Method, device and control equipment for determining abnormal operation of power equipment maintenance

Abstract

The application provides an abnormal operation determination method and device for power equipment maintenance and a control device, and belongs to the technical field of power equipment maintenance. The method comprises the following steps: collecting image data of operation behaviors in a maintenance area, and determining a time point of the operation behaviors according to the image data, wherein the time point belongs to any preset time period; determining an abnormality discrimination strategy corresponding to the time point from a first configuration file, wherein each preset time period corresponds to an abnormality discrimination strategy; and outputting key behavior data in the image data according to the abnormality discrimination strategy corresponding to the time point, and determining whether the key behavior data is abnormal. According to the method provided by the application, the combination of fine time period division and targeted discrimination strategy can be used to determine abnormal operation by outputting only the key behavior data in the image data of the operation behaviors according to the discrimination strategy, so that the processing is simple and the time consumption is small.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power equipment maintenance, and in particular to an abnormal operation determination method and device for power equipment maintenance and a control device. BACKGROUND

[0002] With the rapid development of the power industry in China, the problems in the process of power equipment maintenance and maintenance personnel operation are paid more and more attention. Ensuring the compliance and safety of the power equipment maintenance process is crucial to prevent potential accidents and ensure the stable operation of the power system.

[0003] In related technologies, the abnormal operation behavior detection for power maintenance and maintenance mainly includes: target recognition on sample data based on a YOLO algorithm, calculating the distance between the recognized ROI region frame and the camera; processing the ROI region with a distance less than a threshold value, and outputting the behavior data of the target human body; performing abnormal behavior evaluation on the behavior data of the target human body, and when the abnormal evaluation score is greater than a preset threshold value, it is a low-risk early warning, when the next frame behavior is still evaluated as abnormal, it is a medium-risk early warning, and continuous multiple frames of abnormality are determined as a high-risk early warning, and if the evaluation result is normal, the human body abnormal behavior detection is completed. Among them, each illegal behavior is defined and given a corresponding weight, and then a score accumulation system is used to evaluate and calculate the behavior data to determine whether the behavior is normal.

[0004] However, when using the above method to perform abnormal evaluation on each frame of operation behavior, all behavior data needs to be output, which increases the complexity and time consumption of processing, and is not conducive to timely taking preventive measures. SUMMARY

[0005] The present application provides an abnormal operation determination method and device for power equipment maintenance and a control device to solve the problems in related technologies.

[0006] In a first aspect, the present application provides an abnormal operation determination method for power equipment maintenance, comprising:

[0007] Collecting image data of operation behavior in a maintenance area, and determining the time of occurrence of the operation behavior according to the image data, wherein the time belongs to any preset time period;

[0008] Determining the abnormality discrimination strategy corresponding to the time from the first configuration file, wherein each preset time period corresponds to an abnormality discrimination strategy, and the abnormality discrimination strategy is one of the following abnormal behavior categories: operation position abnormality, operation sequence abnormality and operation frequency abnormality;

[0009] According to the abnormality discrimination strategy corresponding to the time, output the key behavior data in the image data, and determine whether the key behavior data is abnormal.

[0010] In a possible implementation, the time point is a starting time point of any preset time period, and before determining the abnormality discrimination strategy corresponding to the time point from the first configuration file, the method further includes:

[0011] obtaining a plurality of historical abnormal behavior categories in each preset time period from a database;

[0012] giving a corresponding weight to each historical abnormal behavior category in the same preset time period through a preset weighted probability calculation model, and accumulating to calculate a respective weighted probability value, determining a historical abnormal behavior category corresponding to a maximum value in the respective weighted probability value as an abnormality discrimination strategy of the preset time period, wherein the weight is inversely proportional to time;

[0013] collecting the abnormality discrimination strategies to form the first configuration file.

[0014] In a possible implementation, the preset weighted probability calculation model is

[0015]

[0016] wherein (t1, t2) is a preset time period, K is a historical abnormal behavior category set in (t1, t2), e -t represents the weight, represents a probability that the historical abnormal behavior category is k at time t, and is calculated through the following model:

[0017]

[0018] wherein X k is a quantity of the historical abnormal behavior category k at time t, and X K is a total sum of the historical abnormal behavior category at time t;

[0019] In a possible implementation, the key behavior data includes one of position information, a target device being contacted, and a speed value.

