Power distribution network operation monitoring method and system and storage medium

Abstract

The invention discloses a power distribution network operation monitoring method and system and a storage medium, and the method comprises the steps: carrying out the dynamic optimization of an equipment health monitoring process, a power grid load abnormality monitoring process and a power grid operation monitoring process based on a self-learning feedback mechanism; obtaining an equipment health index of each equipment, and a power grid load abnormal index and an operation risk abnormal index of the target power distribution network; inputting a time sequence correlation analysis result of the equipment health index, the power grid load anomaly index and the operation risk anomaly index, the equipment health index, the power grid load anomaly index and the operation risk anomaly index into a pre-trained deep learning model to obtain a risk monitoring result; and controlling a functional component matched with the target power distribution network to execute a risk control strategy corresponding to the risk monitoring result. According to the power distribution network operation monitoring method and system and the storage medium provided by the embodiment of the invention, the monitoring accuracy of the power distribution network is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method, system and storage medium for monitoring the operation of a distribution network. Background Art

[0002] As an important part of the power system, the safe operation of the distribution network ensures the uninterrupted supply of electricity and the safety of user power consumption, and is of crucial significance to the stability, economy of the entire power system and the normal order of social life. By monitoring the distribution network, the operation status of the power grid can be monitored in real time. Once an abnormality or potential fault is found, a warning can be issued immediately, enabling the operation and maintenance personnel to take measures quickly to prevent the expansion of the fault and ensure the safe operation of the power grid.

[0003] However, the existing technology for monitoring the distribution network is a single and static analysis, which cannot cope with the complex operation conditions and real-time changes of the power grid and equipment, resulting in a low accuracy of monitoring the distribution network.

[0004] Therefore, how to improve the accuracy of monitoring the distribution network has become a technical problem that needs to be urgently solved by those skilled in the art. Summary of the Invention

[0005] The present invention provides a method, system and storage medium for monitoring the operation of a distribution network to solve the technical problem that the existing technology for monitoring the distribution network is a single and static analysis, which cannot cope with the complex operation conditions and real-time changes of the power grid and equipment, resulting in a low accuracy of monitoring the distribution network.

[0006] To solve the above technical problem, an embodiment of the present invention provides a method for monitoring the operation of a distribution network.

[0007] Optimize the device health monitoring process based on a self-learning feedback mechanism to obtain the device health index of each device, wherein the device health monitoring process is designed to perform dynamic weighting processing on the obtained device status parameters;

[0008] Optimize the grid load anomaly monitoring process based on a self-learning feedback mechanism to obtain the grid load anomaly index of the target distribution network, wherein the grid load anomaly monitoring process reflects the change analysis results of the grid load rate in different time dimensions;

[0009] Optimize the grid operation monitoring process based on a self-learning feedback mechanism to obtain the operation risk anomaly index of the target distribution network, wherein the grid operation monitoring process reflects the abnormal analysis results of the predicted operation parameter;

[0010] Input the time series correlation analysis results of the device health index, the grid load anomaly index, and the operation risk anomaly index, the device health index, the grid load anomaly index, and the operation risk anomaly index into a pre-trained deep learning model to obtain a risk monitoring result;

[0011] Control the functional components corresponding to the target distribution network to execute the risk control strategy corresponding to the risk monitoring result.

[0012] As one of the preferred solutions, optimizing the device health monitoring process based on the self-learning feedback mechanism to obtain the device health index of each device includes:

[0013] Perform dynamic weighting processing on the collected device status parameters of each device and calculate the device health index;

[0014] The dynamic weighting processing is designed to adjust the weight of each device status parameter in real time based on the deviation between the first real-time data and the corresponding first historical data of each device status parameter, and perform weighted summation on the first real-time data with the updated weight obtained by real-time adjustment as the device health index.

[0015] As one of the preferred solutions, optimizing the grid load anomaly monitoring process based on the self-learning feedback mechanism to obtain the grid load anomaly index of the target distribution network includes:

[0016] Calculate the grid load rate according to the linear relationship between the rated power and the operating power of the collected target distribution network;

[0017] Perform dynamic threshold comparison analysis on the second real-time data of the grid load rate based on the distribution characteristics of the second historical data of the grid load rate, and calculate the grid load anomaly index.

