial solution based on the material movement time and detection waiting time;

[0033] The optimization function of the first optimization model or the second optimization model determines a final solution of the first optimization model or the second optimization model based on the initial solution.

[0034] The initial solution is determined based on the movement time and the detection waiting time, including:

[0035] For each material, a detection device is selected in each detection step, and based on the movement time and the detection waiting time, the total predicted time of the material in the detection device is minimized. The total predicted time satisfies the following relationship:

[0036] S[i][k] = M[i][k] + W[k]

[0037] Where M[i][k] is the predicted time for material i to move from the current location to detection device k, W[k] is the predicted waiting time of detection device k, and S[i][k] is the total predicted time of material i at detection device k;

[0038] The solution process to minimize the predicted movement time and detection waiting time includes:

[0039] Initialize S[i][k] to infinity;

[0040] For each material i, traverse all the detection devices k in each detection step, iteratively calculate the new value S[i][k] S[i][k'], satisfying the following relationship:

[0041] S[i][k'] = min(S[i][k], M[i][k'] + W[k'])

[0042] Where S[i][k'] is the total predicted time of material i at detection device k' after iteration, S[i][k] is the total predicted time of material i at detection device k' before iteration, M[i][k'] is the predicted time for material i to move from the current position to detection device k'. This time is fixed. For a specific material i and detection device k', it is a known value and will not change during the iteration process. W[k'] is the predicted waiting time of detection device k'.

[0043] After each iteration, check whether all S[i][k] and the new value S[i][k'] are the same. If they are the same, that is, all S[i][k'] have maintained the value of the previous iteration, it means that the optimal solution has been reached, and the corresponding detection device k of each material i in each detection step is the initial solution.

[0044] Among them, the first optimization model satisfies the following relationship:

[0045] ;

[0046]

[0047] ;

[0048] ;

[0049] , ;

[0050] ;

[0051] ;

[0052] ;

[0053]

[0054] In the formula, I is a set of materials, and the material number is i; J is a set of detection steps, and the detection step number is j; K is a set of detection equipment, and the detection equipment number is k; The predicted moving time of material i from the previous testing device to the testing device k in testing step j; is the predicted waiting time of the detection device k in the detection step j; is the processing time required for material i to use testing equipment k in testing step j; To detect the maximum capacity of device k; is a binary variable, indicating whether material i is assigned to testing equipment k in testing step j, and its value is 0 or 1; is the time when material i is inspected and completed using inspection equipment k in inspection step j, where k' is the inspection equipment in the previous inspection step; is the total completion time of material i; is the total operation time of the detection device k.

[0055] The optimization function of the first optimization model determines the final solution of the first optimization model based on the initial solution, including:

[0056] A linear programming or integer programming solver is used to solve the first optimization model, with the goal of minimizing the maximum total operating time of all equipment.

[0057] Wherein, the second optimization model satisfies the following relationship:

[0058] ;

[0059]

[0060] ;

[0061] ;

[0062] ;

[0063] ;

[0064] .

[0065] In the formula, The predicted moving time of material i from the current position to the detection device k of detection step j; is the predicted waiting time of detection device k for detection step j; is the processing time of material i on testing device k in testing step j; To detect the capacity of device k; is the weight of the total completion time; The weight of the maximum completion time is a binary variable, indicating whether material i selects detection device k in detection step j, and its value is 0 or 1; It is a continuous variable, indicating the total time that material i spends in detection step j, including moving time, waiting time, and processing time; is the inspection completion time of material i; The maximum completion time for all materials.

[0066] The optimization function of the second optimization model determines the final solution of the second optimization model based on the initial solution, including:

[0067] The second optimization model is solved using a linear programming or integer programming solver, with the goal of minimizing the weighted sum of the total completion time and the maximum completion time of all materials.

