Intelligent electric power operation and maintenance management method and system for work order closed-loop processing

Abstract

The invention relates to the technical field of operation and maintenance management, and particularly discloses an intelligent power operation and maintenance management method and system for work order closed-loop processing, and the method comprises the steps: remotely obtaining the main body data of a power main body, recognizing the main body data, and determining the operation correlation degree of the power main body; obtaining a movement track of a worker, and determining the spatial correlation degree of the power main body; clustering the power main body according to the operation relevancy and the space relevancy; when the main body data of any power main body reaches a preset operation and maintenance condition, generating a direct operation and maintenance work order pointing to the power main body and an indirect operation and maintenance work order pointing to other power main bodies of the same kind; according to the method, the data correlation and the space correlation between different power main bodies are obtained, when a certain power main body needs to be operated and maintained, the related power main bodies are operated and maintained at the same time, one-time operation and maintenance are converted into multiple times of homeopathic operation and maintenance on different main bodies, and the utilization rate of operation and maintenance resources is greatly improved.

Description

Technical Field

[0001] This invention relates to the field of operation and maintenance management technology, specifically to an intelligent power operation and maintenance management method and system with closed-loop work order processing. Background Technology

[0002] Traditional power operation and maintenance management suffers from significant manual intervention, complex work order processes, difficulty in tracking execution progress, and low closed-loop efficiency. Especially given the increasing complexity of power systems, the variety of equipment, and the massive workload of operation and maintenance tasks, efficiency is paramount. An inefficient work order management architecture makes it difficult to fully utilize existing operation and maintenance resources. Therefore, improving the utilization rate of operation and maintenance resources is the technical problem that this invention aims to solve. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent power operation and maintenance management method and system for closed-loop processing of work orders, so as to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A smart power operation and maintenance management method for closed-loop processing of work orders, the method comprising:

[0006] Remotely acquire the main data of the power entity, identify the main data, and determine the operational relevance of the power entity;

[0007] Based on the movement trajectory of the staff obtained from the monitoring system, the spatial correlation of the power entity is determined based on the movement trajectory.

[0008] The power system is clustered based on the operational relevance and spatial relevance; wherein the clustering conditions are determined by the amount of operation and maintenance resources.

[0009] When the main data of any power entity reaches the preset operation and maintenance conditions, a direct operation and maintenance work order is generated for that power entity and an indirect operation and maintenance work order is generated for other power entities of the same type; among them, the workload of the indirect operation and maintenance work order is less than that of the direct operation and maintenance work order.

[0010] As a further aspect of the present invention: the step of remotely acquiring the main data of the power entity, identifying the main data, and determining the operational relevance of the power entity includes:

[0011] Establish a connection channel with the power entity to remotely acquire the power entity's main data; the main data is an array containing time tags, and each element in the array corresponds to a data type;

[0012] The main data is identified, and the data of each element at different times is fitted into a curve, retaining the element index to obtain a curve sequence;

[0013] Compare the curve sequences of any two power entities and calculate the curve correlation of the curve corresponding to each element number.

[0014] Query the weights of the elements corresponding to different curves, and sum the curve relevance based on the weights to obtain the running relevance;

[0015] The process of calculating curve correlation includes: calculating the DTW distance between the two curves involved in the calculation, and determining the curve correlation based on the inverse ratio of the DTW distance.

[0016] As a further aspect of the present invention: the step of obtaining the movement trajectory of the staff from the monitoring system and determining the spatial correlation of the power entity based on the movement trajectory includes:

[0017] Establish a connection channel with the monitoring system to acquire scene videos of the power scenario;

[0018] The scene video is subjected to target recognition, target location, and target feature extraction;

[0019] Obtain all positions of the same target feature within a preset time period, fit all positions, and obtain the target's motion trajectory;

[0020] For any two power entities, query their distances to each motion trajectory, and determine the spatial correlation of the power entities based on the queried distances;

[0021] The spatial correlation calculation process is as follows: when the distance between two power entities and a certain motion trajectory is less than a preset distance threshold, the motion trajectory is marked, the number of marked motion trajectories is recorded, and the spatial correlation is determined according to the direct proportion of the number.

