Substation equipment operation inspection device and health analysis method

Abstract

The invention relates to the technical field of substation operation inspection, in particular to a substation equipment operation inspection device and a health analysis method, and the method comprises the steps: S1, configuring an inspection task according to a demand, and returning the inspection data of an inspection target in real time; s2, performing abnormity judgment on image video data, temperature data and partial discharge data of the inspection target based on the inspection data to obtain an image video data abnormity rating, a temperature data abnormity rating and a partial discharge data abnormity rating; s3, performing comprehensive judgment on all the abnormal ratings to obtain the health state of the inspection target; s4, when the health state is assessed to be sub-health, predicting the running state trend of the inspection target based on a neural network model, comparing and analyzing the predicted running state trend of the inspection target with the running state trend in the health state, and performing active early warning on an analysis result at a monitoring terminal; and when the health state is evaluated to be abnormal, the fault state of the inspection target is judged based on the abnormal rating, and the monitoring terminal sends out a fault alarm.

Description

Technical Field

[0001] The present invention relates to the technical field of substation operation and inspection, and in particular to a substation equipment operation and inspection device and a health analysis method. Background Art

[0002] Substations are critical facilities within power systems used for transforming voltage, receiving and distributing electrical energy, controlling power flow, and adjusting voltage. Substation maintenance and inspection refers to the operation and overhaul of substations. This primarily involves inspecting, monitoring, maintaining, overhauling, and commissioning equipment within the substation to ensure the proper functioning of substation equipment and the safe and stable power supply of the power grid. The traditional substation maintenance and inspection process involves developing an inspection plan, having maintenance personnel prepare for inspections, conducting inspections, recording and reporting equipment defects and failures, and maintaining equipment. As the contradiction between the increasing number of substations and the shrinking number of maintenance personnel becomes increasingly prominent, traditional power operations, maintenance, and management models are no longer adequate for the rapidly evolving needs of the power grid.

[0003] For example, a patrol management system and patrol management method with publication number CN116186616A includes: a task management module for formulating patrol tasks; an execution management module for receiving patrol tasks, instructing the patrol operation end to perform patrols according to the patrol tasks, and receiving patrol fault events reported by the patrol operation end; an event management module for receiving patrol fault events sent by the execution management module, sending a processing task work order to the maintenance operation end according to the fault event level, instructing the maintenance operation end to perform fault processing, and reporting the processing results; an experience management module for receiving the processing results sent by the event management module, and closing the processing task work order according to the processing results. The present invention can realize the integration of various modules, improve patrol efficiency, send a processing task work order to the maintenance operation end according to the fault event level, and improve patrol maintenance efficiency.

[0004] Although inspection tasks are assigned through the system and inspection results are uploaded through the system, equipment status information still relies on inspections and judgments by operation and maintenance personnel, and equipment failures cannot be responded to quickly. The timeliness and accuracy of discovering defects in substation equipment continue to decline, and potential problems of the equipment cannot be detected. The operating life cycle of the equipment is greatly shortened, posing a huge hidden danger to the operation of the power grid. Summary of the Invention

[0005] In order to enable timely discovery and response to faults and potential problems in substation equipment, the present invention provides a substation equipment operation and inspection device and a health analysis method.

[0006] In a first aspect, the present invention provides a substation equipment operation and health analysis method, which adopts the following technical solutions: A method for analyzing the health of substation equipment operation and inspection, comprising the following steps: S1, configure inspection tasks according to requirements and return inspection data of inspection targets in real time, including image and video data, temperature data and partial discharge data; S2, based on the inspection data, performing abnormality judgment on the image and video data, temperature data, and partial discharge data of the inspection target, and obtaining an abnormality rating of the image and video data, an abnormality rating of the temperature data, and an abnormality rating of the partial discharge data; S3, comprehensively judge all abnormal ratings to obtain the health status of the inspection target; S4, when the health status is assessed as sub-healthy, the operating status trend of the inspection target is predicted based on the neural network model, the predicted operating status trend of the inspection target is compared and analyzed with the operating status trend in the healthy state, and the analysis results are actively warned on the monitoring terminal; When the health status is assessed as abnormal, the fault status of the inspection target is determined based on the abnormality rating, and the monitoring terminal issues a fault alarm.