[0068] The present invention also provides a device for optimizing the detection trajectory of electric power materials, which adopts the above-mentioned method for optimizing the detection trajectory of electric power materials, and the device comprises:

[0069] A mobile data acquisition module, which is used to collect mobile video data of materials entering and circulating in the target area;

[0070] The detection data acquisition module is used to determine the start and end time of the detection of materials in the detection area, and to collect video data after the detection equipment in the detection area is started;

[0071] The trajectory optimization module is used to optimize the power material detection trajectory based on the video data of each detection device and the mobile video data of all materials within a preset time period.

[0072] The present invention also discloses a power material detection trajectory optimization system, the system includes a memory and a processor, the memory stores a computer program, and the processor is used to execute the computer program to implement the above-mentioned power material detection trajectory optimization method.

[0073] By applying this invention, the following technical effects can be achieved:

[0074] Through video surveillance and real-time tracking technology, the flow path and location of materials in the target area can be accurately recorded, reducing human recording errors and improving the management efficiency of the target area.

[0075] When the materials arrive at the inspection area, the system can automatically record the start and end time of the inspection to ensure the accuracy and timeliness of the records, which helps to optimize the inspection process.

[0076] By analyzing monitoring data and material movement videos within a preset time period, the system can dynamically adjust the usage sequence of detection equipment, optimize the material circulation path, improve detection efficiency, and reduce detection queue time.

[0077] In addition, by optimizing processes and real-time monitoring, detection accidents caused by prolonged operation time can be prevented, improving the safety and reliability of the detection process. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present disclosure will become readily understood. In the accompanying drawings, several embodiments of the present disclosure are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0079] Figure 1 is a flow chart showing a method for optimizing a power material detection trajectory according to an embodiment of the present invention;

[0080] Figure 2 is a schematic diagram showing an optimization device for power material detection trajectory according to an embodiment of the present invention. DETAILED DESCRIPTION

[0081] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0082] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two.

[0083] It should be understood that although the terms first, second, third, etc. may be used to describe ... in the embodiments of the present invention, these ... should not be limited to these terms. These terms are only used to distinguish .... For example, without departing from the scope of the embodiments of the present invention, the first ... may also be referred to as the second ..., and similarly, the second ... may also be referred to as the first ....

[0084] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0085] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0086] It should also be noted that the term "includes", "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, so that a commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprising a ..." do not exclude the existence of other identical elements in the commodity or device including the elements.

[0087] Traditional power material detection trajectory optimization methods rely on manual recording of material locations, which is error-prone and inefficient, resulting in inaccurate tracking of material flows in the target area and incomplete automatic recording of detection start and end times, which may lead to recording errors and affect the efficiency of the overall detection process. At the same time, long operation time may lead to detection accidents, and there is a lack of effective monitoring and optimization methods to prevent these accidents.

[0088] The present invention provides a method for optimizing the detection trajectory of power materials, which aims to automatically record the start and end time of detection through video monitoring and real-time tracking technology, optimize the use sequence (use order) of detection equipment, dynamically adjust the material circulation path, and achieve the improvement of detection efficiency and the safety of the detection process. The application of this method will significantly improve the overall level of power material detection management and meet the needs of the modern power industry for efficient, safe and accurate detection.

[0089] like Figure 1As shown, the present invention discloses a method for optimizing the detection trajectory of electric power materials, comprising:

[0090] Collect mobile video data of materials entering and circulating in the target area; the target area is a warehouse;

[0091] Determine the start and end time of material inspection in the inspection area, and collect video data after the inspection equipment in the inspection area is started;

[0092] Based on the video data of each detection device and the mobile video data of all materials within the preset time period, it is determined whether the preset conditions are triggered, and different optimization models are selected to optimize the power material detection trajectory according to different triggering preset conditions.

[0093] Specifically, when materials enter the target area, the location and status of the materials entering the target area are recorded through video surveillance;

[0094] When materials are transferred within the target area, their locations are tracked in real time based on multiple surveillance cameras within the target area, and mobile videos are saved;

[0095] After the materials arrive at the inspection area, the inspection start and end time will be automatically recorded;

[0096] A surveillance camera is set up in each detection area for the material, and the surveillance camera records the entire video surveillance after the detection equipment in the detection area starts to start;

[0097] Based on the full video monitoring of each detection equipment in the target area within a preset time period and the mobile video of all materials, the detection trajectory of power materials is optimized.