[0020] In a possible implementation, determining whether the key behavior data is abnormal includes:

[0021] obtaining a first discrimination criterion corresponding to the abnormality discrimination strategy corresponding to the time point from a discrimination criterion library, the first discrimination criterion being one of a preset position range, a preset operation sequence, and a preset frequency range;

[0022] if the key behavior data does not conform to the first discrimination criterion, determining that the operation behavior is abnormal, and issuing a warning, or if the key behavior data conforms to the first discrimination criterion, determining that the operation behavior is normal.

[0023] In a possible implementation, if the key behavior data does not meet the first discrimination criterion, the method further includes:

[0024] retrieving a solution from a preset expert knowledge base.

[0025] In a possible implementation, the database includes artificially diagnosed or automatically diagnosed abnormal operation cases, and the expert knowledge base includes emergency levels, field information and solutions corresponding to the abnormal behavior categories.

[0026] In a second aspect, the application provides an abnormal operation determination apparatus for power equipment maintenance, including a collection module, a determination module and an execution module, wherein:

[0027] The collection module is configured to collect image data of operation behaviors in a maintenance area, and determine a time point of the operation behaviors according to the image data, wherein the time point belongs to any preset time period.

[0028] The determination module is configured to determine an abnormality discrimination strategy corresponding to the time point from a first configuration file, wherein each preset time period corresponds to an abnormality discrimination strategy, and the abnormality discrimination strategy is one of the following abnormal behavior categories: abnormal operation position, abnormal operation sequence and abnormal operation frequency.

[0029] The execution module is configured to output key behavior data in the image data according to the abnormality discrimination strategy corresponding to the time point, and determine whether the key behavior data is abnormal.

[0030] In a third aspect, the application provides a control device, including a memory and a processor, the memory stores program instructions, and the processor is configured to call the program instructions in the memory to execute the abnormal operation determination method for power equipment maintenance according to any one of the first aspect.

[0031] In a fourth aspect, the application further provides a computer readable storage medium, the computer readable storage medium stores computer execution instructions, and the computer execution instructions are configured to be executed by a processor to implement the abnormal operation determination method for power equipment maintenance according to any one of the first aspect.

[0032] The method, device and control equipment for determining abnormal operation of power equipment maintenance provided by the application first acquire image data of operation behavior in the maintenance area, determine the time when the operation behavior occurs according to the image data, and then determine the corresponding discrimination strategy of the time from the first configuration file. The discrimination strategy is one of the following abnormal behavior categories: operation position abnormality, operation sequence abnormality and operation frequency abnormality. Thus, according to the corresponding discrimination strategy of the time, the key behavior data in the image data is output in a targeted manner, and it is determined whether the key behavior data is abnormal, so as to determine whether the operation behavior is correct. The method provided by the application combines the fine time period division with the targeted discrimination strategy. Only the key behavior data in the image data of the operation behavior needs to be output according to the discrimination strategy, which can be used to determine the abnormal operation, is simple to handle and consumes less time, and is beneficial to improve the sensitivity to abnormal operation behavior in the maintenance process, thereby improving the overall operation safety and standardization. BRIEF DESCRIPTION OF DRAWINGS

[0033] The accompanying drawings, which are incorporated herein and constitute part of the specification, illustrate embodiments consistent with the application and, together with the description, serve to explain the principles of the application.

[0034] Figure 1 A flowchart of a method for determining abnormal operation of power equipment maintenance provided by an embodiment of the application;

[0035] Figure 2 A flowchart of another method for determining abnormal operation of power equipment maintenance provided by an embodiment of the application;

[0036] Figure 3 A structure diagram of a device for determining abnormal operation of power equipment maintenance provided by an embodiment of the application;

[0037] Figure 4 A hardware structure diagram of a control equipment provided by an embodiment of the application. DETAILED DESCRIPTION

[0038] In order to make the objects, technical solutions and advantages of the present application clearer, the following will be combined with the accompanying drawings for the preferred embodiments of the present application to make the technical solutions in the embodiments of the present application more clearly and in more detail. In the drawings, the same or similar notations represent the same or similar parts or parts with the same or similar functions throughout. The described embodiments are part of the embodiments of the present application, rather than all the embodiments. The embodiments described below by referring to the drawings are exemplary and are intended to explain the present application, and cannot be understood as a limitation of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application. The embodiments of the present application will be described in detail below with reference to the drawings.