[0007] By adopting the above technical solution, the health status of the equipment can be obtained through real-time judgment of the inspection data. When the health status of the equipment is sub-healthy, the movement status trend of the equipment is analyzed, and through active early warning, the staff can track the status and discover faults in time. When the health status of the equipment is abnormal, the fault analysis and repair of the equipment are directly carried out, which greatly improves the accuracy of fault analysis and early warning.

[0008] Optionally, in step S2, the specific steps of performing abnormality judgment on the image and video data, temperature data, and partial discharge data of the inspection target based on the inspection data and obtaining the image and video data abnormality rating, the temperature data abnormality rating, and the partial discharge data abnormality rating include: Determine abnormal signs of the inspection target based on the image and video data combined with the sign threshold, and determine the abnormal sign rating of the inspection target based on the first level threshold; Determine abnormal temperature changes of the inspection target based on the temperature data combined with the temperature threshold, and determine the abnormal temperature change rating of the inspection target based on the second level threshold; Determine the partial discharge characteristics of the inspection target based on the partial discharge data combined with the partial discharge threshold, and determine the partial discharge characteristic rating of the inspection target based on the third-level threshold; The abnormal sign rating, the abnormal temperature change rating and the partial discharge characteristic rating are all set to four levels, among which level one is normal, level two is slightly abnormal, level three is moderately abnormal and level four is severely abnormal.

[0009] By adopting the above technical solution, abnormal characteristics of image and video data, temperature data and partial discharge data are judged, and the abnormal characteristics are rated to obtain the specific abnormal characteristics of the device based on each data.

[0010] Optionally, in step S3, the specific steps of comprehensively judging all abnormal ratings and obtaining the health status of the inspection target include: When all abnormality ratings are judged to be level one, the health status of the inspection target is assessed as healthy; If at least one of the abnormal sign rating, abnormal temperature change rating, or partial discharge characteristic rating is at level 4, or at least one of the abnormal temperature change rating, or partial discharge characteristic rating is at level 3, the health status of the inspection target is assessed as abnormal. In other abnormal rating cases, the health status of the inspection target is assessed as sub-healthy.

[0011] By adopting the above technical solution and integrating the results of all abnormal conditions, the health status assessment result of the equipment is obtained.

[0012] Optionally, the determination of abnormal temperature change includes: Use infrared thermal imaging equipment to image the inspection target and obtain a thermal image of the inspection target; Analyze abnormal hot spots of inspection targets through temperature distribution of thermal images; The imaging frequency of the infrared thermal imaging device is set according to the inspection target conditions to obtain abnormal temperature changes.

[0013] By adopting the above technical solution, effective abnormal temperature change parameters are obtained through processing, which helps to accurately diagnose and handle equipment faults.

[0014] Optionally, the determination of the partial discharge characteristics includes: Using electromagnetic wave spatial positioning technology to receive partial discharge electromagnetic wave signals leaked into space; After pre-processing the partial discharge electromagnetic wave signal, it undergoes high-speed AD conversion to obtain the partial discharge characteristics of the inspection target.

[0015] By adopting the above technical solution, the effective abnormal partial discharge characteristics are obtained through preliminary processing, which helps to accurately diagnose and handle equipment faults.

[0016] Optionally, the specific steps of establishing the neural network model include: All historical abnormal marking data, historical abnormal temperature change data, and historical partial discharge characteristic data of the inspection target are taken as the historical abnormal data set, and the historical motion state data of the inspection target are taken as the historical state data set; Perform feature extraction on historical anomaly datasets and historical status datasets, and fuse different features at the same time point to obtain training and test sets; The neural network model is trained using the training set to obtain the initial prediction model, and the output of the model is the corresponding inspection target state; Test the initial prediction model using the test set and output the predicted inspection target status; Determine a corresponding loss value based on the predicted inspection target state output by the model and the mean square error loss function, and judge whether the loss value is less than a threshold; If it is less than the threshold, stop training and use the initial prediction model as the corresponding prediction model; If it is not less than the threshold, the parameters corresponding to the initial prediction model are corrected based on the loss value, and the prediction model is updated.

