Electric power system intelligent operation and maintenance method based on artificial intelligence

Abstract

The invention discloses a power system intelligent operation and maintenance method based on artificial intelligence, relates to the technical field of power system operation and maintenance, and solves the technical problems that when service equipment of a power system is abnormal, monitoring of the power system is not in place, and failure report may occur. The method comprises the following steps: acquiring a network structure topological graph of a power system, and setting a plurality of service equipment detection points; collecting real-time operation data of a plurality of service equipment detection points; when the operation of the plurality of service equipment detection points is abnormal; if so, generating an abnormal alarm signal; otherwise, analyzing the real-time operation data; detecting the detection point of the service equipment, determining an abnormal reason, and maintaining the detection point of the service equipment; calling real-time operation data of a plurality of service equipment detection points in a set time period, and calling an anomaly prediction model to predict anomaly of the plurality of service equipment detection points; according to the prediction result, selecting a time period to check and maintain the power system service equipment; and normal operation of the service equipment is ensured.

Description

Technical Field

[0001] The present invention belongs to the field of power system operation and maintenance, and relates to an intelligent power system operation and maintenance technology based on artificial intelligence, and specifically is an intelligent power system operation and maintenance method based on artificial intelligence. Background Art

[0002] Intelligent operation and maintenance of power systems is an inevitable trend in the development of the modern power industry. It integrates advanced information technology, artificial intelligence technology and big data analysis technology to conduct comprehensive monitoring, dispatching, fault diagnosis and predictive analysis of power systems to improve the reliability, safety and efficiency of the system. Intelligent operation and maintenance uses high-precision sensors and monitoring equipment to monitor the operating status and equipment health of the power system in real time, collect and analyze large amounts of data, and thus detect anomalies in a timely manner. Based on real-time data analysis, intelligent operation and maintenance can formulate more accurate maintenance plans, reduce ineffective operations, avoid excessive or insufficient maintenance, and further improve operation and maintenance efficiency. Using big data analysis and machine learning algorithms, intelligent operation and maintenance can predict possible failures of equipment in the power system, take preventive measures in advance, and avoid accidents. It can reduce downtime caused by failures and improve system reliability and stability.

[0003] The prior art (invention patent with application number 2017110931408) discloses an intelligent on-site operation and maintenance system for an electric power system, which includes hierarchical classification and organization of data, normalization and feature extraction of status detection data, establishment of specifications between instruments and intelligent operation and maintenance boxes, establishment of communication specifications between intelligent operation and maintenance boxes and the center side, wireless real-time transmission of video data streams, high-rate video compression, establishment of an efficient multi-level comprehensive index model for status detection data, and personalized display of analysis and diagnosis technology for status detection data. It can realize standardization of on-site operations and data normalization, realize the sharing of data and analysis algorithms between the on-site side and the center side, improve the data analysis capabilities of the on-site side, and strengthen the remote control and real-time technical support capabilities of on-site operation and maintenance; however, the prior art strengthens on-site operation and maintenance through real-time analysis of the power system on site. When the service equipment of the power system is abnormal, the monitoring of the power system will not be in place, and faults may be missed.

[0004] The present invention provides an intelligent operation and maintenance method for a power system based on artificial intelligence to solve the above technical problems. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes an intelligent operation and maintenance method for an electric power system based on artificial intelligence, which is used to solve the technical problem in the prior art that when the service equipment of the electric power system is abnormal, the monitoring of the electric power system may not be in place, and faults may be missed.

[0006] To achieve the above objectives, the first aspect of the present invention provides an intelligent operation and maintenance method for a power system based on artificial intelligence, comprising:

[0007] Step S1: obtaining a network structure topology diagram of the power system, and setting a number of service equipment detection points according to the network structure topology diagram; collecting real-time operation data of the number of service equipment detection points through data sensors; wherein the real-time operation data includes real-time voltage, real-time current, real-time temperature and real-time vibration;

[0008] Step S2: judging whether the operation of several service equipment detection points is normal based on the real-time operation data; if yes, continuously analyzing the real-time operation data; if no, generating an abnormal alarm signal;

[0009] Step S3: Detect the service equipment detection point according to the abnormal alarm signal to determine the abnormal cause; repair the service equipment detection point according to the abnormal cause;

[0010] Step S4: Retrieve real-time operation data of several service equipment detection points within a set time period, and call an abnormality prediction model to predict abnormalities of several service equipment detection points;

[0011] Step S5: Select a time period to inspect and maintain the power system service equipment according to the prediction results.