[0039] First of all, it should be noted that in the embodiments of the present application, the terms "comprise", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the sentence "comprises a" does not exclude the presence of another identical element in the process, method, article or device comprising the element.

[0040] Secondly, it should be noted that in the embodiments of the present application, the words "exemplary" or "for example" are used to represent examples, illustrations or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. On the contrary, the words "exemplary" or "for example" are intended to present the relevant concept in a specific way.

[0041] In the related art, a human abnormal behavior detection method includes: performing target recognition on sample data based on a YOLO algorithm, calculating the distance of the recognized ROI region frame from the camera; processing the ROI region whose distance is less than a threshold value, and outputting the behavior data of the target human body; performing abnormal behavior evaluation on the behavior data of the target human body, and when the abnormal evaluation score is greater than a preset threshold value, it is a low-risk early warning, when the next frame behavior is still evaluated as abnormal, it is a medium-risk early warning, and if there are continuous multiple frames of abnormality, it is determined as a high-risk early warning, and if the evaluation result is normal, the human abnormal behavior detection is completed. Wherein, each illegal behavior is defined and given a corresponding weight, and then the score accumulation system is used to evaluate and calculate the behavior data to determine whether the behavior is normal.

[0042] However, when the above method is used to perform abnormality assessment on each frame of operation behavior, all behavior data needs to be fully output on each frame of operation behavior, resulting in complex processing process and time consumption, which affects the timely response to abnormal conditions and the adoption of preventive measures.

[0043] In addition, if the illegal behavior weight distribution is unreasonable, it may cause misjudgment and reduce the accuracy of abnormal behavior detection.

[0044] Therefore, the present application provides an abnormal operation determination method, device and control equipment for power equipment maintenance. First, image data of operation behavior in the maintenance area is collected, and the time when the operation behavior occurs is determined according to the image data. The time belongs to any preset time period. Different abnormality discrimination strategies can be used for operation behaviors occurring at different times. Each preset time period corresponds to an abnormality discrimination strategy. Then, the abnormality discrimination strategy corresponding to the time is determined from the first configuration file. The abnormality discrimination strategy is one of the following abnormal behavior categories: operation position abnormality, operation sequence abnormality and operation frequency abnormality. Finally, the key behavior data in the image data is output in a targeted manner according to the abnormality discrimination strategy corresponding to the time, and it is determined whether the key behavior data is abnormal, so as to determine whether the operation behavior is correct.

[0045] Further, in order to effectively improve the accuracy of abnormal behavior detection, appropriate abnormality discrimination strategies need to be set for the time when the operation behavior occurs. By taking the time when the operation behavior occurs as the starting time of any preset time period, first, the multiple historical abnormal behavior categories and their occurrence probabilities that have occurred in the time period are obtained and analyzed. Then, considering the influence of time factor on the reference value of abnormal behavior, that is, when the operation behavior is discriminated, the historical abnormal behavior category consistent or close to the time when the operation behavior occurs is selected as the standard. Therefore, by introducing a time weight, the occurrence probability of each historical abnormal behavior category is weighted and processed, and then the weighted probability is calculated. Finally, the historical abnormal behavior category with the maximum probability after weighting is selected as the abnormality discrimination strategy for judging abnormal operation.

[0046] By the above method, the processing flow can be simplified and the time consumption can be reduced, thereby facilitating to improve the recognition speed and accuracy of abnormal operation behavior in the maintenance process, and improving the safety and standardization level of the whole operation.

[0047] Figure 1 A flowchart of an abnormal operation determination method for power equipment maintenance provided by an embodiment of the present application is shown in Figure 1 The method can include:

[0048] S101. Collect image data of the operation behavior in the inspection and maintenance area, and determine the time when the operation behavior occurs based on the image data, wherein the time belongs to any preset time period.

[0049] In the context of power equipment maintenance, image data refers to video streams or still images captured by cameras installed at the work site. The information contained in this image data can include operator location, operational procedures, frequency of operation, tool usage, environmental conditions, and the implementation status of safety measures.

[0050] In practice, monitoring cameras or other image acquisition devices installed in the power equipment maintenance area can be used to acquire image data of operators' actions in real time or periodically.

[0051] The image data of the operation behavior is any single frame image in the video stream that contains operation behavior information. The operation behavior information may specifically include: the operator's location information, operation sequence, operation frequency and operation order, and tool usage.

[0052] By timestamping the image sequence of consecutive frames with precision down to the frame level, and combining image processing techniques and computer vision algorithms, the timing of each operation can be identified and located.