[0017] Optionally, the step of comparing and analyzing the predicted operating status trend of the inspection target with the operating status trend in a healthy state, and actively issuing an early warning based on the analysis result at the monitoring terminal includes: Calculate the Euclidean distance between the predicted data points and the data points in the healthy state at multiple different time points; Determine the changing trend of the Euclidean distance value; If the value of the Euclidean distance becomes larger and larger, the health status of the inspection target is at risk of deterioration, and an early warning window will pop up on the interface of the monitoring system, which displays the inspection target's operating status trend information and health status deterioration risk information.

[0018] Optionally, the step of determining the fault state of the inspection target based on the abnormality rating and causing the monitoring terminal to issue a fault alarm includes: If one of the ratings is rated as level 4, the inspection target is in a dangerous state. The monitoring terminal pops up an alarm window and sounds an alarm. The alarm window displays the abnormal rating and an "urgent maintenance" reminder. If it is determined that there is at least one level three rating and no level one rating in all ratings, the inspection target is in a defective state, and an alarm window pops up on the monitoring terminal and sounds an alarm. The alarm window displays the abnormal rating and a reminder to "determine the maintenance time for maintenance"; Other rating situations are judged as suspicious, and an alarm window pops up on the monitoring terminal. The alarm window displays the abnormal rating and the reminder to "increase the inspection frequency" to continuously track and monitor the inspection targets.

[0019] In a second aspect, a substation equipment operation and inspection device includes a body, which is deployed at a substation end and includes: The data acquisition module is equipped with multiple detection devices for real-time acquisition of inspection data of the inspection target, including image and video data, temperature data, and partial discharge data; The main computer module configures inspection tasks and generates drive instructions according to needs. It is connected to the power supply unit, image acquisition unit, partial discharge detection unit and temperature sensor unit, receives inspection data and analyzes the health status of the inspection target. It is connected to the monitoring terminal and sends active early warning or fault alarm to the monitoring terminal. The device driving module is connected to the main computer unit, receives driving instructions and drives the device to perform inspections.

[0020] By adopting the above technical solution, the operation and maintenance process is carried out through the above-mentioned device, and no manual monitoring by operation and maintenance personnel is required, thus avoiding problems such as basic data conflicts, duplications, and data missing caused by information islands and fragmented maintenance within the substation.

[0021] Optionally, in the main computer module, the steps of receiving inspection data and analyzing the health status of the inspection target include: Based on the inspection data, the abnormality of the image and video data, temperature data and partial discharge data of the inspection target is judged, and the abnormality rating of the image and video data, the abnormality rating of the temperature data and the abnormality rating of the partial discharge data are obtained; Comprehensively judge all abnormal ratings to obtain the health status of the inspection target; The steps of issuing an active early warning or fault alarm to the monitoring terminal include: When the health status is assessed as sub-healthy, the operating status trend of the inspection target is predicted based on the neural network model. The predicted operating status trend of the inspection target is compared and analyzed with the operating status trend in the healthy state, and the analysis results are actively warned at the monitoring terminal; When the health status is assessed as abnormal, the fault status of the inspection target is determined based on the abnormality rating, and the monitoring terminal issues a fault alarm.

[0022] In summary, the present invention has the following beneficial technical effects: The health status of equipment is assessed through analysis and judgment of the image and video data, temperature data, and partial discharge data returned from inspections. Movement status trend analysis is performed on equipment that currently has no obvious faults, significantly improving the timeliness and accuracy of equipment defect detection. The fault status of the equipment is judged in combination with abnormality ratings, and fault alarms are issued, significantly improving the accuracy of fault analysis and early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 The present invention provides a flow chart of a method for analyzing the health of substation equipment.

[0024] Figure 2 2 is an analysis diagram of a fault state of an embodiment of the present invention.

[0025] Figure 3 The present invention is a schematic structural diagram of a substation equipment operation and maintenance device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The following is combined with Figure 1-3 The present invention is described in further detail.