[0012] Preferably, the step of setting a plurality of service equipment detection points according to the network structure topology diagram includes:

[0013] Retrieving a network structure topology diagram of the power system, selecting a number of nodes from the network structure topology diagram; marking the number of nodes as service equipment detection points;

[0014] The service equipment detection points are numbered i according to their positional relationships; and the historical data of several service equipment detection points are stored in the power database according to the numbers; wherein i=1, 2, ..., n, and n is a positive integer.

[0015] The present invention selects a number of nodes according to a network structure topology diagram of the power system, marks the several nodes as service equipment detection points and numbers them, which can facilitate the management of the service equipment detection points and facilitate the subsequent positioning of the service equipment detection points.

[0016] Preferably, judging whether the operation of several service equipment detection points is normal based on the real-time operation data includes:

[0017] Retrieve real-time operating data of several service equipment detection points;

[0018] Determine whether the real-time operation data of several service equipment detection points are all within the standard range; if yes, determine that the operation of several service equipment detection points is normal; if not, determine that the operation of the corresponding service equipment detection points is abnormal.

[0019] It should be noted that the standard range is determined based on historical data, and different data types of real-time operation data correspond to different standard ranges. During the comparison process, the data of each data type of the real-time operation data needs to be compared with the corresponding standard range.

[0020] The present invention analyzes the real-time operation data based on the real-time operation data of several power nodes. When the real-time operation data is not within the standard range, it determines that the operation of the corresponding service equipment detection point is abnormal. It can analyze several service equipment detection points in real time and handle the faults of the power system service equipment in time, which is conducive to ensuring the normal operation of the power system service equipment.

[0021] Preferably, the detecting of the service equipment detection point according to the abnormal alarm signal to determine the abnormal cause includes:

[0022] Retrieve the abnormal alarm signal, determine the abnormal service equipment detection point according to the abnormal alarm signal, and mark the corresponding service equipment detection point as an abnormal point;

[0023] Retrieve the real-time operation data of the abnormal point and obtain the data abnormality library; match the real-time operation data with the data abnormality library to obtain the corresponding abnormal cause.

[0024] The present invention marks abnormal service equipment detection points as abnormal points, obtains real-time operation data of the abnormal points, matches the real-time operation data with a data abnormality library to obtain the cause of the abnormality, and can analyze the abnormality based on the real-time operation data, which is conducive to timely processing the abnormality according to the cause of the abnormality.

[0025] Preferably, the repairing of the service equipment detection point according to the abnormal cause includes:

[0026] According to the abnormal cause, check the power lines connected to the service equipment detection point to determine the specific location of the abnormal node; formulate abnormal maintenance measures based on the abnormal cause and the specific location of the abnormal node;

[0027] Before repairing the abnormal node, stop the operation of the power system in the corresponding area; repair the abnormal node according to the abnormal repair measures; after the repair is completed, detect the abnormal node and restore the operation of the power system in the corresponding area according to the detection results.

[0028] The present invention investigates abnormal nodes according to the causes of the abnormalities, and repairs the abnormal nodes according to the specific locations and causes of the abnormal nodes, and can formulate the most appropriate maintenance plan according to the actual abnormal conditions; it is beneficial to ensure the normal operation of the service equipment of the power system and guarantee the real-time monitoring of the power system.

[0029] Preferably, the step of restoring the operation of the power system in the corresponding area according to the detection result includes:

[0030] Retrieve the test results after maintenance;

[0031] Determine whether the data of the test results are all within the standard range; if yes, restore the operation of the power system in the corresponding area; if not, re-inspect and repair the abnormal points.

[0032] The present invention detects the abnormal nodes after maintenance, and analyzes whether the abnormal nodes after maintenance can operate normally according to the detection results. When it can operate normally, the power supply of the power system is restored. Otherwise, the abnormal nodes are continuously inspected and repaired for abnormalities. It can ensure that the power supply is restored only after the service equipment of the power system has returned to normal, which is conducive to ensuring effective monitoring of the power system.

[0033] Preferably, the calling of the anomaly prediction model to predict anomalies of several service equipment detection points includes:

[0034] Retrieve the real-time operation data of several service equipment detection points within a set time period, combine the real-time operation data of the same service equipment detection point into an operation array; integrate several operation arrays to obtain an abnormal prediction sequence;

[0035] Call the anomaly prediction model and input the anomaly prediction sequence into the anomaly prediction model; obtain the anomaly prediction label; match the corresponding anomaly information according to the anomaly prediction label; wherein the anomaly information includes the anomaly detection point, the anomaly time and the anomaly cause; the anomaly prediction model is constructed according to the artificial intelligence model.