[0022] As a further aspect of the present invention: the step of clustering the power entities based on the operational relevance and spatial relevance includes:

[0023] Query the resource quantity of operation and maintenance resources, and determine the number of clusters based on the resource quantity;

[0024] Using the number of clusters as the K value, K-means clustering is performed on the power system based on operational relevance and spatial relevance;

[0025] Real-time query of ongoing maintenance work orders, and update the resource quantity of maintenance resources and the power entities participating in clustering based on the maintenance work orders.

[0026] As a further aspect of the present invention: the step of generating a direct maintenance work order pointing to that power entity and an indirect maintenance work order pointing to other power entities of the same type when the main data of any power entity reaches the preset maintenance conditions includes:

[0027] The first time range is determined based on the current moment and the preset first time span;

[0028] Acquire the main data of each power entity within the first time range and construct an instantaneous data matrix;

[0029] The instantaneous data matrix is ​​input into the trained judgment model, and the judgment status is output; the judgment status includes reached and not reached, which is used to characterize whether the main data of the power entity has reached the preset operation and maintenance conditions.

[0030] When the status is determined to be reached, a direct maintenance work order is generated for the power entity and an indirect maintenance work order is generated for other power entities of the same type.

[0031] Among them, the workload of indirect maintenance work orders is less than that of direct maintenance work orders.

[0032] As a further aspect of the present invention, the method further includes:

[0033] The operation and maintenance process is recorded using a memory device installed on the main power unit.

[0034] Input the operation and maintenance process into a preset quantitative model to obtain the operation and maintenance quantity;

[0035] Statistics on the maintenance volume of the same power entity at different times, and based on the maintenance volume at different times;

[0036] The operation and maintenance volume at different times is fitted to obtain the operation and maintenance volume change function;

[0037] A second time range is determined based on the current moment and a preset second time span. The integral of the operation and maintenance quantity change function is calculated within the second time range, and the upload frequency of the main data is adjusted according to the integral.

[0038] The upload frequency is inversely proportional to the integral.

[0039] The present invention also provides an intelligent power operation and maintenance management system for closed-loop processing of work orders, the system comprising:

[0040] The main data identification module is used to remotely acquire the main data of the power entity, identify the main data, and determine the operational relevance of the power entity.

[0041] The motion trajectory analysis module is used to obtain the motion trajectory of the staff from the monitoring system and determine the spatial correlation of the power entity based on the motion trajectory.

[0042] The power entity clustering module is used to cluster power entities based on the operational relevance and spatial relevance; wherein, the clustering conditions of the clustering process are determined by the resource quantity of operation and maintenance resources;

[0043] The maintenance work order generation module is used to generate direct maintenance work orders for any power entity and indirect maintenance work orders for other power entities of the same type when the main data of any power entity reaches the preset maintenance conditions; wherein, the workload of indirect maintenance work orders is less than that of direct maintenance work orders.

[0044] As a further aspect of the present invention: the main data identification module includes:

[0045] The main data acquisition unit is used to establish a connection channel with the power entity and remotely acquire the main data of the power entity; the main data is an array containing time tags, and each element in the array corresponds to a data type;

[0046] The curve sequence acquisition unit is used to identify the main data, fit the data of each element at different times into a curve, retain the element index, and obtain the curve sequence.

[0047] The correlation calculation unit is used to compare the curve sequences of any two power entities and calculate the curve correlation of the curve corresponding to each element number.

[0048] The relevance accumulation unit is used to query the weights of elements corresponding to different curves, and accumulate the curve relevances based on the weights to obtain the running relevance.

[0049] The process of calculating curve correlation includes: calculating the DTW distance between the two curves involved in the calculation, and determining the curve correlation based on the inverse ratio of the DTW distance.

[0050] As a further aspect of the present invention: the motion trajectory analysis module includes:

[0051] The scene video acquisition unit is used to establish a connection channel with the monitoring system and acquire scene videos of the power scene.