[0053] For example, the operation may occur within any preset time period. The specific division of the preset time period can be set and adjusted according to the actual operation of the power plant. For example, three time periods can be set: morning (6:00-12:00), afternoon (12:00-18:00), and evening (18:00-6:00). In addition, twenty-four time periods can be set, that is, each hour from 0:00 to 24:00 is a time period.

[0054] S102. Determine the anomaly detection strategy corresponding to the time from the first configuration file. Each preset time period corresponds to an anomaly detection strategy. The anomaly detection strategy is one of the following abnormal behavior categories: abnormal operation location, abnormal operation sequence, and abnormal operation frequency.

[0055] In other words, once the specific time when a certain operation occurs is determined, the anomaly detection strategy within the preset time period of that time can be retrieved from the first configuration file.

[0056] It should be noted that in actual implementation, the abnormality discrimination strategy corresponding to the preset time period can be set based on historical data statistical analysis, job specification requirements, expert experience and other factors. Each preset time period corresponds to an abnormality discrimination strategy, such as operation position abnormality, operation sequence abnormality or operation frequency abnormality. In this way, the selected abnormality discrimination strategy has good reliability, and it is convenient to determine whether the operation behavior is abnormal based on the abnormality discrimination strategy.

[0057] It can be understood that reasonable division of the preset time period helps to apply different abnormality discrimination strategies, thereby improving the accuracy and real-time performance of abnormal behavior detection.

[0058] For example, in the first configuration file, different abnormality discrimination strategies can be configured for the morning (6:00-12:00), afternoon (12:00-18:00) and evening (18:00-6:00) time periods. In other examples, an abnormality discrimination strategy can be set for each hour (0:00-24:00).

[0059] S103, output the key behavior data in the image data according to the abnormality discrimination strategy corresponding to the time, and determine whether the key behavior data is abnormal.

[0060] That is, according to the abnormality discrimination strategy corresponding to the time when the operation behavior occurs determined in step S102, the key behavior data related to the current abnormality discrimination strategy needs to be output from the image data of the operation behavior. For example, if the abnormality discrimination strategy is operation position abnormality, the position information of the operator in the image needs to be focused on; if the abnormality discrimination strategy is operation sequence abnormality, the execution sequence information of the operation steps in the image needs to be focused on; if the abnormality discrimination strategy is operation frequency abnormality, the number of times or interval information of the operation action in the image needs to be focused on.

[0061] It can be understood that the collected image data of the operation behavior can be combined with computer vision, machine learning and pattern recognition techniques to output the key behavior data related to the current abnormality discrimination strategy.

[0062] Further, operation position abnormality indicates that the human operation position is inconsistent with the preset position, operation sequence abnormality indicates that the human operation sequence is inconsistent with the preset sequence, and operation frequency abnormality indicates that the human operation frequency is outside the preset frequency range. Based on this, it can be determined whether the key behavior data is abnormal, and further determine whether the operation behavior is correct.

[0063] The abnormal operation determination method for power equipment maintenance of the embodiment of the application determines the abnormal operation by combining the fine time period division and the targeted determination strategy, and only needs to output the key behavior data in the image data of the operation behavior according to the determination strategy, so as to determine the abnormal operation, which is simple in processing and low in time consumption, and is beneficial to improving the sensitivity to the abnormal operation behavior in the maintenance process, thereby improving the overall operation safety and standardization.

[0064] In some embodiments, before determining the abnormal determination strategy corresponding to the time point from the first configuration file, the first configuration file can be reasonably set to further improve the accuracy of abnormal behavior detection. Figure 2 A flowchart of an abnormal operation determination method for power equipment maintenance provided by the embodiment of the application is shown in FIG. 1. Figure 2 The method can include:

[0065] S201, image data of operation behavior in a maintenance area is collected, and a time point of occurrence of the operation behavior is determined according to the image data, wherein the time point is a starting time point of any preset time period.

[0066] Specifically, when the preset time period is specifically divided, the time point of occurrence of the operation behavior is the starting time point of any preset time period. That is, the time points of occurrence of the operation behavior of two adjacent frames are adjacent, and there is a part of the time points overlapping between the preset time periods corresponding to the two.

[0067] S202, a plurality of historical abnormal behavior categories in each preset time period are obtained from a database.