[0027] Example 1 Reference Figure 1 A method for analyzing the health of substation equipment in this embodiment includes the following steps: S1, configure inspection tasks according to requirements and return inspection data of inspection targets in real time, including image and video data, temperature data and partial discharge data; S2, based on the inspection data, performing abnormality judgment on the image and video data, temperature data, and partial discharge data of the inspection target, and obtaining an abnormality rating of the image and video data, an abnormality rating of the temperature data, and an abnormality rating of the partial discharge data; S3, comprehensively judge all abnormal ratings to obtain the health status of the inspection target; S4, when the health status is assessed as sub-healthy, the operating status trend of the inspection target is predicted based on the neural network model, the predicted operating status trend of the inspection target is compared and analyzed with the operating status trend in the healthy state, and the analysis results are actively warned on the monitoring terminal; When the health status is assessed as abnormal, the fault status of the inspection target is determined based on the abnormality rating, and the monitoring terminal issues a fault alarm.

[0028] In step S2, the specific steps of performing abnormality judgment on the image and video data, temperature data, and partial discharge data of the inspection target based on the inspection data and obtaining the abnormality rating of the image and video data, the abnormality rating of the temperature data, and the abnormality rating of the partial discharge data include: Based on the image video data combined with the sign threshold, the abnormal sign of the equipment is judged, and the abnormal sign rating of the equipment is judged based on the first-level threshold; first, the deep learning algorithm is used to perform intelligent image recognition and analysis on the video image data; the automatic recognition of the secondary pressure plate, SF6 pressure, circuit breaker and knife switch position, cabinet indicator light status, oil leakage of oil-filled equipment, meter reading and other equipment status is realized; the monitoring terminal stores the initial data of the equipment, and uses the initial sign data of the equipment as the sign threshold to determine whether there is an abnormal sign, where the abnormal signs include: blurred dial, damaged dial, damaged shell, insulation Cracks on the sub-devices, ruptured insulators, oil stains on the surface of components, oil stains on the ground, metal rust, broken silicone barrels, abnormal closure of the box door, hanging suspended objects, bird nests, damage to doors, windows, walls and floors, damaged cover plates, unlocked frame ladders, dirty surfaces, abnormal separation and connection of pressure plates, abnormal meter readings, abnormal oil level of respirator seals, discoloration of silicone, etc.; among all the abnormal signs, some have little impact on the equipment, while some have a more serious impact on the equipment. Different first-level thresholds are configured for different initial signs. Therefore, after identifying the abnormal signs, the abnormal sign ratings are determined for different abnormal signs according to the first-level thresholds.

[0029] The abnormal temperature change of the equipment is judged based on the temperature data combined with the temperature threshold, and the abnormal temperature change rating of the equipment is judged based on the second-level threshold; specifically, the monitoring terminal stores the temperature of the equipment during normal operation, and sets key equipment in the substation as detection equipment, such as the moving and static contacts of the switch cabinet, knife switches, cable joints, low-voltage cabinet contacts, high-voltage busbar joints, dry-type transformers and other electrical devices; infrared thermal imaging is used to convert the infrared energy emitted by the equipment into a thermal image; the abnormal hot spots of the insulator are analyzed through the temperature distribution of the thermal image; the frequency of infrared thermal imaging is set according to the equipment condition to obtain abnormal temperature change parameters; the frequency setting range is 1s-1h. The more serious the abnormal condition of the equipment, the higher the frequency should be set for data acquisition and analysis. At the same time, the second-level threshold is set according to the normal operating temperature, and the abnormal temperature change is rated according to the abnormal heating condition of the equipment and the second-level threshold.