[0052] The target recognition unit is used to perform target recognition on the scene video, locate the target, and extract target features;

[0053] The position fitting unit is used to obtain all positions of the same target feature within a preset time period, fit all positions, and obtain the motion trajectory of the target.

[0054] The distance application unit is used to query the distance between any two power entities and each movement trajectory, and determine the spatial correlation of the power entities based on the queried distances;

[0055] The spatial correlation calculation process is as follows: when the distance between two power entities and a certain motion trajectory is less than a preset distance threshold, the motion trajectory is marked, the number of marked motion trajectories is recorded, and the spatial correlation is determined according to the direct proportion of the number.

[0056] As a further aspect of the present invention: the power entity clustering module includes:

[0057] Query the resource quantity of operation and maintenance resources, and determine the number of clusters based on the resource quantity;

[0058] Using the number of clusters as the K value, K-means clustering is performed on the power system based on operational relevance and spatial relevance;

[0059] Real-time query of ongoing maintenance work orders, and update the resource quantity of maintenance resources and the power entities participating in clustering based on the maintenance work orders.

[0060] Compared with the prior art, the beneficial effects of the present invention are: the present invention obtains the correlation between different power entities in data and in space. When a certain power entity needs operation and maintenance, the related power entities are operated and maintained at the same time, so that one operation and maintenance is transformed into multiple operation and maintenance for different entities in a timely manner, which greatly improves the utilization rate of operation and maintenance resources. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention.

[0062] Figure 1 A flowchart of an intelligent power operation and maintenance management method for closed-loop processing of work orders.

[0063] Figure 2 A structural diagram of the intelligent power operation and maintenance management system for closed-loop processing of work orders. Detailed Implementation

[0064] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0065] Figure 1 A flowchart illustrating an intelligent power operation and maintenance management method for closed-loop work order processing is provided in this embodiment of the invention. The method includes:

[0066] Step S100: Remotely acquire the main data of the power entity, identify the main data, and determine the operational relevance of the power entity;

[0067] The power entity refers to the power equipment used in a power scenario. The power equipment is equipped with sensors and a data transmission module. The sensors are used to collect data, and the data transmission module is used to upload the data to the control center. The control center is the executing entity of the technical solution of this invention. For the control center, remotely acquiring the main data of the power entity and identifying the main data can determine the correlation between different power entities in the data. It focuses on the correlation of the data change process. If the changes of two power entities are similar, they are considered to have a high degree of correlation, which is represented by the parameter of operational correlation.

[0068] Regarding step S100, the step of remotely acquiring the main data of the power entity, identifying the main data, and determining the operational relevance of the power entity includes:

[0069] Establish a connection channel with the power entity to remotely acquire the power entity's main data; the main data is an array containing time tags, and each element in the array corresponds to a data type;

[0070] The main data is identified, and the data of each element at different times is fitted into a curve, retaining the element index to obtain a curve sequence;

[0071] Compare the curve sequences of any two power entities and calculate the curve correlation of the curve corresponding to each element number.

[0072] Query the weights of the elements corresponding to different curves, and sum the curve relevance based on the weights to obtain the running relevance.

[0073] Establish a connection channel with the power system to remotely acquire the power system's main data, which is the sensor data of the sensors built into the power system. The main data is statistically analyzed using arrays, where each element corresponds to a data type. In addition, there is an array for each acquisition time, thus obtaining an array set.

[0074] The main data is identified, and the data of each element at different times is fitted into a curve. The element number is retained to obtain a curve sequence. The curve sequences of any two power entities are compared, and the curve correlation of the curve corresponding to each element number is calculated. Its physical meaning is to determine the data change curve of each sensor, query the weight of the element corresponding to different curves (sensors), and accumulate the curve correlation according to the weight to obtain the operational correlation.

[0075] The above content involves the curve comparison process. The curve correlation calculation process includes: calculating the DTW distance between the two curves involved in the calculation, and determining the curve correlation based on the inverse ratio of the DTW distance; the DTW distance is the distance obtained by applying DTW (Time Warping) algorithm.