[0018] As one of the preferred solutions, optimizing the grid operation monitoring process based on the self-learning feedback mechanism to obtain the operation risk anomaly index of the target distribution network includes:

[0019] Use a long short-term memory network model to perform predictive analysis on the third historical data of each operation parameter collected, obtain the predicted data of each operation parameter, and perform anomaly detection on all the predicted data to obtain the operation risk anomaly index.

[0020] As one of the preferred solutions, optimizing the grid operation monitoring process based on the self-learning feedback mechanism to obtain the operation risk anomaly index of the target distribution network further includes:

[0021] Feedback-adjust the weight parameters of the long short-term memory network model based on the deviation between the predicted data based on each of the operation parameters of the job and the third real-time data at the corresponding moment.

[0022] As one of the preferred solutions, the real-time adjustment of the weight of each of the device state parameters based on the deviation between the first real-time data based on each of the device state parameters and the corresponding first historical data includes:

[0023] Take the average value of all the first historical data of each of the device state parameters as the state reference value of each of the device state parameters;

[0024] Take the absolute value of the difference between the first real-time data of each of the device state parameters and the corresponding state reference value as the degree of state deviation of each of the device state parameters;

[0025] Take the result obtained by normalizing all the degrees of state deviation as the updated weight of each of the device state parameters.

[0026] As one of the preferred solutions, the dynamic threshold comparison analysis of the second real-time data of the grid load rate based on the distribution characteristics of the second historical data of the grid load rate, and calculate the grid load anomaly index, including:

[0027] Take the sum of the average value and the corresponding standard deviation of all the second historical data as the warning upper limit, and take the absolute value of the difference between the second real-time data and the warning upper limit as the upper limit deviation degree;

[0028] Take the difference between the average value of all the second historical data and the corresponding standard deviation as the warning lower limit, and take the absolute value of the difference between the second real-time data and the warning lower limit as the lower limit deviation degree;

[0029] Take the minimum value of the upper limit deviation degree and the lower limit deviation degree as the grid load anomaly index.

[0030] As one of the preferred solutions, the risk monitoring results include mild risk, moderate risk and severe risk;

[0031] The risk control strategies include a preventive reminder strategy corresponding to the mild risk, a mandatory reminder strategy corresponding to the moderate risk, and a mandatory intervention strategy corresponding to the severe risk.

[0032] Another embodiment of the present invention provides a distribution network operation monitoring system, including:

[0033] The device monitoring module is used to optimize the device health monitoring process based on a self-learning feedback mechanism to obtain the device health index of each device, wherein the device health monitoring process is designed to perform dynamic weighting processing on the obtained device status parameters;

[0034] The power grid monitoring module is used to optimize the power grid load anomaly monitoring process based on a self-learning feedback mechanism to obtain the power grid load anomaly index of the target distribution network, wherein the power grid load anomaly monitoring process reflects the analysis results of the changes in the power grid load rate in different time dimensions;

[0035] The operation monitoring module is used to optimize the power grid operation monitoring process based on a self-learning feedback mechanism to obtain the operation risk anomaly index of the target distribution network, wherein the power grid operation monitoring process reflects the analysis results of the anomalies of the predicted operation parameter;

[0036] The three-risk linkage evaluation module is used to input the time-series correlation analysis results of the device health index, the power grid load anomaly index and the operation risk anomaly index, the device health index, the power grid load anomaly index and the operation risk anomaly index into a pre-trained deep learning model to obtain a risk monitoring result;

[0037] The risk control module is used to control the functional components matching the target distribution network to execute the risk control strategy corresponding to the risk monitoring result.

[0038] Another embodiment of the present invention provides a computer-readable storage medium storing a computer program, wherein when the device where the computer-readable storage medium is located executes the computer program, the above-mentioned method for monitoring the operation of a distribution network is implemented.