[0098] Among them, the detection trajectory of power materials is optimized, including optimizing the circulation path by adjusting the use sequence of detection equipment, realizing dynamic optimization of optimizing detection efficiency and reducing detection queues, and preventing detection accidents caused by excessive operation time.

[0099] By applying the method of the present invention, the following technical effects can be achieved:

[0100] Through video monitoring and real-time tracking technology, the flow path and location of materials in the target area can be accurately recorded, reducing human recording errors and improving the management efficiency of the target area. When the materials arrive at the inspection area, the system can automatically record the start and end time of the inspection to ensure the accuracy and timeliness of the records, which helps to optimize the inspection process. By analyzing the monitoring data and material movement videos within the preset time period, the system can dynamically adjust the use sequence of the inspection equipment, optimize the material flow path, improve the inspection efficiency, and reduce the inspection queue time. In addition, by optimizing the process and real-time monitoring, it can prevent inspection accidents caused by excessive operation time and improve the safety and reliability of the inspection process.

[0101] Among them, after the optimization process is started, the system will make real-time corrections to the manual operation time based on the machine operation time to obtain the corrected manual operation time. The corrected operation time data is more accurate, which helps to improve the optimization effect.

[0102] After the optimization process is started, the system will make real-time corrections to the manual operation time based on the machine operation time to obtain the corrected manual operation time, which may include:

[0103] The system obtains the key nodes of the machine operation time reported by the detection equipment. These key nodes may include the operation start time, operation end time, and some important intermediate operation time (such as some stage operations of the equipment).

[0104] In the video segment of the full video surveillance where the detection equipment area corresponding to the start time point and the end time point of the machine operation is located, the system obtains the manual operation time period in the video through computer vision and behavior recognition technology.

[0105] For the manual operation time period based on video action recognition statistics, the key nodes of machine operation time are matched and calibrated with the manual operation time period of video recognition. For example, if the start time of a manual operation in video recognition matches the time of a key node of machine operation, then this manual operation time period can be considered accurate; conversely, if a manual operation time period does not match the time of any key node of machine operation, then this time period may need to be corrected or excluded.

[0106] Based on the results of matching and calibration, the system can correct the manual operation time of video recognition. For those time periods that do not match the key nodes of the machine operation time, corrections can be made, such as adjusting the length or position of the time period, or directly excluding it.

[0107] Based on the results of matching and calibration, the system can correct the manual operation time of video recognition in the following ways:

[0108] If there is a slight deviation between the start and end time of the manual operation time period and the key node of the corresponding machine operation, specifically, if the difference between the start and end time of the manual operation time period and the key node of the corresponding machine operation is less than a preset specific value, it may be caused by an error in visual recognition or a delay in the device reporting time. In this case, the manual operation time period can be fine-tuned, such as slightly extending or shortening it to align it with the key node of the machine operation.

[0109] For example, assuming that the time deviation threshold set by the system is 3 seconds, if the start time of a manual operation time period identified from the video is 08:00:05 and the end time is 08:01:05 (i.e., the operation duration is 60 seconds), but the actual start time of the corresponding machine operation key node is 08:00:03 and the end time is 08:01:06 (i.e., the operation duration is 63 seconds). Since the difference between the start and end time of the manual operation time period and the key node time of the machine operation (2 seconds and 1 second) is less than the preset specific value (3 seconds), the system will determine that this may be caused by an error in visual recognition or a delay in the device reporting time. Therefore, the system will fine-tune the manual operation time period and extend it to 63 seconds to align it with the key node of the machine operation.

[0110] If a manual operation time period does not match any key node of machine operation, specifically, the difference between the start and end time of the manual operation time period and the key node of the corresponding machine operation is greater than a preset specific value, it may be because the operation in this time period is not a valid detection operation, such as a worker's non-working behavior or a misoperation of the equipment. In this case, this time period can be excluded and not counted in the total manual operation time.