[0076] Step S200: Obtain the movement trajectory of the staff from the monitoring system, and determine the spatial correlation of the power main body based on the movement trajectory;

[0077] The monitoring system is a cluster of cameras installed in the power sector for real-time monitoring of the power sector. This is the infrastructure of the power sector, used for managing staff. In this invention's technical solution, staff are granted monitoring permissions by default. In reality, management personnel will explicitly inform staff of the location of the cameras and that their work will be conducted under surveillance. This monitoring is essentially monitoring of public areas. Staff are granted permissions by default, but the acquired video cannot be arbitrarily shared, and the management is fully aware of this. For this invention's technical solution, the execution end is also the management party, which acquires the monitoring video. The acquisition and application of the video also require informing staff. Since this process does not involve privacy, staff generally grant permissions. Therefore, this invention's technical solution assumes that staff have both monitoring and data application permissions.

[0078] By obtaining the movement trajectory of the staff from the monitoring system, it can be determined which power entities the staff frequently move between. If the power entities frequently appear on the same movement trajectory, it can be considered that their work is closely related and has a higher degree of correlation, which is represented by the parameter of spatial correlation.

[0079] It should be noted that spatial correlation is only a weak correlation. It is possible that two devices are only spatially correlated. For example, to reach a certain device, one must pass through another device. This is also considered a correlation. Two devices with high spatial correlation actually indicate that they are related in space, but it does not mean that the two devices are necessarily closely related.

[0080] Regarding step S200, the step of obtaining the movement trajectory of the staff from the monitoring system and determining the spatial correlation of the power entity based on the movement trajectory includes:

[0081] Establish a connection channel with the monitoring system to acquire scene videos of the power scenario;

[0082] The scene video is subjected to target recognition, target location, and target feature extraction;

[0083] Obtain all positions of the same target feature within a preset time period, fit all positions, and obtain the target's motion trajectory;

[0084] For any two power entities, query their distances to each motion trajectory, and determine the spatial correlation of the power entities based on the queried distances;

[0085] The spatial correlation calculation process is as follows: when the distance between two power entities and a certain motion trajectory is less than a preset distance threshold, the motion trajectory is marked, the number of marked motion trajectories is recorded, and the spatial correlation is determined according to the direct proportion of the number.

[0086] The extraction of motion trajectories described above is not complicated and can be accomplished by existing video recognition processes. The method to convert it into spatial correlation is as follows: for any two electrical entities, query their distances to each motion trajectory. When the distance is small enough, it is determined that they appear on the same motion trajectory. Then, count the number of trajectories in which they appear on the same motion trajectory. The spatial correlation is determined based on the direct proportion of the number. That is, the more times two electrical entities appear on the same motion trajectory, the greater their spatial correlation.

[0087] Step S300: Cluster the power entities based on the operational relevance and spatial relevance;

[0088] Operational relevance and spatial relevance are used to characterize the differences between power entities. Another meaning of difference is distance. By using distance to apply clustering algorithms, power entities can be clustered. The clustering conditions in the clustering process are determined by the amount of operation and maintenance resources. This means that how similar each type of power entity is in the clustering results is related to the clustering process. The more resources invested in the clustering process, the more similar the power entities of the same type are. Therefore, the clustering conditions are determined by the amount of operation and maintenance resources. The specific method of determination is not limited in the technical solution of this invention, only that they are related, and the specific degree of correlation is not constrained.

[0089] Regarding step S300, the step of clustering the power entities based on the operational relevance and spatial relevance includes:

[0090] Query the resource quantity of operation and maintenance resources, and determine the number of clusters based on the resource quantity;

[0091] Using the number of clusters as the K value, K-means clustering is performed on the power system based on operational relevance and spatial relevance;

[0092] Real-time query of ongoing maintenance work orders, and update the resource quantity of maintenance resources and the power entities participating in clustering based on the maintenance work orders.

[0093] The above content limits the clustering scheme to K-means clustering, and limits the relationship between resource quantity and clustering conditions to the relationship between resource quantity and number of clusters. The more resources there are, the more clusters there are, and the more accurate the classification is.