[0030] The system uses PD data combined with PD thresholds to determine the device's PD characteristics, and then uses the third-level threshold to determine the device's PD characteristic rating. This system utilizes electromagnetic wave spatial positioning technology, using an omnidirectional UHF sensing antenna to detect electromagnetic wave signals generated by PD excitation. This system can receive PD electromagnetic wave signals leaking into space over a wide range, effectively detecting and locating faults at a distance of 5-20 meters and with a detection bandwidth of 500MHz to 1500MHz. The PD electromagnetic wave signals are pre-processed by filtering out interference, amplifying them appropriately with a programmable amplifier, and then detecting the signal envelope and intermediate frequency signal using a detector and mixer. High-speed analog-to-digital conversion is used to detect weak PD signals, and PD characteristics are derived from these signals. These characteristics include discharge volume, number of discharges, and discharge phase, making PD characteristic monitoring more accurate. Under normal circumstances, PD should not occur in equipment, and due to interference, the acquired PD electromagnetic waves may not accurately reflect the device's condition. Therefore, a third-level threshold is set based on the level of interference. The processed PD characteristics, combined with the third-level threshold, are used to determine the device's PD characteristic rating.

[0031] The abnormal sign rating, the abnormal temperature change rating and the partial discharge characteristic rating are all set to four levels, among which level one is normal, level two is slightly abnormal, level three is moderately abnormal and level four is severely abnormal.

[0032] In step S3, all abnormal ratings are comprehensively judged to obtain the health status of the inspection target. The specific steps include: If all ratings are determined to be level one, no anomalies exist and the device's health status is assessed as healthy. If at least one of the abnormality mark rating, abnormal temperature change rating, or partial discharge characteristic rating is at level 4, or at least one of the abnormality mark rating, abnormal temperature change rating, or partial discharge characteristic rating is at level 3, the health status of the equipment is assessed as abnormal. In other rating cases, the health status of the equipment is assessed as sub-healthy.

[0033] In step S4, the specific steps of establishing the neural network model include: All historical abnormal marking data, historical abnormal temperature change data, and historical partial discharge characteristic data of the inspection target are taken as the historical abnormal data set, and the historical motion state data of the inspection target are taken as the historical state data set; Perform feature extraction on historical anomaly datasets and historical status datasets, and fuse different features at the same time point to obtain training and test sets; The neural network model is trained using the training set to obtain the initial prediction model, and the output of the model is the corresponding inspection target state; Test the initial prediction model using the test set and output the predicted inspection target status; Determine a corresponding loss value based on the predicted inspection target state output by the model and the mean square error loss function, and judge whether the loss value is less than a threshold; If it is less than the threshold, stop training and use the initial prediction model as the corresponding prediction model; If it is not less than the threshold, the parameters corresponding to the initial prediction model are corrected based on the loss value, and the prediction model is updated.

[0034] In step S4, the steps of comparing and analyzing the predicted operating status trend of the inspection target with the operating status trend in the healthy state and actively issuing an early warning based on the analysis results at the monitoring terminal include: Calculate the Euclidean distance between the predicted data points and the data points in the healthy state at multiple different time points; Determine the changing trend of the Euclidean distance value; If the Euclidean distance increases, the health of the inspection target is at risk of deterioration, and an alert window will pop up on the monitoring system interface, displaying the inspection target's operating status trend and health deterioration risk information. If the Euclidean distance decreases, it indicates that the inspection target's health is returning to normal. If the Euclidean distance value remains unchanged, it indicates that the inspection target is currently stable in a sub-healthy state. If the change trend is more complex, such as decreasing and then increasing, specific regional analysis can be performed.

[0035] Reference Figure 2 In step S4, the steps of determining the fault status of the inspection target based on the abnormality rating and the monitoring terminal issuing a fault alarm include: When it is determined that there is a level four among all the ratings, the inspection target is in a dangerous state, and the monitoring terminal pops up an alarm window and sounds an alarm. The alarm window displays the abnormal rating and an "urgent maintenance" reminder; this indicates that the fault is already very serious and staff need to perform maintenance in a timely manner according to the emergency maintenance alarm.