[0036] It should be noted that the time period is set manually based on experience. When setting the time period, multiple time periods can be set; the model is used for prediction and verification respectively, and the accuracy of the corresponding prediction results of multiple time periods is analyzed; the time period with the highest accuracy is selected as the final set time period; the abnormal prediction label is set to a positive integer.

[0037] The present invention predicts abnormal information of the power system based on the real-time operating data of the service equipment detection points within a set time period, can analyze the abnormal situation based on historical data, and detect the power system before the abnormality occurs, which is beneficial to the accuracy of the prediction and avoids the inability to detect power system failures due to abnormalities in the power system service equipment, thereby affecting the normal power consumption of residents.

[0038] Preferably, the abnormality prediction model is constructed according to an artificial intelligence model, including:

[0039] Acquire standard training data, wherein the standard training data includes standard input data consistent with the content attributes of the abnormal prediction sequence, and standard output data consistent with the content attributes of the abnormal prediction label;

[0040] The artificial intelligence model is trained using standard training data, and the trained artificial intelligence model is marked as an abnormal prediction model; wherein the artificial intelligence model includes a convolutional neural network model or a long short-term memory neural network model.

[0041] It should be noted that standard training data is data based on actual simulation. Using a large amount of standard training data to train the artificial intelligence model can improve the accuracy of the artificial intelligence model; and the trained artificial intelligence model needs to be verified; when the verification indicators all meet the standards, the artificial intelligence model training is completed, otherwise, the artificial intelligence model is retrained.

[0042] Preferably, the step of selecting a time period according to the prediction result to inspect and maintain the power system service equipment includes:

[0043] Retrieve abnormal information; obtain maintenance time based on abnormal nodes and the time required for maintenance determined by abnormal nodes, and divide the time period before the abnormal time into several maintenance times;

[0044] Obtain historical electricity consumption, analyze the electricity consumption corresponding to several maintenance times based on the historical electricity consumption, select the maintenance time with the least electricity consumption to inspect and maintain the power system service equipment; and notify the user of the corresponding maintenance time.

[0045] The present invention determines the time required for maintenance based on abnormal information, divides the time period before the abnormal time into several maintenance times, and selects the final maintenance time based on the power consumption of the several maintenance times; it can detect the power system before the abnormality occurs in the power system, and select the time period with the least power consumption for maintenance; it is beneficial to ensure the power demand of most users and reduce the impact of power outages.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. The present invention obtains a network structure topology diagram of an electric power system, selects a number of nodes according to the network structure topology diagram of the electric power system, marks the number of nodes as service equipment detection points and numbers them, which can facilitate the management of the service equipment detection points and facilitate the subsequent positioning of the service equipment detection points; collects real-time operation data of a number of service equipment detection points through data sensors, analyzes the real-time operation data according to the real-time operation data of a number of power nodes, and determines that the operation of the corresponding service equipment detection point is abnormal when the real-time operation data is not within the standard range; can analyze a number of service equipment detection points in real time, and promptly handle the faults of the electric power system, which is beneficial to ensuring the normal operation of the service equipment of the electric power system and improving the monitoring of the electric power system.

[0048] 2. Mark the abnormal service equipment detection point as an abnormal point, obtain the real-time operation data of the abnormal point, match the real-time operation data with the data anomaly library to obtain the cause of the abnormality, and be able to analyze the abnormality based on the real-time operation data, which is conducive to timely handling the abnormality according to the cause of the abnormality; check the abnormal nodes according to the cause of the abnormality, and repair the abnormal nodes according to the specific location and cause of the abnormal node. According to the actual abnormal situation, the most appropriate maintenance plan can be formulated; it is conducive to ensuring the normal operation of the power system service equipment and ensuring real-time monitoring of the power system.