[0094] In addition, real-time queries of ongoing maintenance work orders consume maintenance resources during execution, thus requiring updates to resource quantities. The execution process affects a group of power entities, meaning that a group of power entities will be subject to maintenance and will no longer participate in the remaining clustering process.

[0095] Step S400: When the main data of any power entity reaches the preset operation and maintenance conditions, a direct operation and maintenance work order is generated for that power entity and an indirect operation and maintenance work order is generated for other power entities of the same type; wherein, the workload of the indirect operation and maintenance work order is less than that of the direct operation and maintenance work order.

[0096] Identifying the main data of a power entity can determine whether maintenance is required. When the main data of any power entity reaches the preset maintenance conditions, it indicates that maintenance is required, and a direct maintenance work order is generated for that power entity. At the same time, this invention also generates indirect maintenance work orders for other power entities of the same type. The workload of indirect maintenance work orders is less than that of direct maintenance work orders. For example, a direct maintenance work order is a repair task, while an indirect maintenance work order is an inspection task.

[0097] The advantage of this process is that the technical solution of this invention does not generate a single work order, but generates a batch of work orders pointing to related power entities when a problem occurs in a certain power entity. The actual physical meaning is that when a problem occurs in a certain power entity, it is assumed that the related power entities may also have problems. During operation and maintenance, the related power entities are also simplified and maintained. This allows most power entities to undergo multiple advance incidental operation and maintenance, so that operation and maintenance is not carried out only after a problem occurs. This is a preventive architecture.

[0098] Regarding step S400, the step of generating a direct maintenance work order for that power entity and an indirect maintenance work order for other power entities of the same type when the main data of any power entity reaches the preset maintenance conditions includes:

[0099] The first time range is determined based on the current moment and the preset first time span;

[0100] Acquire the main data of each power entity within the first time range and construct an instantaneous data matrix;

[0101] The instantaneous data matrix is ​​input into the trained judgment model, and the judgment status is output; the judgment status includes reached and not reached, which is used to characterize whether the main data of the power entity has reached the preset operation and maintenance conditions.

[0102] When the status is determined to be reached, a direct maintenance work order is generated for the power entity and an indirect maintenance work order is generated for other power entities of the same type.

[0103] Among them, the workload of indirect maintenance work orders is less than that of direct maintenance work orders.

[0104] The above describes the application phase. The first time range is determined based on the current moment and a preset first time span. For example, with the current moment as the last moment, data within one minute is acquired to construct a data matrix. Because it is data within one minute, relative to a long workflow, it is equivalent to an instantaneous data matrix. The instantaneous data matrix is ​​input into a trained judgment model, which outputs a judgment status. The judgment status includes "reached" and "not reached," used to characterize whether the main data of the power entity meets the preset operation and maintenance conditions. The judgment model is a conventional classification model, and the technical solution of this invention will not be elaborated further.

[0105] When the status is determined to be met, a direct maintenance work order is generated for the power entity and an indirect maintenance work order is generated for other power entities of the same type; the workload of the indirect maintenance work order is less than that of the direct maintenance work order.

[0106] As a preferred embodiment of the technical solution of the present invention, the method further includes:

[0107] The operation and maintenance process is recorded using a memory device installed on the main power unit.

[0108] Input the operation and maintenance process into a preset quantitative model to obtain the operation and maintenance quantity;

[0109] Statistics on the maintenance volume of the same power entity at different times, and based on the maintenance volume at different times;

[0110] The operation and maintenance volume at different times is fitted to obtain the operation and maintenance volume change function;

[0111] A second time range is determined based on the current moment and a preset second time span. The integral of the operation and maintenance quantity change function is calculated within the second time range, and the upload frequency of the main data is adjusted according to the integral.

[0112] The upload frequency is inversely proportional to the integral.