[0036] When it is determined that at least one of all ratings is level three and no level one exists, the inspection target is in a defective state, and an alarm window pops up on the monitoring terminal and sounds an alarm. The alarm window displays the abnormal rating and a reminder to "determine the maintenance time for maintenance." In this embodiment, the staff can analyze the abnormal signs, abnormal temperature changes, and partial discharge characteristics based on historical data to determine the time when the fault first appeared, the time when the fault began to develop rapidly, and whether the fault developed gradually, and determine the maintenance type based on the abnormal rating, which includes: Determine whether the consequences of the fault are serious based on the abnormality rating of the inspection target; when the abnormality rating of the inspection target of the target contains two or more level threes, it is determined that the consequences of the fault are serious; when the consequences of the fault are determined to be serious, analyze whether the fault is gradually developing based on the abnormal signs, abnormal temperature changes and partial discharge characteristics to determine whether the fault is a gradual type; when it is determined that the fault is not a gradual type or the fault is a gradual type but has large dispersion, determine whether the fault is a time-delayed type based on whether there is a delay between the first appearance time of the abnormal signs, abnormal temperature changes and partial discharge characteristics and the time when the fault begins to develop rapidly; when it is determined that the fault is a time-delayed type but the state is unmeasurable, determine whether the inspection target can be energized based on the abnormality rating of the inspection target; when all abnormality ratings are level three, determine that the inspection target cannot be energized; when it is determined that the fault is not a time-delayed type, the fault is a sudden fault, and determine whether the inspection target can be energized based on the abnormality rating of the inspection target; when all abnormality ratings are level three, determine that the inspection target cannot be energized; When the consequences of a fault are judged to be minor and the frequency of occurrence is high, the inspection target is redesigned and optimized, with improvements such as repairs and component replacement performed on the inspection target. When the consequences of a fault are judged to be minor and the frequency of occurrence is low, or the inspection target cannot be powered on, fault maintenance is performed. Existing faults are repaired to restore normal equipment operation. When the fault is judged to be gradual and the dispersion is low, regular maintenance is performed. Due to the gradual nature of the fault and the current minor consequences, equipment maintenance can be performed at regular intervals, determined primarily by the gradual nature of the fault. When the fault is judged to be time-delayed and its condition is measurable, condition-based maintenance is performed. By analyzing and evaluating the status of the inspection target, it is determined whether the inspection target requires maintenance, as well as the content and timing of the maintenance. This maintenance method is more scientific and accurate, can avoid over- or under-maintenance, and improve equipment reliability and cost-effectiveness. When the inspection target is judged to be powered on, live maintenance is performed. Carrying out maintenance on inspection targets while they are operating normally and energized can avoid the impact of equipment power outages on grid operation and improve the power supply reliability of the grid.

[0037] Other rating situations are judged as suspicious, and an alarm window pops up on the monitoring terminal. The alarm window displays the abnormal rating and the reminder to "increase the inspection frequency" to continuously track and monitor the inspection targets.

[0038] Example 2 The difference between this embodiment and embodiment 1 is that this embodiment provides a substation equipment operation and inspection device, including a body, which is deployed at the substation end, referring to Figure 3 , the body includes: The data acquisition module is equipped with multiple detection devices for real-time collection of inspection data of the inspection target, including image and video data, temperature data, and partial discharge data. Specifically, it includes: an image acquisition unit equipped with image and video acquisition devices such as a camera for collecting image and video data; a partial discharge detection unit equipped with a partial discharge sensor for collecting partial discharge data; and a temperature sensing unit equipped with a temperature sensor for collecting temperature data of the device. The main computer module configures inspection tasks and generates drive instructions based on demand. It is signal-connected to the power supply unit, image acquisition unit, partial discharge detection unit, and temperature sensor unit. It receives inspection data and analyzes the health status of the inspection target. It is connected to the monitoring terminal and issues active early warnings or fault alarms to the monitoring terminal. Specifically, it includes a communication unit. Through data transmission and communication between the communication unit and the remote monitoring center or other intelligent devices, it can upload real-time data, fault alarms and other information to the remote monitoring center, and receive instructions or parameter settings from the monitoring center, which helps to realize remote monitoring and intelligent operation and maintenance of the substation. The device drive module is connected to the main computer unit, receives drive instructions and drives the device for inspection, and specifically includes a power supply unit for providing power to the device; a drive unit, which receives drive instructions and drives the device for inspection; the drive unit adopts crawler movement and is composed of drive wheels, guide wheels, traction wheels, track shoes and track frames. The crawler movement structure has a large ground support area, low grounding pressure, and low rolling friction. It is similar to a tank in appearance and has good off-road performance. It can easily cross the raised cable trench and reach every corner of the substation to realize smart inspection tasks.