[0049] 3. The present invention detects the abnormal nodes after maintenance, and analyzes whether the abnormal nodes after maintenance can operate normally according to the detection results. When it can operate normally, the power supply of the power system is restored, otherwise, the abnormal nodes are continuously inspected and repaired for abnormalities; it can ensure that the power supply is restored only after the power system is restored to normal, which is beneficial to ensuring the normal operation of the power system; the abnormal information of the power system is predicted according to the real-time operation data of the service equipment detection point within the set time period, the abnormal situation can be analyzed according to the historical data, and the power system is detected before the abnormality occurs, which is beneficial to the accuracy of the prediction and avoids affecting the normal electricity consumption of residents due to the abnormality of the power system service equipment; the time required for maintenance is determined according to the abnormal information, the time period before the abnormal time is divided into several maintenance times, and the final maintenance time is selected according to the power consumption of several maintenance times; before the abnormality occurs in the power system service equipment, the power system service equipment can be detected, and the time period with the least power consumption is selected for maintenance; it is beneficial to guarantee the power demand of most users and reduce the impact of power outages. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 It is a schematic diagram of the specific steps of the present invention;

[0052] Figure 2 A schematic diagram of the steps of analyzing abnormal causes of the present invention;

[0053] Figure 3 A schematic diagram of abnormal maintenance steps of the present invention;

[0054] Figure 4 Schematic diagram of abnormality prediction and processing of the present invention. DETAILED DESCRIPTION

[0055] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] See also Figure 1 The first embodiment of the present invention provides an intelligent operation and maintenance method of a power system based on artificial intelligence, comprising:

[0057] Step S1: obtaining a network structure topology diagram of the power system, and setting a number of service equipment detection points according to the network structure topology diagram; collecting real-time operation data of the number of service equipment detection points through data sensors; wherein the real-time operation data includes real-time voltage, real-time current, real-time temperature and real-time vibration;

[0058] Step S2: judging whether the operation of several service equipment detection points is normal based on the real-time operation data; if yes, continuously analyzing the real-time operation data; if no, generating an abnormal alarm signal;

[0059] Step S3: Detect the service equipment detection point according to the abnormal alarm signal to determine the abnormal cause; repair the service equipment detection point according to the abnormal cause;

[0060] Step S4: Retrieve real-time operation data of several service equipment detection points within a set time period, and call an abnormality prediction model to predict abnormalities of several service equipment detection points;

[0061] Step S5: Select a time period to inspect and maintain the power system service equipment according to the prediction results.

[0062] See also Figure 2 , obtain a network structure topology diagram of the power system, select several nodes from the network structure topology diagram; mark several nodes as service equipment detection points; number the service equipment detection points according to their positional relationships; and store historical data of several service equipment detection points in the power database according to the numbers.

[0063] For example, suppose that an electric power system is analyzed, a network topology diagram of the electric power system is obtained, and a total of 15 nodes are selected from the network topology diagram. These 15 nodes are marked as service equipment detection points and numbered.

[0064] Collect the real-time operation data of several service equipment detection points through data sensors to determine whether the real-time operation data of several service equipment detection points are all within the standard range; if yes, continue to analyze the real-time operation data; if not, generate an abnormal alarm signal; determine the abnormal service equipment detection point according to the abnormal alarm signal, and mark the corresponding service equipment detection point as an abnormal point; retrieve the real-time operation data of the abnormal point and obtain the data abnormality library; match the real-time operation data with the data abnormality library to obtain the corresponding abnormal cause.

[0065] For example, suppose that 15 service equipment detection points are analyzed, and the real-time operation data of each service equipment detection point is collected and analyzed. If the real-time operation data of service equipment detection point 9 and service equipment detection point 13 are not within the standard range, an abnormal alarm signal is generated;

[0066] The real-time operation data of the service equipment detection point 9 and the service equipment detection point 13 are respectively obtained, and the data anomaly library is obtained; the real-time operation data of the service equipment detection point 9 and the service equipment detection point 13 are matched with the data anomaly library, and the anomaly cause of the service equipment detection point 9 is D, and the anomaly cause of the service equipment detection point 13 is B.

[0067] See also Figure 3 , according to the cause of the abnormality, check the power lines connected to the service equipment detection point to determine the specific location of the abnormal node; formulate abnormal maintenance measures based on the cause of the abnormality and the specific location of the abnormal node; before repairing the abnormal node, stop the operation of the power system in the corresponding area; repair the abnormal node according to the abnormal maintenance measures; after the maintenance is completed, detect the abnormal node to determine whether the data of the test results are all within the standard range; if yes, restore the operation of the power system in the corresponding area; if not, re-check and repair the abnormal point.