[0113] In one embodiment of the technical solution of this invention, a self-regulation scheme applied to a power system is provided. Based on a memory installed on the power system recording the operation and maintenance process, the operation and maintenance process is input into a preset quantification model to obtain the operation and maintenance quantity, indicating when and to what extent the power system performed operation and maintenance. Since staff upload their work type during operation and maintenance, the operation and maintenance quantity corresponding to the work type can be directly queried from a preset table (the table is pre-determined by management personnel for each type of operation and maintenance work). The operation and maintenance quantity of the same power system at different times is statistically analyzed, and the operation and maintenance quantity at different times is fitted to obtain an operation and maintenance quantity change function.

[0114] Finally, a second time range is determined based on the current moment and a preset second time span. For example, 24 hours backward from the current moment is used as the second time range. The integral of the maintenance volume change function is calculated within the second time range. The integral of the maintenance volume itself has no practical meaning because maintenance work is carried out irregularly. That is, there are multiple moments with maintenance volume on the time axis. For moments when no maintenance is performed, the maintenance volume is actually zero. However, after fitting the function, these positions are not zero. At this time, calculating the maintenance volume over a period of time seems to have some meaning. It can be used to characterize the effectiveness of maintenance behavior. The larger the maintenance volume, the larger the integral, and the better the effectiveness. It should be noted that this relationship is not certain. It is only a trend and a high-probability event. For example, replacing parts has a larger maintenance volume than repairing parts, and the effective time should theoretically be longer. However, considering practical factors, special cases may occur, making the effective time shorter. Such special cases are not within the scope of the technical solution of this invention. This invention only considers the correlation under normal conditions.

[0115] Figure 2 The block diagram of the intelligent power operation and maintenance management system for closed-loop processing of work orders is shown. As a preferred embodiment of the technical solution of the present invention, the present invention also provides an intelligent power operation and maintenance management system for closed-loop processing of work orders, wherein the system 10 includes:

[0116] The main data identification module 11 is used to remotely acquire the main data of the power entity, identify the main data, and determine the operational relevance of the power entity;

[0117] The motion trajectory analysis module 12 is used to obtain the motion trajectory of the staff from the monitoring system and determine the spatial correlation of the power body based on the motion trajectory.

[0118] The power entity clustering module 13 is used to cluster power entities based on the operational relevance and spatial relevance; wherein, the clustering conditions of the clustering process are determined by the resource quantity of operation and maintenance resources;

[0119] The operation and maintenance work order generation module 14 is used to generate direct operation and maintenance work orders pointing to the power entity and indirect operation and maintenance work orders pointing to other power entities of the same type when the main data of any power entity reaches the preset operation and maintenance conditions; wherein, the workload of indirect operation and maintenance work orders is less than that of direct operation and maintenance work orders.

[0120] Furthermore, the main data identification module 11 includes:

[0121] The main data acquisition unit is used to establish a connection channel with the power entity and remotely acquire the main data of the power entity; the main data is an array containing time tags, and each element in the array corresponds to a data type;

[0122] The curve sequence acquisition unit is used to identify the main data, fit the data of each element at different times into a curve, retain the element index, and obtain the curve sequence.

[0123] The correlation calculation unit is used to compare the curve sequences of any two power entities and calculate the curve correlation of the curve corresponding to each element number.

[0124] The relevance accumulation unit is used to query the weights of elements corresponding to different curves, and accumulate the curve relevances based on the weights to obtain the running relevance.

[0125] The process of calculating curve correlation includes: calculating the DTW distance between the two curves involved in the calculation, and determining the curve correlation based on the inverse ratio of the DTW distance.

[0126] Specifically, the motion trajectory analysis module 12 includes:

[0127] The scene video acquisition unit is used to establish a connection channel with the monitoring system and acquire scene videos of the power scene.

[0128] The target recognition unit is used to perform target recognition on the scene video, locate the target, and extract target features;

[0129] The position fitting unit is used to obtain all positions of the same target feature within a preset time period, fit all positions, and obtain the motion trajectory of the target.