[0039] The LED display module is the device's human-machine interface, used to display real-time data such as temperature, images, partial discharge detection, and equipment operating status information. Operation and maintenance personnel can intuitively understand the situation within the substation and perform corresponding operations or decisions as needed; The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from the other embodiments. References to the same or similar parts between the various embodiments are sufficient. The methods disclosed in the embodiments are described briefly because they correspond to the systems disclosed in the embodiments. For relevant details, refer to the method description.

[0040] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0041] The above disclosure is only a preferred embodiment of the present invention, but the present invention is not limited thereto. Any non-creative changes that can be thought of by those skilled in the art, as well as several improvements and modifications made without departing from the principles of the present invention, should fall within the scope of protection of the present invention.

Claims

1. A method for analyzing the health of substation equipment operation and maintenance, characterized in that: The following steps are involved: S1, configure inspection tasks according to requirements and return inspection data of inspection targets in real time, including image and video data, temperature data and partial discharge data; S2, based on the inspection data, performing abnormality judgment on the image and video data, temperature data, and partial discharge data of the inspection target, and obtaining an abnormality rating of the image and video data, an abnormality rating of the temperature data, and an abnormality rating of the partial discharge data; S3, comprehensively judge all abnormal ratings to obtain the health status of the inspection target; S4, when the health status is assessed as sub-healthy, the operating status trend of the inspection target is predicted based on the neural network model, the predicted operating status trend of the inspection target is compared and analyzed with the operating status trend in the healthy state, and the analysis results are actively warned on the monitoring terminal; When the health status is assessed as abnormal, the fault status of the inspection target is determined based on the abnormality rating, and the monitoring terminal issues a fault alarm.

2. The substation equipment operation and health analysis method according to claim 1, characterized in that: In step S2, the specific steps of performing abnormality judgment on the image and video data, temperature data, and partial discharge data of the inspection target based on the inspection data and obtaining the abnormality rating of the image and video data, the abnormality rating of the temperature data, and the abnormality rating of the partial discharge data include: Determine abnormal signs of the inspection target based on the image and video data combined with the sign threshold, and determine the abnormal sign rating of the inspection target based on the first level threshold; Determine abnormal temperature changes of the inspection target based on the temperature data combined with the temperature threshold, and determine the abnormal temperature change rating of the inspection target based on the second level threshold; Determine the partial discharge characteristics of the inspection target based on the partial discharge data combined with the partial discharge threshold, and determine the partial discharge characteristic rating of the inspection target based on the third-level threshold; The abnormal sign rating, the abnormal temperature change rating and the partial discharge characteristic rating are all set to four levels, among which level one is normal, level two is slightly abnormal, level three is moderately abnormal and level four is severely abnormal.

3. The substation equipment operation and health analysis method according to claim 2, characterized in that: In step S3, all abnormal ratings are comprehensively judged to obtain the health status of the inspection target. The specific steps include: When all abnormality ratings are judged to be level one, the health status of the inspection target is assessed as healthy; If at least one of the abnormal sign rating, abnormal temperature change rating, or partial discharge characteristic rating is at level 4, or at least one of the abnormal temperature change rating, or partial discharge characteristic rating is at level 3, the health status of the inspection target is assessed as abnormal. In other abnormal rating cases, the health status of the inspection target is assessed as sub-healthy.

4. The substation equipment operation and health analysis method according to claim 2, characterized in that: The determination of abnormal temperature change includes: Use infrared thermal imaging equipment to image the inspection target and obtain a thermal image of the inspection target; Analyze abnormal hot spots of inspection targets through temperature distribution of thermal images; The imaging frequency of the infrared thermal imaging device is set according to the inspection target conditions to obtain abnormal temperature changes.

5. The substation equipment operation and health analysis method according to claim 2, characterized in that: The determination of the partial discharge characteristics includes: Using electromagnetic wave spatial positioning technology to receive partial discharge electromagnetic wave signals leaked into space; After pre-processing the partial discharge electromagnetic wave signal, it undergoes high-speed AD conversion to obtain the partial discharge characteristics of the inspection target.