[0068] Here, the historical abnormal behavior categories are consistent with the foregoing embodiments, including operation position abnormality, operation sequence abnormality and operation frequency abnormality.

[0069] Illustratively, the database includes artificial abnormal operation cases diagnosed by human or automatically diagnosed. Thus, each historical abnormal behavior category appearing in each preset time period is obtained therefrom.

[0070] It can be understood that according to different historical situations, each preset time period can include one or more of operation position abnormality, operation sequence abnormality and operation frequency abnormality, and each type can have one or more.

[0071] S203, a preset weighted probability calculation model is used to give a corresponding weight to each historical abnormal behavior category in the same preset time period and to calculate a weighted probability value of each, and a historical abnormal behavior category corresponding to a maximum value in the weighted probability values is determined as an abnormal determination strategy of the preset time period, wherein the weight is inversely proportional to the time.

[0072] The step can be understood as introducing a weighted probability calculation model that combines time correlation and frequency statistical factors to weight the historical abnormal behavior categories, ensuring that the historical abnormal behavior categories closer to the time when the collected operation behavior occurs obtain higher weights.

[0073] In this way, by calculating the weighted probability values of the operation position abnormality, operation sequence abnormality and operation frequency abnormality respectively, the historical abnormal behavior category with the maximum weighted probability value is selected as the abnormality discrimination strategy in the preset time period, so as to more accurately reflect the abnormal situation that may occur in the actual environment.

[0074] In a specific example, the preset weighted probability calculation model refers to

[0075]

[0076] where (t1, t2) is the preset time period, K is the set of historical abnormal behavior categories in (t1, t2), e -t represents the weight, represents the probability of the historical abnormal behavior category being k at time t, which is calculated by the following model:

[0077]

[0078] where X k is the number of historical abnormal behavior categories k at time t, and X K is the sum of historical abnormal behavior categories at time t.

[0079] S204, form a first configuration file by collecting the abnormality discrimination strategies.

[0080] That is, through the above steps, an optimized abnormality discrimination strategy is obtained for each preset time period. The optimal abnormality discrimination strategies corresponding to each time period are uniformly collected together to form a complete set of abnormality discrimination strategies including all preset time periods, i.e. the first configuration file.

[0081] Therefore, for the image data of the collected operation behavior at different times, the corresponding abnormality discrimination strategy can be determined from the first configuration file to determine whether the operation behavior is correct.

[0082] S205, determine the abnormality discrimination strategy corresponding to the time from the first configuration file, wherein each preset time period corresponds to an abnormality discrimination strategy, and the abnormality discrimination strategy is one of the following abnormal behavior categories: operation position abnormality, operation sequence abnormality and operation frequency abnormality.

[0083] It should be noted that the execution process of step S205 can refer to the execution process of S102, which will not be described here.

[0084] S206, according to the abnormality discrimination strategy corresponding to the time, output the key behavior data in the image data, and determine whether the key behavior data is abnormal.

[0085] It should be noted that the execution process of step S206 can refer to the execution process of S103, which will not be described here.

[0086] The abnormal operation determination method for power equipment maintenance of the embodiments of the application determines the abnormality discrimination strategy through analysis of historical abnormal behavior categories and application of a reasonable weighting mechanism, which is beneficial to improve the accuracy and reliability of abnormal operation judgment in the process of power equipment maintenance.

[0087] In some embodiments, the key behavior data includes one of position information, a target device being contacted, and a frequency value.

[0088] For example, when the abnormal behavior strategy is operation position abnormality, position information needs to be extracted from the image data of the operation behavior. Here, the key parts of the operator and the equipment can be recognized in combination with target detection and positioning technology (such as YOLO, Faster R-CNN, etc.), and then the relative position coordinates or distance information of the operator relative to the key parts of the equipment are calculated.

[0089] In a specific example, for the image data of the real-time collected operation behavior, there are two cases of operation behavior abnormality at the current time and operation behavior normality at the current time. In order to ensure that the work personnel can maintain and repair in a standard manner, response measures need to be taken for these two cases to improve the safety of the work process. Therefore, in some embodiments, determining whether the key behavior data is abnormal includes:

[0090] S1, obtaining a first discrimination criterion corresponding to the abnormality discrimination strategy corresponding to the time from the discrimination standard library, the first discrimination criterion being one of a preset position range, a preset operation sequence, and a preset frequency range.