[0039] Compared with the prior art, the beneficial effects of the embodiments of the present invention are at least one of the following:

[0040] Considering that the operating conditions of the distribution network change in real time, static analysis is difficult to cope with the complex operating conditions and real-time changes of the power grid. The device health monitoring process is optimized based on a self-learning feedback mechanism to obtain the device health index of each device. Among them, the device health monitoring process is designed to perform dynamic weighting on the obtained status parameters of each device; the grid load anomaly monitoring process is optimized based on the self-learning feedback mechanism to obtain the grid load anomaly index of the target distribution network. Among them, the grid load anomaly monitoring process reflects the analysis results of the changes in the grid load rate under different time dimensions; the grid operation monitoring process is optimized based on the self-learning feedback mechanism to obtain the operation risk anomaly index of the target distribution network. Among them, the grid operation monitoring process reflects the analysis results of the anomalies of the predicted operation parameter. Through the self-learning feedback mechanism, the monitoring of the power grid, devices, and operations is dynamically optimized, improving the accuracy of the distribution network monitoring.

[0041] The time-series correlation analysis results of the device health index, the grid load anomaly index, and the operation risk anomaly index, the device health index, the grid load anomaly index, and the operation risk anomaly index are input into a pre-trained deep learning model to obtain the risk monitoring results, realizing the dynamic fusion and comprehensive analysis of the power grid operation, device status, and operation. The functional components corresponding to the target distribution network are controlled to execute the risk control strategy corresponding to the risk monitoring results. Through the dynamic monitoring of the power grid, devices, and operations and the correlation analysis among the three, the adaptive adjustment of the risk monitoring results is realized, enhancing the response ability of the system, avoiding the risk of operation errors caused by only monitoring a single factor, ensuring the efficient and safe operation of the system, and further improving the accuracy of the distribution network monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flowchart of a distribution network operation monitoring method in one embodiment of the present invention;

[0043] Figure 2 It is a structural block diagram of a distribution network operation monitoring system in one embodiment of the present invention;

[0044] Figure 3 It is a system architecture diagram of a distribution network operation monitoring system in one embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0045] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0046] In the description of the present application, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first", "second", "third", etc. may explicitly or implicitly include one or more of such features. In the description of the present application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0047] In the description of the present application, it should be noted that unless otherwise clearly defined and limited, the terms "installed", "connected", "connected" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two components. The terms "vertical", "horizontal", "left", "right", "up", "down" and similar expressions used herein are only for the purpose of illustration and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be understood as a limitation of the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0048] In the description of the present application, it should be noted that unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those of ordinary skill in the technical field to which this technology belongs. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0049] An embodiment of the present invention provides a method for monitoring the operation of a distribution network. Specifically, please refer to Figures 1 to 3 , Figure 1 which shows a flowchart of a method for monitoring the operation of a distribution network in one of the embodiments of the present invention, Figure 2 which shows a structural block diagram of a system for monitoring the operation of a distribution network in one of the embodiments of the present invention.Figure 3 It is shown as the system architecture diagram of a distribution network operation monitoring system in one embodiment of the present invention.

[0050] The flowchart of a distribution network operation monitoring method in one embodiment of the present invention includes the following steps S1 to S5, specifically as follows:

[0051] Step S1: Optimize the device health monitoring process based on the self - learning feedback mechanism to obtain the device health index of each device. Among them, the device health monitoring process is designed to perform dynamic weighting processing on the obtained various device status parameters.

[0052] It should be noted that through the widely deployed sensor network, the system continuously collects power grid operation parameters, device status parameters, and operation parameters. After these data are processed, they are transmitted to the central data processing unit through wireless communication or optical fiber for the next analysis link.

[0053] Furthermore, it should be noted that considering that the operation status of the distribution network changes in real time, static analysis is difficult to cope with the complex operation conditions and real - time changes of the power grid. In order to improve the real - time performance of distribution network monitoring, it is necessary to perform dynamic monitoring and analysis on the power grid, equipment, and operations.

[0054] Specifically, perform dynamic weighting processing on the collected various device status parameters, and calculate the device health index; the dynamic weighting processing is designed to adjust the weight of each device status parameter in real time based on the deviation between the first real - time data and the corresponding first historical data of each device status parameter, and perform weighted summation on the first real - time data with the updated weight obtained by real - time adjustment as the device health index.

[0055] In this step, adjusting the weight of each device status parameter in real time based on the deviation between the first real - time data and the corresponding first historical data of each device status parameter includes: taking the average value of all the first historical data of each device status parameter as the status reference value of each device status parameter; taking the absolute value of the difference between the first real - time data of each device status parameter and the corresponding status reference value as the status deviation degree of each device status parameter; taking the result of normalizing all the status deviation degrees as the updated weight of each device status parameter.