6. The substation equipment operation and health analysis method according to claim 1, characterized in that: The specific steps to build a neural network model include: All historical abnormal marking data, historical abnormal temperature change data, and historical partial discharge characteristic data of the inspection target are taken as the historical abnormal data set, and the historical motion state data of the inspection target are taken as the historical state data set; Perform feature extraction on historical anomaly datasets and historical status datasets, and fuse different features at the same time point to obtain training and test sets; The neural network model is trained using the training set to obtain the initial prediction model, and the output of the model is the corresponding inspection target state; Test the initial prediction model using the test set and output the predicted inspection target status; Determine a corresponding loss value based on the predicted inspection target state output by the model and the mean square error loss function, and judge whether the loss value is less than a threshold; If it is less than the threshold, stop training and use the initial prediction model as the corresponding prediction model; If it is not less than the threshold, the parameters corresponding to the initial prediction model are corrected based on the loss value, and the prediction model is updated.

7. The substation equipment operation and health analysis method according to claim 6, characterized in that: The steps of comparing and analyzing the predicted operating status trend of the inspection target with the operating status trend in the healthy state and actively issuing an early warning based on the analysis results at the monitoring terminal include: Calculate the Euclidean distance between the predicted data points and the data points in the healthy state at multiple different time points; Determine the changing trend of the Euclidean distance value; If the value of the Euclidean distance becomes larger and larger, the health status of the inspection target is at risk of deterioration, and an early warning window will pop up on the interface of the monitoring system, which displays the inspection target's operating status trend information and health status deterioration risk information.

8. The substation equipment operation and health analysis method according to claim 3, characterized in that: In step S4, the steps of determining the fault status of the inspection target based on the abnormality rating and the monitoring terminal issuing a fault alarm include: If one of the ratings is rated as level 4, the inspection target is in a dangerous state. An alarm window will pop up on the monitoring terminal and an alarm will sound. The alarm window will display the abnormal rating and an "urgent maintenance" reminder. If all ratings are determined to contain at least one level three rating and no level one rating, the inspection target is in a defective state. An alarm window will pop up on the monitoring terminal and sound an alarm. The alarm window will display the abnormal rating and a reminder to "determine the maintenance time for maintenance." Other rating situations are judged as suspicious, and an alarm window pops up on the monitoring terminal. The alarm window displays the abnormal rating and the reminder to "increase inspection frequency" to continuously track and monitor the inspection targets.

9. A substation equipment operation and inspection device, characterized in that: The machine body is deployed at the substation end and includes: The data acquisition module is equipped with multiple detection devices for real-time acquisition of inspection data of the inspection target, including image and video data, temperature data, and partial discharge data; The main computer module configures inspection tasks and generates drive instructions according to needs. It is connected to the power supply unit, image acquisition unit, partial discharge detection unit and temperature sensor unit, receives inspection data and analyzes the health status of the inspection target. It is connected to the monitoring terminal and sends active early warning or fault alarm to the monitoring terminal. The device driving module is connected to the main computer unit, receives driving instructions and drives the device to perform inspections.

10. The substation equipment operation and inspection device according to claim 9, characterized in that: In the main computer module, the steps of receiving inspection data and analyzing the health status of the inspection target include: Based on the inspection data, the abnormality of the image and video data, temperature data and partial discharge data of the inspection target is judged, and the abnormality rating of the image and video data, the abnormality rating of the temperature data and the abnormality rating of the partial discharge data are obtained; Comprehensively judge all abnormal ratings to obtain the health status of the inspection target; The steps of issuing an active early warning or fault alarm to the monitoring terminal include: When the health status is assessed as sub-healthy, the operating status trend of the inspection target is predicted based on the neural network model. The predicted operating status trend of the inspection target is compared and analyzed with the operating status trend in the healthy state, and the analysis results are actively warned at the monitoring terminal; When the health status is assessed as abnormal, the fault status of the inspection target is determined based on the abnormality rating, and the monitoring terminal issues a fault alarm.

Images