[0111] For example, assuming that the time deviation threshold set by the system is 3 seconds, if a manual operation time period is identified from the video with a start time of 08:05:00 and an end time of 08:06:00 (i.e., the operation duration is 60 seconds), but there is no corresponding machine operation key node in this time period. In other words, the difference between the start and end time of the manual operation time period and the time of any key node of machine operation is greater than the preset specific value (3 seconds). In this case, the system will determine that the operation in this time period is not a valid detection operation, which may be a non-working behavior of the worker or an erroneous operation of the equipment. Therefore, the system will exclude this time period and will not include it in the total manual operation time.

[0112] Based on computer vision technology, including target detection and behavior recognition, the system first defines and identifies key manual actions, such as moving materials and starting equipment. Then, the system identifies the start and end time points of these key actions in the video, thereby calculating the time of each operation.

[0113] In order to obtain the manual operation time and evaluate the worker's busyness during the preset time period, the algorithm monitors the video stream throughout the process and marks the time period when the worker is operating. The following are the specific steps:

[0114] First, use the object detection algorithm in computer vision (such as YOLO, Faster RCNN) to identify the workers in the video. This step is to determine the location and activity range of the workers in the video.

[0115] Then, a behavior recognition algorithm (such as 3D Convolutional Neural Networks) is used to identify the worker's actions. Some key work actions need to be defined for the detection worker, such as carrying materials, operating detection equipment, etc. If these actions are recognized in the video, it can be considered that the worker is working.

[0116] After that, the total time that the worker is in working state within the preset time period is counted. This time is the manual operation time. Specifically, it is the sum of the duration of all actions of the worker identified within the preset time period.

[0117] In addition, some rules can be set to filter out some non-detection operation time, so as to obtain more accurate manual operation time, including further improving the accuracy of manual operation time by identifying invalid work behaviors. Invalid work behaviors can be non-work-related activities performed by workers during working hours, such as chatting, using mobile phones, taking breaks, etc. These behaviors can be identified by the behavior recognition algorithm 3D CNN algorithm, and these times can be excluded from the preset time period before identifying the time occupied by manual operation.

[0118] After completing the identification of invalid work behaviors, marking the workers' behavior during the operation period can more accurately reflect the work intensity and efficiency of the workers.

[0119] 3D Convolutional Neural Network (3D CNN) is a powerful computer vision model that performs particularly well when processing video data. 3D CNN can process not only spatial information (width and height) in images, but also temporal information, which makes it particularly suitable for video action recognition tasks.

[0120] For example, when 3D CNN performs video behavior recognition tasks, it is necessary to first preprocess the original video, including frame sampling (selecting some frames from the video as input), scale normalization (ensuring that all input video frames are of the same size), color normalization (normalizing pixel values ​​to between 0 and 1), etc.

[0121] And define a 3D CNN model. This model usually includes multiple 3D convolutional layers, pooling layers, and fully connected layers. The 3D convolutional layer is used to extract temporal and spatial features from the input video, the pooling layer is used to reduce the dimension of the features and increase the robustness of the model, and the fully connected layer is used for the final classification.

[0122] Then use the dataset containing the annotated behavior videos to train the 3D CNN model. In this process, the model will learn how to recognize various behaviors from the video. Use the trained 3D CNN model to perform behavior recognition on new videos. The model will output the probability of each behavior, and the behavior with the highest probability can be selected as the recognition result.

[0123] Based on the full video monitoring of each detection device in the target area within a preset time period and the moving video of all materials, the detection trajectory of power materials is optimized, including:

[0124] During the operation, the system monitors the manual operation time, machine reported operation time and material movement time obtained through camera analysis in real time. When a certain optimization trigger target triggers a preset condition, the power material detection trajectory is optimized. The preset condition includes that the corrected manual operation time in a certain detection area exceeds the first preset threshold, and the preset condition also includes that the sample detection time exceeds the second preset threshold.