[0056] It should be noted that the evaluation of the device health index is based on key parameters such as device temperature, vibration amplitude, and load rate, and is dynamically adjusted through the correlation analysis of real - time data and historical operation modes. By comparing the deviation amplitude of real - time data with the historical mode, when the real - time vibration or temperature deviates from the normal range, the weight of the corresponding parameter is automatically increased, so that the evaluation result can better reflect the current health status of the device.

[0057] Step S2: Optimize the power grid load anomaly monitoring process based on the self-learning feedback mechanism to obtain the power grid load anomaly index of the target distribution network. Among them, the power grid load anomaly monitoring process reflects the analysis results of the changes in the power grid load rate in different time dimensions.

[0058] It should be noted that the power grid operation parameters at least include the operating power of the target distribution network.

[0059] Specifically, according to the linear relationship between the rated power and the collected operating power of the target distribution network, the power grid load rate is calculated; based on the distribution characteristics of the second historical data of the power grid load rate, a dynamic threshold comparison analysis is performed on the second real-time data of the power grid load rate, and the power grid load anomaly index is calculated.

[0060] As an embodiment of the present application, the power grid load rate is the ratio of the operating power of the target distribution network to the rated power.

[0061] In this step, based on the distribution characteristics of the second historical data of the power grid load rate, a dynamic threshold comparison analysis is performed on the second real-time data of the power grid load rate, and the power grid load anomaly index is calculated, including: taking the sum of the average value of all second historical data and the corresponding standard deviation as the warning upper limit, and taking the absolute value of the difference between the second real-time data and the warning upper limit as the upper limit deviation degree; taking the difference between the average value of all second historical data and the corresponding standard deviation as the warning lower limit, and taking the absolute value of the difference between the second real-time data and the warning lower limit as the lower limit deviation degree; taking the minimum value of the upper limit deviation degree and the lower limit deviation degree as the power grid load anomaly index.

[0062] It should be noted that in order to improve the adaptability of the calculation of the power grid load anomaly index, the present application dynamically adjusts the range of the power grid load rate in combination with the fluctuation range of the historical data of the power grid load rate to avoid false alarms or missed alarms. When the historical data fluctuates little, the range will be narrowed to improve sensitivity; when the fluctuation is large, the range will be appropriately widened to reduce false alarms.

[0063] Step S3: Optimize the power grid operation monitoring process based on the self-learning feedback mechanism to obtain the operation risk anomaly index of the target distribution network. Among them, the power grid operation monitoring process reflects the analysis results of the anomalies of the predicted operation parameter.

[0064] Specifically, the long short-term memory network model is used to perform predictive analysis on the third historical data of each collected operation parameter to obtain the predicted data of each operation parameter, and an anomaly detection is performed on all the predicted data to obtain the operation risk anomaly index.

[0065] During the process of predictive analysis, the weight parameters of the long short-term memory network model are feedback-adjusted based on the deviation between the predictive data of each job operation parameter and the third real-time data at the corresponding moment.

[0066] It should be noted that after each job is completed, the real-time data and the predictive results are compared, and the deviation between the actual risk and the predictive risk is analyzed. When a high predictive error is found, the system will automatically adjust the input weights and training parameters of the long short-term memory network model.

[0067] Step S4: Input the time-series correlation analysis results of the equipment health index, the grid load anomaly index, and the job risk anomaly index, the equipment health index, the grid load anomaly index, and the job risk anomaly index into a pre-trained deep learning model to obtain the risk monitoring results.

[0068] It should be noted that only considering a certain factor in the power grid, equipment, or job operation is likely to ignore the influence of other factors on the system, resulting in incomplete risk assessment. Through the linkage assessment, the interaction between the power grid operation state, the equipment health status, and the job operation process can be comprehensively considered to identify potential systematic risks.

[0069] Specifically, input the time-series correlation analysis results of the equipment health index, the grid load anomaly index, and the job risk anomaly index, the equipment health index, the grid load anomaly index, and the job risk anomaly index into a pre-trained deep learning model to obtain the risk monitoring results.

[0070] It should be noted that the risk monitoring results include low risk, medium risk, and high risk.