[0130] The distance application unit is used to query the distance between any two power entities and each movement trajectory, and determine the spatial correlation of the power entities based on the queried distances;

[0131] The spatial correlation calculation process is as follows: when the distance between two power entities and a certain motion trajectory is less than a preset distance threshold, the motion trajectory is marked, the number of marked motion trajectories is recorded, and the spatial correlation is determined according to the direct proportion of the number.

[0132] Furthermore, the power entity clustering module 13 includes:

[0133] Query the resource quantity of operation and maintenance resources, and determine the number of clusters based on the resource quantity;

[0134] Using the number of clusters as the K value, K-means clustering is performed on the power system based on operational relevance and spatial relevance;

[0135] Real-time query of ongoing maintenance work orders, and update the resource quantity of maintenance resources and the power entities participating in clustering based on the maintenance work orders.

[0136] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A smart power operation and maintenance management method with closed-loop processing of work orders, characterized in that, The method includes: Remotely acquire the main data of the power entity, identify the main data, and determine the operational relevance of the power entity; Based on the movement trajectory of the staff obtained from the monitoring system, the spatial correlation of the power entity is determined based on the movement trajectory. The power system is clustered based on the operational relevance and spatial relevance; wherein the clustering conditions are determined by the amount of operation and maintenance resources. When the main data of any power entity reaches the preset operation and maintenance conditions, a direct operation and maintenance work order is generated for that power entity and an indirect operation and maintenance work order is generated for other power entities of the same type; among them, the workload of the indirect operation and maintenance work order is less than that of the direct operation and maintenance work order.

2. The intelligent power operation and maintenance management method for closed-loop processing of work orders according to claim 1, characterized in that, The steps of remotely acquiring the main data of the power entity, identifying the main data, and determining the operational relevance of the power entity include: Establish a connection channel with the power entity to remotely acquire the power entity's main data; the main data is an array containing time tags, and each element in the array corresponds to a data type; The main data is identified, and the data of each element at different times is fitted into a curve, retaining the element index to obtain a curve sequence; Compare the curve sequences of any two power entities and calculate the curve correlation of the curve corresponding to each element number. Query the weights of the elements corresponding to different curves, and sum the curve relevance based on the weights to obtain the running relevance; The process of calculating curve correlation includes: calculating the DTW distance between the two curves involved in the calculation, and determining the curve correlation based on the inverse ratio of the DTW distance.

3. The intelligent power operation and maintenance management method for closed-loop processing of work orders according to claim 1, characterized in that, The step of obtaining the movement trajectory of the staff from the monitoring system and determining the spatial correlation of the power entity based on the movement trajectory includes: Establish a connection channel with the monitoring system to acquire scene videos of the power scenario; The scene video is subjected to target recognition, target location, and target feature extraction; Obtain all positions of the same target feature within a preset time period, fit all positions, and obtain the target's motion trajectory; For any two power entities, query their distances to each motion trajectory, and determine the spatial correlation of the power entities based on the queried distances; The spatial correlation calculation process is as follows: when the distance between two power entities and a certain motion trajectory is less than a preset distance threshold, the motion trajectory is marked, the number of marked motion trajectories is recorded, and the spatial correlation is determined according to the direct proportion of the number.

4. The intelligent power operation and maintenance management method for closed-loop processing of work orders according to claim 1, characterized in that, The step of clustering the power entities based on the operational relevance and spatial relevance includes: Query the resource quantity of operation and maintenance resources, and determine the number of clusters based on the resource quantity; Using the number of clusters as the K value, K-means clustering is performed on the power system based on operational relevance and spatial relevance; Real-time query of ongoing maintenance work orders, and update the resource quantity of maintenance resources and the power entities participating in clustering based on the maintenance work orders.

5. The intelligent power operation and maintenance management method for closed-loop processing of work orders according to claim 4, characterized in that, The step of generating a direct maintenance work order for that power entity and an indirect maintenance work order for other power entities of the same type when the entity's main data reaches the preset maintenance conditions includes: The first time range is determined based on the current moment and the preset first time span; Acquire the main data of each power entity within the first time range and construct an instantaneous data matrix; The instantaneous data matrix is ​​input into the trained judgment model, and the judgment status is output; the judgment status includes reached and not reached, which is used to characterize whether the main data of the power entity has reached the preset operation and maintenance conditions. When the status is determined to be reached, a direct maintenance work order is generated for the power entity and an indirect maintenance work order is generated for other power entities of the same type. Among them, the workload of indirect maintenance work orders is less than that of direct maintenance work orders.