[0125] Based on multiple surveillance cameras in the target area, the location of materials is tracked in real time and the sample detection time is determined. When materials enter the target area, the materials are identified in the video through target detection algorithms (such as YOLO). Once the materials are identified, their positions need to be tracked in real time. This can be achieved through target tracking algorithms (such as KALMAN filter and MeanShift algorithm). In a multi-camera environment, seamless tracking of materials in different camera fields of view can be achieved through technologies such as camera switching or fusion.

[0126] As materials move into the inspection area and begin to be inspected, it is necessary to identify this behavior and then record the time when each key behavior occurs. This can be achieved through a behavior recognition algorithm (3D CNN), which includes defining a series of key behaviors for the material inspection process, such as when materials are moved into the inspection equipment and when the inspection ends.

[0127] The sample detection time is the time from when the material enters the target area to when the detection is completed. This time can be calculated by recording the time when the material enters the target area (that is, the time when the material is identified) and the end time of the detection (obtained through the behavior recognition algorithm). The specific calculation method is: sample detection time = end time of detection - time when the material enters the target area.

[0128] Optionally, depending on the reason for triggering the optimization, the system will select different optimization models, including:

[0129] If the optimization is triggered because the manual operation time is too long, the first optimization model of the model that focuses on optimizing the operation time will be selected;

[0130] And, if the optimization is triggered due to the long sample detection time, a second optimization model of the model focusing on optimizing the sample detection time will be selected, and the second optimization model focuses on optimizing the movement time and / or the detection waiting time. The sample detection time is the time taken for the material to enter the target area and complete the detection.

[0131] Among them, the model that focuses on optimizing the operation time is the first optimization model, which is triggered by the excessive manual operation time. Among them, the system first determines the initial solution based on the actuation time and the detection waiting time, and then optimizes the expected working time to obtain the optimized final solution, which is applied to the next batch of inspection materials. The AGV equipment determines the movement trajectory of the inspection materials based on the final solution.

[0132] For each batch of inspection materials with multiple inspection operation steps, each inspection operation step corresponds to at least one inspection device, and the first optimization model and the second optimization model both include first determining an initial solution based on the actuation time and the inspection waiting time, and then determining the final solution of the first optimization model or the second optimization model based on the optimization function of the first optimization model or the second optimization model.

[0133] In order to determine the initial solution, it is necessary to analyze past data to predict the average time it takes for each material to move from entering the target area to reaching each detection site in the first detection step, and between each site in the subsequent consecutive steps. This can be done using the historical average value, and in the initial solution determination process, it can be assumed that all materials follow this average time when moving to each point.

[0134] By analyzing the data of the current testing materials, the waiting time of each testing station is predicted, and the waiting time of each testing station is determined based on the waiting queue length of the testing station. The initial solution determination process can assume that the waiting time of the materials at each testing station is based on this waiting time.

[0135] In a target area, a batch of materials needs to be processed. These materials need to go through multiple inspection steps in the target area. There may be multiple optional inspection equipment in each step. First, find an initial solution, that is, select a inspection equipment for each material in each step to minimize the moving time and inspection waiting time. Determining the initial solution includes the following process:

[0136] Step 1: It is necessary to predict the time it takes for materials to move from the previous inspection device to the next inspection device. You can analyze historical data and take the average of the moving time between each pair of devices as the predicted moving time.

[0137] Step 2: It is necessary to predict the waiting time of materials on each testing device. The current equipment usage, such as equipment utilization rate and waiting queue length, can be analyzed to predict the waiting time of the equipment.

[0138] Step 3: Determine the initial solution based on the travel time and the detection waiting time. Specifically, for each material in each step, the device that minimizes the predicted travel time plus the detection waiting time can be selected. In this way, the device selection for each material in each step, i.e., the initial solution, is obtained.

[0139] In step 3, it is necessary to select a detection device for each material in each step so that the predicted movement time and detection waiting time are minimized. This can be achieved by the following formula:

[0140] Let M[i][k] be the predicted time for material i to move from its current location to detection device k, W[k] be the predicted waiting time for detection device k, and S[i][k] be the total predicted time for material i at detection device k, i.e., S[i][k] = M[i][k] + W[k]. The goal is to find a device selection scheme that minimizes the total predicted time for all materials, i.e., minimize sum(S[i][k]).