[0068] For example, suppose that the service device detection point 9 and the service device detection point 13 are respectively subjected to exception processing, and the specific steps are as follows:

[0069] Service equipment detection point 9: According to the abnormal cause D, the power line is checked to determine the specific location of the abnormal node; according to the abnormal cause D and the specific location, abnormal maintenance measures C are formulated; the operation of the power system in the corresponding area is stopped, and the abnormal cause D is repaired using the abnormal maintenance measures C; after the maintenance is completed, the abnormality and the single are tested, and if the test results are within the standard range, the power supply to the corresponding service equipment detection point 9 area is restored;

[0070] Service equipment detection point 13: According to the abnormal cause B, the power line is checked to determine the specific location of the abnormal node; according to the abnormal cause B and the specific location, abnormal maintenance measures R are formulated; the operation of the power system in the corresponding area is stopped, and the abnormal cause B is repaired using the abnormal maintenance measures R; after the maintenance is completed, the abnormality and the single are tested, and it is found that the current in the test result is not within the standard range. The abnormality is re-checked and the cause is found to be A. After the abnormal cause A is repaired and tested again, the data of the test results are all within the standard range, and the power supply to the corresponding service equipment detection point 13 area is restored.

[0071] Please refer to 4, retrieve the real-time operation data of several service equipment detection points within the set time period, combine the real-time operation data of the same service equipment detection point into an operation array; integrate several operation arrays to obtain an abnormal prediction sequence; call the abnormal prediction model, input the abnormal prediction sequence into the abnormal prediction model; obtain the abnormal prediction label; match the corresponding abnormal information according to the abnormal prediction label.

[0072] For example: suppose an abnormality prediction is performed on the service equipment detection point, the real-time operation data within 5 minutes is integrated into an abnormality prediction sequence, and the abnormality prediction model is called to analyze the abnormality prediction sequence, and the abnormality prediction label is 8; the corresponding abnormal detection point is service equipment detection point 5, the abnormal reason is E, and the abnormal time is 3 hours later.

[0073] The maintenance time is obtained based on the abnormal nodes and the time required for maintenance determined by the abnormal nodes, and the time period before the abnormal time is divided into several maintenance times; the historical power consumption is obtained, and the power consumption corresponding to several maintenance times is analyzed based on the historical power consumption, and the maintenance time with the least power consumption is selected to inspect and maintain the power system service equipment; and the corresponding maintenance time is notified to the user.

[0074] For example: assuming that the maintenance time of service equipment detection point 5 is determined, and the predicted time of abnormality is 19:00, the time when the abnormality will occur is 22:00; according to the analysis of abnormality cause E, the time required for maintenance is 20 minutes, so the time period is divided into 8, and the power consumption of several maintenance times is analyzed according to historical data. It is found that the time period with the least power consumption is from 21:40 to 22:00, so this time period is used as the maintenance time period, and the user is notified at 19:00 that a power outage will occur from 21:40 to 22:00, and the power system will be inspected.

[0075] It is worth noting that the anomaly prediction model is built based on the artificial intelligence model, including:

[0076] Acquire standard training data, wherein the standard training data includes standard input data consistent with the content attributes of the abnormal prediction sequence, and standard output data consistent with the content attributes of the abnormal prediction label;

[0077] The artificial intelligence model is trained using standard training data, and the trained artificial intelligence model is marked as an abnormal prediction model; wherein the artificial intelligence model includes a convolutional neural network model or a long short-term memory neural network model.

[0078] The working principle of the present invention is as follows: the present invention obtains a network structure topology diagram of an electric power system, and sets a number of service equipment detection points according to the network structure topology diagram; collects real-time operation data of the number of service equipment detection points through data sensors; determines whether the operation of the number of service equipment detection points is normal based on the real-time operation data; if yes, continuously analyzes the real-time operation data; if not, generates an abnormal alarm signal; detects the service equipment detection points according to the abnormal alarm signal, and determines the cause of the abnormality; repairs the service equipment detection points according to the cause of the abnormality; retrieves the real-time operation data of the number of service equipment detection points within a set time period, and calls an abnormal prediction model to predict the abnormality of the number of service equipment detection points; selects a time period according to the prediction result to inspect and maintain the service equipment of the electric power system.

[0079] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent operation and maintenance method for power system based on artificial intelligence, characterized in that: include: Step S1: obtaining a network structure topology diagram of the power system, and setting a number of service equipment detection points according to the network structure topology diagram; Collecting real-time operation data of several service equipment detection points through data sensors; wherein the real-time operation data includes real-time voltage, real-time current, real-time temperature and real-time vibration; Step S2: judging whether the operation of several service equipment detection points is normal based on the real-time operation data; if yes, continuously analyzing the real-time operation data; if no, generating an abnormal alarm signal; Step S3: Detect the service equipment detection point according to the abnormal alarm signal to determine the abnormal cause; repair the service equipment detection point according to the abnormal cause; Step S4: Retrieve real-time operation data of several service equipment detection points within a set time period, and call an abnormality prediction model to predict abnormalities of several service equipment detection points; Step S5: Select a time period to inspect and maintain the power system service equipment according to the prediction results.