6. The intelligent power operation and maintenance management method for closed-loop processing of work orders according to claim 1, characterized in that, The method further includes: The operation and maintenance process is recorded using a memory device installed on the main power unit. Input the operation and maintenance process into a preset quantitative model to obtain the operation and maintenance quantity; Statistics on the maintenance volume of the same power entity at different times, and based on the maintenance volume at different times; The operation and maintenance volume at different times is fitted to obtain the operation and maintenance volume change function; A second time range is determined based on the current moment and a preset second time span. The integral of the operation and maintenance quantity change function is calculated within the second time range, and the upload frequency of the main data is adjusted according to the integral. The upload frequency is inversely proportional to the integral.

7. An intelligent power operation and maintenance management system for closed-loop processing of work orders, characterized in that, The system includes: The main data identification module is used to remotely acquire the main data of the power entity, identify the main data, and determine the operational relevance of the power entity. The motion trajectory analysis module is used to obtain the motion trajectory of the staff from the monitoring system and determine the spatial correlation of the power entity based on the motion trajectory. The power entity clustering module is used to cluster power entities based on the operational relevance and spatial relevance; wherein, the clustering conditions of the clustering process are determined by the resource quantity of operation and maintenance resources; The maintenance work order generation module is used to generate direct maintenance work orders for any power entity and indirect maintenance work orders for other power entities of the same type when the main data of any power entity reaches the preset maintenance conditions; wherein, the workload of indirect maintenance work orders is less than that of direct maintenance work orders.

8. The intelligent power operation and maintenance management system for closed-loop processing of work orders according to claim 7, characterized in that, The main data identification module includes: The main data acquisition unit is used to establish a connection channel with the power entity and remotely acquire the main data of the power entity; the main data is an array containing time tags, and each element in the array corresponds to a data type; The curve sequence acquisition unit is used to identify the main data, fit the data of each element at different times into a curve, retain the element index, and obtain the curve sequence. The correlation calculation unit is used to compare the curve sequences of any two power entities and calculate the curve correlation of the curve corresponding to each element number. The relevance accumulation unit is used to query the weights of elements corresponding to different curves, and accumulate the curve relevances based on the weights to obtain the running relevance. The process of calculating curve correlation includes: calculating the DTW distance between the two curves involved in the calculation, and determining the curve correlation based on the inverse ratio of the DTW distance.

9. The intelligent power operation and maintenance management system for closed-loop processing of work orders according to claim 7, characterized in that, The motion trajectory analysis module includes: The scene video acquisition unit is used to establish a connection channel with the monitoring system and acquire scene videos of the power scene. The target recognition unit is used to perform target recognition on the scene video, locate the target, and extract target features; The position fitting unit is used to obtain all positions of the same target feature within a preset time period, fit all positions, and obtain the motion trajectory of the target. The distance application unit is used to query the distance between any two power entities and each movement trajectory, and determine the spatial correlation of the power entities based on the queried distances; The spatial correlation calculation process is as follows: when the distance between two power entities and a certain motion trajectory is less than a preset distance threshold, the motion trajectory is marked, the number of marked motion trajectories is recorded, and the spatial correlation is determined according to the direct proportion of the number.

10. The intelligent power operation and maintenance management system for closed-loop processing of work orders according to claim 7, characterized in that, The power entity clustering module includes: Query the resource quantity of operation and maintenance resources, and determine the number of clusters based on the resource quantity; Using the number of clusters as the K value, K-means clustering is performed on the power system based on operational relevance and spatial relevance; Real-time query of ongoing maintenance work orders, and update the resource quantity of maintenance resources and the power entities participating in clustering based on the maintenance work orders.

Images