[0141] The specific calculation process can be implemented using a greedy algorithm, that is, for each material and each step, select the device k that minimizes S[i][k]. This process can be implemented by the following steps: Initialize S[i][k] = inf (infinity). For each material i and each step k, calculate M[i][k] and W[k], and update S[i][k] = min(S[i][k], M[i][k] + W[k]).

[0142] Repeat step 2 until all S[i][k] no longer change.

[0143] In this way, the equipment selection for each material in each step, that is, the initial solution, is obtained.

[0144] The optimization goal of the optimization function of the first optimization model is to minimize the maximum manual operation time corresponding to all detection areas, and the optimization of the first optimization model includes optimizing the order of adjustment operations or adjusting the assigned detection equipment in multiple detection equipment corresponding to the detection operation steps.

[0145] Wherein, the following parameters are defined in the first optimization model:

[0146] Material collection, material number is ;

[0147] A collection of steps, with step numbers ;

[0148] : Device collection, device number is ;

[0149] Supplies Move to step from the previous detection device Medium Equipment The predicted moving time;

[0150] step Medium Equipment The predicted waiting time;

[0151] Supplies In step Use equipment the processing time required;

[0152] equipment Maximum capacity;

[0153] Binary variable, representing material In step Is it assigned to the device? ;

[0154] Supplies In step Use equipment completion time;

[0155] Supplies Total completion time;

[0156] : equipment Total operation time;

[0157] The first optimization model needs to satisfy the following constraints at the same time:

[0158] Constraint 1: Each material can only be assigned to one device at each step: ;

[0159] Constraint 2: Each device can only process materials within its capacity in each time period:

[0160] ;

[0161] Constraint 3: Materials In step Use equipment Completion time:

[0162] , ;

[0163] Constraint 4: Materials The completion time of is the maximum of the completion times of all its steps:

[0164] , ;

[0165] Constraint 5: Ensure that materials are carried out in the correct order of steps:

[0166]

[0167] ;

[0168] In the first optimization model, the optimization goal is to minimize the maximum operation time of all devices, and the optimization function of the first optimization model is:

[0169] ;

[0170] Among them, equipment Operation time is the sum of the processing time of all its assigned tasks:

[0171] , .

[0172] In summary, the first optimization model can be expressed as:

[0173] ;

[0174]

[0175] ;

[0176] ;

[0177] , ;

[0178] ;

[0179] ;

[0180] ;

[0181] ;

[0182] The first optimization model starts to optimize based on the initial solution, and constructs the optimization model according to the initial solution.

[0183] The model is solved using a linear programming or integer programming solver with the goal of minimizing the maximum operating time of all devices, adjusting parameters during the solution process.

[0184] Collect and organize all relevant data, including material collection , Step Collection , Equipment Collection And various parameters , , , determine the capacity constraints of each device and define the decision variables , , , .

[0185] After generating the initial solution through the greedy algorithm, select a suitable solver, such as CPLEX, Gurobi, or SCIP, and set the solver parameters and solution strategy, including setting the time limit, number of iterations, etc., to ensure the solution efficiency. Run the solver to obtain the optimized solution.

[0186] Apply the optimized solution to the actual production process and allocate the next batch of inspection materials to the equipment corresponding to the optimized solution.

[0187] Among them, for the second optimization model focusing on the long moving time or the detection waiting time, when the optimization is triggered due to the long sample detection time, it will focus on optimizing the moving time and / or the detection waiting time.

[0188] The sample detection time is the time taken for the sample to enter the target area and complete the detection.