2. According to claim 1, an intelligent operation and maintenance method for a power system based on artificial intelligence is characterized in that: The method of setting a plurality of service equipment detection points according to the network structure topology diagram includes: Retrieving a network structure topology diagram of the power system, selecting a number of nodes from the network structure topology diagram; marking the number of nodes as service equipment detection points; The service equipment detection points are numbered i according to their positional relationships; and the historical data of several service equipment detection points are stored in the power database according to the numbers; wherein i=1, 2, ..., n, and n is a positive integer.

3. The intelligent operation and maintenance method of a power system based on artificial intelligence according to claim 1, characterized in that: The determining whether the operation of several service equipment detection points is normal based on the real-time operation data includes: Retrieve real-time operating data of several service equipment detection points; Determine whether the real-time operation data of several service equipment detection points are all within the standard range; if yes, determine that the operation of several service equipment detection points is normal; if not, determine that the operation of the corresponding service equipment detection points is abnormal.

4. The method for intelligent operation and maintenance of a power system based on artificial intelligence according to claim 3 is characterized in that: The detecting point of the service equipment is detected according to the abnormal alarm signal to determine the abnormal cause, including: Retrieve the abnormal alarm signal, determine the abnormal service equipment detection point according to the abnormal alarm signal, and mark the corresponding service equipment detection point as an abnormal point; Retrieve the real-time operation data of the abnormal point and obtain the data abnormality library; match the real-time operation data with the data abnormality library to obtain the corresponding abnormal cause.

5. The method for intelligent operation and maintenance of a power system based on artificial intelligence according to claim 4 is characterized in that: The maintenance of the service equipment detection point according to the abnormal cause includes: According to the abnormal cause, check the power lines connected to the service equipment detection point to determine the specific location of the abnormal node; formulate abnormal maintenance measures based on the abnormal cause and the specific location of the abnormal node; Before repairing the abnormal node, stop the operation of the power system in the corresponding area; repair the abnormal node according to the abnormal repair measures; after the repair is completed, detect the abnormal node and restore the operation of the power system in the corresponding area according to the detection results.

6. The method for intelligent operation and maintenance of a power system based on artificial intelligence according to claim 5, characterized in that: The restoring the operation of the power system in the corresponding area according to the detection result includes: Retrieve the test results after maintenance; Determine whether the data of the test results are all within the standard range; if yes, restore the operation of the power system in the corresponding area; if not, re-inspect and repair the abnormal points.

7. The method for intelligent operation and maintenance of a power system based on artificial intelligence according to claim 1, characterized in that: The calling of the abnormality prediction model to predict abnormalities of several service equipment detection points includes: Retrieve the real-time operation data of several service equipment detection points within a set time period, combine the real-time operation data of the same service equipment detection point into an operation array; integrate several operation arrays to obtain an abnormal prediction sequence; Call the anomaly prediction model and input the anomaly prediction sequence into the anomaly prediction model; obtain the anomaly prediction label; match the corresponding anomaly information according to the anomaly prediction label; wherein the anomaly information includes the anomaly detection point, the anomaly time and the anomaly cause; the anomaly prediction model is constructed according to the artificial intelligence model.

8. The method for intelligent operation and maintenance of a power system based on artificial intelligence according to claim 7, characterized in that: The abnormal prediction model is constructed according to an artificial intelligence model, including: Acquire standard training data, wherein the standard training data includes standard input data consistent with the content attributes of the abnormal prediction sequence, and standard output data consistent with the content attributes of the abnormal prediction label; The artificial intelligence model is trained using standard training data, and the trained artificial intelligence model is marked as an abnormal prediction model; wherein the artificial intelligence model includes a convolutional neural network model or a long short-term memory neural network model.

9. The method for intelligent operation and maintenance of a power system based on artificial intelligence according to claim 7, characterized in that: The step of selecting a time period according to the prediction result to inspect and maintain the power system service equipment includes: Retrieve abnormal information; obtain maintenance time based on abnormal nodes and the time required for maintenance determined by abnormal nodes, and divide the time period before the abnormal time into several maintenance times; Obtain historical electricity consumption, analyze the electricity consumption corresponding to several maintenance times based on the historical electricity consumption, select the maintenance time with the least electricity consumption to inspect and maintain the power system service equipment; and notify the user of the corresponding maintenance time.

Images