[0071] Step S5: Control the functional components matching the target distribution network to execute the risk control strategy corresponding to the risk monitoring results.

[0072] Specifically, the risk control strategies include a preventive reminder strategy corresponding to low risk, a mandatory reminder strategy corresponding to medium risk, and a mandatory intervention strategy corresponding to high risk.

[0073] It should be noted that the risks of the target distribution network are divided into multiple levels, namely low, medium, and high risks, according to the risk monitoring results. For different risk levels, the system takes different response measures. In the case of low risk, the system only issues a warning to remind the operator; in the case of medium risk, the system issues a mandatory prompt and recommends that the operator take preventive measures; in the case of high risk, the system automatically executes a forced intervention operation, such as cutting off the faulty line or locking the equipment. When the system identifies a high-level risk, it selects intervention means according to the preset strategy, including power grid load adjustment, equipment shutdown, personnel evacuation, etc. All intervention operations are executed through the automated control system to ensure that risks can be quickly and effectively eliminated in extreme situations and the safety of the power grid is guaranteed.

[0074] The beneficial effects of a distribution network operation monitoring method provided by an embodiment of the present invention compared with the prior art are as follows:

[0075] Considering that the operating conditions of the distribution network change in real time, static analysis is difficult to cope with the complex operating conditions and real-time changes of the power grid. The equipment health monitoring process is optimized based on the self-learning feedback mechanism to obtain the equipment health index of each device. Among them, the equipment health monitoring process is designed to perform dynamic weighting processing on the obtained state parameters of each device; the grid load anomaly monitoring process is optimized based on the self-learning feedback mechanism to obtain the grid load anomaly index of the target distribution network. Among them, the grid load anomaly monitoring process reflects the change analysis results of the grid load rate in different time dimensions; the grid operation monitoring process is optimized based on the self-learning feedback mechanism to obtain the operation risk anomaly index of the target distribution network. Among them, the grid operation monitoring process reflects the anomaly analysis results of the predicted operation parameter. Through the self-learning feedback mechanism, the monitoring of the power grid, equipment, and operation is dynamically optimized, improving the accuracy of the distribution network monitoring.

[0076] The time-series correlation analysis results of the equipment health index, the grid load anomaly index, and the operation risk anomaly index, the equipment health index, the grid load anomaly index, and the operation risk anomaly index are input into a pre-trained deep learning model to obtain the risk monitoring results, realizing the dynamic fusion and comprehensive analysis of power grid operation, equipment status, and operation. The functional components corresponding to the target distribution network are controlled to execute the risk control strategy corresponding to the risk monitoring results. Through the dynamic monitoring of the power grid, equipment, and operation and the correlation analysis among the three, the adaptive adjustment of the risk level is realized, enhancing the response ability of the system, avoiding the risk of operation errors caused by only monitoring a single factor, and further improving the accuracy of the distribution network monitoring.

[0077] Another embodiment of the present invention provides a distribution network operation monitoring system, including:

[0078] The device monitoring module 11 is used to optimize the device health monitoring process based on a self-learning feedback mechanism to obtain the device health index of each device. Among them, the device health monitoring process is designed to perform dynamic weighting processing on the obtained device status parameters;

[0079] The power grid monitoring module 12 is used to optimize the power grid load anomaly monitoring process based on a self-learning feedback mechanism to obtain the power grid load anomaly index of the target distribution network. Among them, the power grid load anomaly monitoring process reflects the change analysis results of the power grid load rate in different time dimensions;

[0080] The operation monitoring module 13 is used to optimize the power grid operation monitoring process based on a self-learning feedback mechanism to obtain the operation risk anomaly index of the target distribution network. Among them, the power grid operation monitoring process reflects the anomaly analysis results of the predicted operation parameter;

[0081] The three-risk linkage evaluation module 14 is used to input the time-series correlation analysis results of the device health index, the power grid load anomaly index, and the operation risk anomaly index, the device health index, the power grid load anomaly index, and the operation risk anomaly index into a pre-trained deep learning model to obtain a risk monitoring result;

[0082] The risk control module 15 is used to control the functional components matching the target distribution network to execute the risk control strategy corresponding to the risk monitoring result.

[0083] It should be noted that a distribution network operation monitoring system provided in this embodiment further includes a central control unit.