[0189] In the second optimization model, the following parameters are defined:

[0190] : Material collection, indexed by ;

[0191] : A collection of steps, indexed by ;

[0192] : Device collection, indexed by ;

[0193] : Supplies Move to step from current position Equipment The predicted moving time;

[0194] step Equipment The predicted waiting time;

[0195] Supplies In step Equipment Processing time on

[0196] :equipment Capacity;

[0197] The weight of the total completion time;

[0198] : The weight of the maximum completion time;

[0199] : Binary variable, representing material Is it in step Select device , the value is 0 or 1;

[0200] : Continuous variable, indicating material In step Total time on the site, including travel time, waiting time, and processing time;

[0201] : Supplies completion time;

[0202] : The maximum completion time of all materials.

[0203] The constraints in the second optimization model are as follows:

[0204] Constraint 1: Each material can only select one device in each step: ;

[0205] Constraint 2: The processing volume of each device at each step cannot exceed its capacity: ;

[0206] Constraint three, calculate the total time of materials in each step, including movement time, waiting time and processing time:

[0207] ;

[0208] Constraint 4: The completion time of a material is the total time of all its steps: ;

[0209] Constraint 5: Maximum Completion Time Should be at least equal to the maximum of the completion times for each material: .

[0210] In the second optimization model, the optimization goal is to minimize the weighted sum of the total completion time and the maximum completion time of all materials. The objective function of the second optimization model can be expressed as:

[0211] .

[0212] In summary, the second optimization model can be comprehensively expressed as:

[0213] ;

[0214]

[0215] ;

[0216] ;

[0217] ;

[0218] ;

[0219] .

[0220] And, solving the second optimization model in a similar manner to solving the first optimization model, applying the optimization solution to the actual production process, and allocating the next batch of inspection materials to the equipment corresponding to the optimization solution.

[0221] Therefore, the method and system of the present invention can be dynamically optimized according to real-time data, timely adjust the use of detection equipment and material circulation path, ensure the efficient operation of the detection process, and adapt to the changing detection needs. The present invention significantly improves the management efficiency of the power material detection process and the use efficiency of the detection equipment, reduces human errors and detection waiting time, and improves the safety and reliability of the detection process.

[0222] like Figure 2 As shown, a device for optimizing the detection trajectory of electric power materials adopts the above-mentioned method for optimizing the detection trajectory of electric power materials, and the device includes:

[0223] A mobile data acquisition module, which is used to collect mobile video data of materials entering and circulating in the target area;

[0224] The detection data acquisition module is used to determine the start and end time of the detection of materials in the detection area, and to collect video data after the detection equipment in the detection area is started;

[0225] The trajectory optimization module is used to optimize the power material detection trajectory based on the video data of each detection device and the mobile video data of all materials within a preset time period.

[0226] The present invention also discloses a power material detection trajectory optimization system, the system includes a memory and a processor, the memory stores a computer program, and the processor is used to execute the computer program to implement the above-mentioned power material detection trajectory optimization method.

[0227] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0228] The computer-readable medium may be included in the electronic device, or may exist independently without being installed in the electronic device.

[0229] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0230] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0231] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of a unit does not, in some cases, limit the unit itself.

[0232] The above introduces the preferred embodiments of the present invention, which is intended to make the spirit of the present invention clearer and easier to understand, but is not intended to limit the present invention. All modifications, substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection outlined by the claims attached to the present invention.

Claims

1. A method for optimizing the detection trajectory of electric power materials, characterized in that: The method comprises: Collect mobile video data of materials entering and circulating in the target area; Determine the start and end time of material inspection in the inspection area, and collect video data after the inspection equipment in the inspection area is started; Based on the video data of each detection device and the mobile video data of all materials within a preset time period, determine whether the preset condition is triggered, and select different optimization models to optimize the power material detection trajectory according to different triggering preset conditions, wherein the preset conditions include that the corrected manual operation time in any detection area exceeds the first preset threshold, and the sample detection time exceeds the second preset threshold; Before optimizing the power material detection trajectory, the manual operation time is corrected in real time based on the machine operation time, and the corrected manual operation time is given, including: Obtain the key nodes of the machine operation time reported by the detection equipment; According to the key nodes of the machine operation time, determine the start time point and the end time point of the machine operation, and provide the video data of the detection area corresponding to the detection equipment between the start time point and the end time point; Perform action recognition analysis on the video data between the start time point and the end time point, and give the recognized manual operation time period; Match and align key nodes of machine operation time with identified human operation time periods; The manual operation time of identification is corrected according to the results of matching and calibration.