[0091] This step can be understood as that different abnormality discrimination strategies correspond to different discrimination criteria. For example, when the abnormality discrimination strategy corresponding to the time of the operation behavior is operation position abnormality, the first discrimination criterion obtained from the pre-constructed discrimination standard library is a preset position range.

[0092] The preset position range can be set according to the actual maintenance work, which is not limited by the application.

[0093] S2. If the key behavioral data does not meet the first discrimination criterion, determine that the operation behavior is abnormal and issue an early warning; or, if the key behavioral data meets the first discrimination criterion, determine that the operation behavior is normal.

[0094] For example, if an operator's position exceeds a preset safe area, it is determined that critical behavioral data is abnormal, meaning the operational behavior does not conform to specifications and an error has occurred. An alarm or indicator light can then be activated to alert the operator to suspend work.

[0095] If the operator is within the preset reasonable area, it indicates that the operation during the inspection and maintenance process is compliant and safe. In this case, the alarm or indicator light will not issue a warning, thereby avoiding unnecessary interference and false alarms.

[0096] As mentioned above, after identifying anomalies in critical behavioral data and issuing an alert, operators can be prompted to suspend operations. However, safety hazards arising in the maintenance area due to these operational anomalies must be addressed promptly to ensure the safety of the maintenance area. Therefore, in some examples, if the critical behavioral data does not meet the first criterion, the following also applies:

[0097] Retrieve solutions from a pre-defined expert knowledge base.

[0098] In practice, the expert knowledge base includes the emergency level, on-site information, and solutions corresponding to the abnormal behavior category.

[0099] For example, emergency response levels can be categorized into different levels based on the severity of the abnormal behavior and the potential risk level, enabling the rapid activation of corresponding emergency plans. On-site information provides background information such as on-site environmental conditions and equipment status related to the abnormal behavior, helping operators better understand the context of the problem and thus take more accurate and effective countermeasures.

[0100] Solutions can include operating guidelines, troubleshooting steps, safety precautions, and procedures for restoring normal operations, to guide on-site personnel to quickly restore operational standards and effectively resolve safety hazards caused by abnormal behavior.

[0101] See Figure 3 As shown, this application provides an abnormal operation determination device 30 for power equipment maintenance, including a data acquisition module 31, a determination module 32, and an execution module 33, wherein:

[0102] The acquisition module 31 is used to acquire image data of the operation behavior in the maintenance area, and determine the time when the operation behavior occurs based on the image data, wherein the time belongs to any preset time period.

[0103] The determination module 32 is used to determine the anomaly detection strategy corresponding to the time from the first configuration file. Each preset time period corresponds to an anomaly detection strategy, and the anomaly detection strategy is one of the following abnormal behavior categories: abnormal operation location, abnormal operation sequence, and abnormal operation frequency.

[0104] The execution module 33 is used to output key behavioral data in the image data according to the anomaly discrimination strategy corresponding to the time, and to determine whether the key behavioral data has anomalies.

[0105] The abnormal operation determination device 30 for power equipment maintenance and repair in this application embodiment can execute the technical solution of the abnormal operation determination method for power equipment maintenance and repair in the above method embodiment. Its implementation principle and technical effect are similar, and will not be repeated here.

[0106] Figure 4 A schematic diagram of the control device provided in an embodiment of this application. (See attached diagram.) Figure 4 The control device 40 includes a processor 41 and a memory 42. The memory 42 is used to store computer programs, and the processor 41 is used to execute the computer programs stored in the memory 42 to implement the abnormal operation determination method for power equipment maintenance as shown in any of the above method embodiments.

[0107] Specifically, the processor 41 and the memory 42 can communicate; for example, the processor 41 and the memory 42 communicate via a communication bus 43.

[0108] For example, the control device 40 may also include a communication interface, which may include a transmitter and / or a receiver.

[0109] For example, the processor 41 described above can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this application can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor.

[0110] This application also provides a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are executed by a processor, the computer performs the above-described method for determining abnormal operations during the maintenance of power equipment.

[0111] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0112] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0113] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in a combination of hardware and software functional units.