[0084] The device monitoring module 11 is mainly responsible for monitoring the operation status of various power devices in the distribution network, including the real-time status of devices such as transformers, switches, and circuit breakers. Sensors will monitor parameters such as the temperature, vibration, leakage current, and insulation resistance of the devices to ensure that the devices operate within a safe range. Once abnormal parameters are detected, such as too high temperature, too large vibration, or reduced insulation resistance, the system will immediately trigger an alarm and record the abnormal data to assist subsequent analysis and decision-making.

[0085] The power grid monitoring module 12 collects the core data of the power grid operation in real time through a distributed sensor network, such as parameters such as voltage, current, power, and frequency. These data can reflect the current operation status of the power grid, especially in the case of load fluctuations, line abnormalities, voltage fluctuations, etc., and can timely feedback the operation risks of the power grid. The data is transmitted to the central control unit through wireless communication or optical fiber communication for further risk analysis.

[0086] The operation monitoring module 13 mainly monitors the operation process of operators, especially high-risk operation steps. In operations such as the maintenance and repair of the distribution network, the operation behavior of personnel is crucial. This module combines video monitoring, sensors, and operation records to monitor the operation actions of operators, environmental conditions, etc. in real time. The system can identify non-standard operation behaviors or high-risk operation states and issue warnings or perform automatic intervention when necessary to ensure operation safety.

[0087] The three-risk linkage assessment module 14 is the core module, responsible for comprehensively analyzing and making a linkage assessment of the data in three aspects: the power grid, equipment, and operation. Through big data analysis and artificial intelligence technology, the system can achieve the fusion analysis of multi-source data and identify potential comprehensive risks. The linkage analysis not only considers the risks of single factors, such as grid overload or equipment failure, but also combines the behavior patterns of operators and the operation environment to form a dynamic and real-time comprehensive risk assessment.

[0088] Based on the risk monitoring results of the three-risk linkage assessment module, the risk control module 15 can judge the risk level of the system in real time and take corresponding warning and intervention measures. When the system detects a relatively high comprehensive risk, it will remind the operator by means of audible and visual alarms, text message notifications, etc., and perform automatic intervention according to the preset strategy. For example, when a serious equipment failure occurs or the personnel operation is improper, the system can automatically disconnect the faulty equipment or stop the operation process to prevent the accident from expanding further.

[0089] Another embodiment of the present invention provides a computer-readable storage medium storing a computer program. When the device where the computer-readable storage medium is located executes the computer program, it implements a method for monitoring the operation of a distribution network as described above.

[0090] The above embodiments only represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be subject to the appended claims.

Claims

1. A distribution network operation monitoring method, characterized in that: The method comprises: Optimizing the equipment health monitoring process based on a self-learning feedback mechanism to obtain an equipment health index for each of the equipment, wherein the equipment health monitoring process is designed to dynamically weight each of the acquired equipment status parameters; Based on the self-learning feedback mechanism, the grid load anomaly monitoring process is optimized to obtain the grid load anomaly index of the target distribution network, wherein the grid load anomaly monitoring process reflects the change analysis results of the grid load rate under different time dimensions; The power grid operation monitoring process is optimized based on a self-learning feedback mechanism to obtain an abnormal operation risk index of the target distribution network, wherein the power grid operation monitoring process reflects an abnormal analysis result of the predicted operation operation parameters; Input the time series correlation analysis results of the equipment health index, the power grid load abnormality index and the operation risk abnormality index, the equipment health index, the power grid load abnormality index and the operation risk abnormality index into a pre-trained deep learning model to obtain a risk monitoring result; Controlling the functional components matching the target power distribution network to execute the risk control strategy corresponding to the risk monitoring result.

2. A distribution network operation monitoring method according to claim 1, characterized in that: The device health monitoring process is optimized based on the self-learning feedback mechanism to obtain the device health index of each device, including: Dynamically weight the collected equipment status parameters to calculate the equipment health index; The dynamic weighted processing is designed to adjust the weight of each device status parameter in real time based on the deviation between the first real-time data of each device status parameter and the corresponding first historical data, and to weighted sum the first real-time data with the updated weight obtained by the real-time adjustment as the device health index.