2. A method for optimizing the detection trajectory of electric power materials according to claim 1, characterized in that: The materials have multiple inspection steps in the target area, and each inspection step has multiple inspection devices corresponding to each step in the inspection area to complete the inspection.

3. A method for optimizing the detection trajectory of electric power materials as claimed in claim 2, characterized in that: Optimize the detection trajectory of power materials, including adjustments to the use sequence of detection equipment.

4. The method for optimizing the detection trajectory of electric power materials according to claim 1, characterized in that: Determination of sample testing time, including: Obtain video data of materials entering the target area in real time to determine the time when the materials enter the target area; Determine, based on the video data after the material enters the target area, the time when a key behavior occurs after the material moves to the detection area and starts to be detected, wherein the key behavior at least includes the end of the detection; The sample detection time is given based on the difference between the detection end time and the time when the materials enter the target area.

5. The method for optimizing the detection trajectory of electric power materials according to claim 1, characterized in that: According to different trigger preset conditions, different optimization models are selected to optimize the power material detection trajectory, including: If the corrected manual operation time in any detection area exceeds a first preset threshold, a first optimization model for optimizing the operation time is selected; And, if the sample detection time exceeds a second preset threshold, a second optimization model for optimizing the sample detection time is selected.

6. A method for optimizing the detection trajectory of electric power materials as claimed in claim 5, characterized in that: Optimize the detection trajectory of power materials, including: Predict the time it takes for materials to move from the previous testing device to the next testing device based on historical data; According to the current usage of testing equipment, predict the waiting time of materials on each testing equipment; Determine the initial solution based on the material movement time and detection waiting time; The optimization function of the optimization model determines the final solution based on the initial solution.

7. A method for optimizing the detection trajectory of electric power materials according to claim 6, characterized in that: Determine the initial solution based on the movement time and detection waiting time, including: Select a detection device for each material in each detection step, and based on the movement time and the detection waiting time, minimize the total predicted time of the material in the detection device, and give the corresponding detection device for each material in each detection step; Among them, the total prediction time satisfies the following relationship: S[i][k] = M[i][k] + W[k] Where M[i][k] is the predicted time for material i to move from the current location to detection device k, W[k] is the predicted waiting time of detection device k, and S[i][k] is the total predicted time of material i at detection device k; The solution process to minimize the total predicted time of materials in the testing equipment includes: Initialize S[i][k] to infinity; For each material i, traverse all the detection devices k in each detection step, iteratively calculate the new value S[i][k] S[i][k'], satisfying the following relationship: S[i][k'] = min(S[i][k], M[i][k'] + W[k']) Where S[i][k'] is the total predicted time of material i at detection device k' after iteration, S[i][k] is the total predicted time of material i at detection device k before iteration, M[i][k'] is the predicted time for material i to move from the current position to detection device k', and W[k'] is the predicted waiting time of detection device k'; After each iteration, check whether all S[i][k] and the new value S[i][k'] are the same. If they are the same, all S[i][k'] maintain the value of the previous iteration, and the optimal solution has been reached. The corresponding detection device k' of each material i in each detection step is the initial solution.

8. A device for optimizing the detection trajectory of electric power materials, characterized in that: The method for optimizing the detection trajectory of electric power materials according to any one of claims 1 to 7 is adopted, wherein the device comprises: A mobile data acquisition module, which is used to collect mobile video data of materials entering and circulating in the target area; The detection data acquisition module is used to determine the start and end time of the detection of materials in the detection area, and to collect video data after the detection equipment in the detection area is started; The trajectory optimization module is used to determine whether to trigger the preset conditions based on the video data of each detection device and the mobile video data of all materials within a preset time period, and select different optimization models to optimize the power material detection trajectory according to the different triggering preset conditions.

Images