[0114] The integrated units implemented as software functional units described above can be stored in a computer-readable storage medium. These software functional units, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0115] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is merely an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0116] The technical solutions of this application have been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of this application is obviously not limited to these specific embodiments. The above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for determining abnormal operations during the inspection and maintenance of power equipment, characterized in that, include: Image data of operational behaviors in the inspection and maintenance area are collected, and the time when the operational behavior occurs is determined based on the image data, wherein the time is the start time of any preset time period; The anomaly detection strategy corresponding to the time is determined from the first configuration file. Each preset time period corresponds to an anomaly detection strategy, and the anomaly detection strategy is one of the following abnormal behavior categories: abnormal operation location, abnormal operation sequence, and abnormal operation frequency. Based on the anomaly detection strategy corresponding to the time, output the key behavioral data in the image data to determine whether the key behavioral data is abnormal. Before determining the anomaly detection strategy corresponding to the time from the first configuration file, the method further includes: Retrieve multiple historical abnormal behavior categories within each preset time period from the database; By using a preset weighted probability calculation model, each of the historical abnormal behavior categories within the same preset time period is assigned a corresponding weight and their respective weighted probability values ​​are calculated cumulatively. The historical abnormal behavior category corresponding to the maximum value among the weighted probability values ​​is determined as the abnormality discrimination strategy for the preset time period. The weights are inversely proportional to time. The set of all the aforementioned anomaly detection strategies is used to form the first configuration file; The preset weighted probability calculation model refers to Where (t1, t2) is a preset time period, K is the set of historical abnormal behavior categories within (t1, t2), and e -t Indicates weight, This represents the probability that the historical abnormal behavior category is k at time t. The following calculations were performed using the model: Among them, X k Let X be the number of historical abnormal behaviors of category k at time t. K This represents the sum of historical abnormal behavior categories at time t.

2. The method according to claim 1, characterized in that, The key behavioral data includes one of the following: location information, the target device being contacted, and speed value.

3. The method according to claim 2, characterized in that, Determining whether the key behavioral data is abnormal includes: Obtain a first discrimination criterion corresponding to the anomaly discrimination strategy corresponding to the time from the discrimination criterion library. The first discrimination criterion is one of a preset location range, a preset operation sequence, and a preset frequency range. If the key behavioral data does not conform to the first discrimination criterion, it is determined that the operation behavior is abnormal and an early warning is issued; or, if the key behavioral data conforms to the first discrimination criterion, it is determined that the operation behavior is normal.

4. The method according to claim 3, characterized in that, If the key behavioral data does not meet the first discrimination criterion, the following further applies: Retrieve solutions from a pre-defined expert knowledge base.

5. The method according to claim 4, characterized in that, The database includes cases of abnormal human operations that have been diagnosed manually or automatically. The expert knowledge base includes emergency levels, on-site information, and solutions corresponding to the abnormal behavior categories.

6. A device for determining abnormal operations during the maintenance and repair of power equipment, characterized in that, include: The module consists of a data acquisition module, a determination module, and an execution module, among which: The acquisition module is used to acquire image data of operational behavior in the maintenance area, and determine the time when the operational behavior occurs based on the image data, wherein the time is the start time of any preset time period; The determining module is used to determine the discrimination strategy corresponding to the time from the first configuration file, wherein each preset time period corresponds to a discrimination strategy, and the discrimination strategy is one of the following abnormal behavior categories: abnormal operation position, abnormal operation sequence, and abnormal operation frequency; The execution module is used to determine whether the collected image data has an operational anomaly based on the discrimination strategy corresponding to the time. The acquisition module is also used to obtain multiple historical abnormal behavior categories within each preset time period from the database; The determining module is further configured to assign corresponding weights to each of the historical abnormal behavior categories within the same preset time period through a preset weighted probability calculation model and accumulate and calculate their respective weighted probability values, determine the historical abnormal behavior category corresponding to the maximum value among the weighted probability values ​​as the anomaly discrimination strategy for the preset time period, wherein the weights are inversely proportional to time; and form the first configuration file by combining the anomaly discrimination strategies. The preset weighted probability calculation model refers to Where (t1, t2) is a preset time period, K is the set of historical abnormal behavior categories within (t1, t2), and e -t Indicates weight, This represents the probability that the historical abnormal behavior category is k at time t. The following calculations were performed using the model: Among them, X k Let X be the number of historical abnormal behaviors of category k at time t. K This represents the sum of historical abnormal behavior categories at time t.

7. A control device, characterized in that, The device includes a memory and a processor, wherein the memory stores program instructions, and the processor is used to call the program instructions in the memory to execute the abnormal operation determination method for power equipment maintenance as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the abnormal operation determination method for power equipment maintenance as described in any one of claims 1 to 5.

Images