3. A distribution network operation monitoring method according to claim 1, characterized in that: The process of optimizing the abnormality monitoring process of the power grid load based on the self-learning feedback mechanism to obtain the abnormality index of the power grid load of the target distribution network includes: The grid load rate is calculated based on the linear relationship between the rated power and the collected operating power of the target distribution network; Based on the distribution characteristics of the second historical data of the power grid load rate, a dynamic threshold comparison analysis is performed on the second real-time data of the power grid load rate to calculate the power grid load anomaly index.

4. A distribution network operation monitoring method according to claim 1, characterized in that: The process of optimizing the power grid operation monitoring process based on the self-learning feedback mechanism to obtain the abnormal operation risk index of the target distribution network includes: A long short-term memory network model is used to perform predictive analysis on the collected third historical data of each operation parameter to obtain predicted data of each operation parameter, and abnormality detection is performed on all the predicted data to obtain an abnormal operation risk index.

5. A distribution network operation monitoring method according to claim 4, characterized in that: The process of optimizing the power grid operation monitoring process based on the self-learning feedback mechanism to obtain the abnormal operation risk index of the target distribution network also includes: Feedback adjustment is performed on the weight parameters of the long short-term memory network model based on the deviation between the predicted data of each of the job operation parameters and the third real-time data at the corresponding moment.

6. A distribution network operation monitoring method according to claim 2, characterized in that: The step of adjusting the weight of each of the device state parameters in real time based on the deviation between the first real-time data of each of the device state parameters and the corresponding first historical data includes: taking an average value of all the first historical data of each of the device state parameters as a state reference value of each of the device state parameters; Taking the absolute value of the difference between the first real-time data of each device state parameter and the corresponding state reference value as the state deviation degree of each device state parameter; The result obtained by normalizing all the state deviation degrees is used as the update weight of each device state parameter.

7. A distribution network operation monitoring method according to claim 3, characterized in that: The step of performing a dynamic threshold comparison analysis on the second real-time data of the power grid load rate based on the distribution characteristics of the second historical data of the power grid load rate to calculate the power grid load anomaly index includes: The sum of the average values ​​of all the second historical data and the corresponding standard deviations is used as the warning upper limit, and the absolute value of the difference between the second real-time data and the warning upper limit is used as the upper limit deviation degree; The difference between the average value of all the second historical data and the corresponding standard deviation is used as the warning lower limit, and the absolute value of the difference between the second real-time data and the warning lower limit is used as the lower limit deviation degree; The minimum value of the upper limit deviation degree and the lower limit deviation degree is used as the power grid load abnormality index.

8. A distribution network operation monitoring method according to claim 1, characterized in that: The risk monitoring results include mild risk, moderate risk and severe risk; The risk control strategy includes a preventive reminder strategy corresponding to the mild risk, a compulsory reminder strategy corresponding to the moderate risk, and a compulsory intervention strategy corresponding to the severe risk.

9. A distribution network operation monitoring system, characterized in that: The system comprises: An equipment monitoring module is used to optimize the equipment health monitoring process based on a self-learning feedback mechanism to obtain the equipment health index of each of the equipment, wherein the equipment health monitoring process is designed to dynamically weight each of the acquired equipment status parameters; A power grid monitoring module is used to optimize the power grid load anomaly monitoring process based on a self-learning feedback mechanism to obtain a power grid load anomaly index of a target distribution network, wherein the power grid load anomaly monitoring process reflects the analysis results of changes in power grid load rates under different time dimensions; An operation monitoring module is used to optimize the power grid operation monitoring process based on a self-learning feedback mechanism to obtain an abnormal operation risk index of the target distribution network, wherein the power grid operation monitoring process reflects an abnormal analysis result of the predicted operation operation parameters; A three-risk linkage assessment module is used to input the time series correlation analysis results of the equipment health index, the power grid load abnormality index and the operation risk abnormality index, the equipment health index, the power grid load abnormality index and the operation risk abnormality index into a pre-trained deep learning model to obtain a risk monitoring result; The risk control module is used to control the functional components matching the target distribution network to execute the risk control strategy corresponding to the risk monitoring result.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein when the device where the computer-readable storage medium is located executes the computer program, a distribution network operation monitoring method as described in any one of claims 1 to 8 is implemented.

Images