A multi-level active power distribution network residual current early warning method and device

Abstract

The embodiment of the application discloses a kind of multi-level active distribution network residual current early warning method and device, comprising: establishing distribution system simulation model, and determine the access point and position of distributed photovoltaic system in distribution system;Obtain historical residual current data and pretreatment;After pretreatment, determine the change trend of residual current by time series, determine the residual current mode of different network levels;Residual current change trend and residual current mode are combined to filter associated factors, build prediction model, and baseline prediction is carried out to residual current;Set pre-set threshold;When residual current exceeds pre-set threshold, trigger early warning.Through historical data processing and residual current prediction, residual current in distribution network can be effectively monitored and managed, and an alarm is issued to remind when residual current exceeds pre-set threshold, which can effectively improve the safety, reliability and economy of distribution network, especially suitable for multi-level active distribution network containing distributed photovoltaic system.

Description

TECHNICAL FIELD

[0001] The embodiment of the application relates to the technical field of power systems, in particular to a residual current early warning method and device for a multi-level active power distribution network. BACKGROUND

[0002] In a traditional power distribution network system, with the wide access of distributed energy resources (such as photovoltaic systems), the network structure becomes more complex and dynamic, which poses new challenges to the monitoring, protection and operation of the power distribution network, especially in the management of residual current (also known as leakage current). Due to the uncertainty of the access point and operation of distributed power generation, it is difficult for traditional residual current protection methods to adapt.

[0003] Residual current generally refers to the current flowing in the non-working current path in the power system, which may be caused by insulation failure, equipment damage or improper wiring, etc. High residual current not only causes waste of electric energy, but more importantly, it may cause electrical fire or electric shock hazards. Therefore, monitoring and early warning of residual current are crucial to ensure the safe operation of the power distribution system. SUMMARY

[0004] The embodiment of the application provides a residual current early warning method and device for a multi-level active power distribution network, which can effectively monitor and manage the residual current in the power distribution network by processing historical data and predicting residual current, and issue an alarm when the residual current exceeds the preset threshold, thereby effectively improving the safety, reliability and economy of the power distribution network, and being especially suitable for a multi-level active power distribution network containing a distributed photovoltaic system.

[0005] According to a first aspect of the application, a residual current early warning method for a multi-level active power distribution network is provided, comprising:

[0006] establishing a power distribution system simulation model based on the network structure of the power distribution system, and determining the access point of the distributed photovoltaic system in the power distribution system and the position of the distributed photovoltaic system in the power distribution network;

[0007] obtaining historical residual current data of different network levels in the network structure of the power distribution system, and preprocessing the historical residual current data;

[0008] determining the change trend of the residual current according to the preprocessed historical residual current data, and determining the residual current mode of different network levels;

[0009] screening the associated factors of the residual current in combination with the change trend of the residual current and the residual current mode, constructing a residual current prediction model based on the grey system theory, and performing baseline prediction of the residual current according to the change trend of the residual current and the residual current prediction model;

[0010] The residual current baseline prediction based on the residual current prediction model and the preset threshold value of the power distribution system simulation model are set;

[0011] When the prediction model predicts that the residual current exceeds the preset threshold value, a warning is triggered.

[0012] Optionally, the grid structure of the power distribution system includes grid levels and nodes and connection methods of the grid levels; the grid levels include a main power distribution layer, a secondary power distribution layer, and an end user access layer.

[0013] Optionally, the historical residual current data includes residual current data of different grid levels without photovoltaic access under different working conditions, residual current data of different grid levels with photovoltaic access under different working conditions, configuration parameters and operation data of distributed photovoltaic access points, and environmental data; wherein the environmental data includes environmental temperature and environmental humidity.

[0014] The correlation factors include at least one of the operating state of the power grid equipment, environmental factors, load conditions, power grid parameters, power grid connection methods, and line states.

[0015] Optionally, the historical residual current data of different grid levels in the grid structure of the power distribution system is obtained, and the historical residual current data is preprocessed, including:

[0016] The historical residual current data is removed from outliers and noise, data dimensionality reduction and feature extraction are performed, and normalization or standardization processing is performed.

[0017] Optionally, the historical residual current data is removed from outliers and noise, data dimensionality reduction and feature extraction are performed, and normalization or standardization processing is performed, including:

[0018] Anomaly values and noise are identified and removed using a machine learning algorithm, and data dimensionality reduction and feature extraction are performed using principal component analysis.

[0019] Optionally, the correlation factors of the residual current are screened in combination with the change trend of the residual current and the residual current mode, including:

[0020] The correlation factors are screened through grey correlation analysis or principal component analysis.

[0021] Optionally, the residual current prediction model is constructed based on the grey system theory, including:

[0022] A preliminary prediction model is established using the grey system theory;

[0023] The parameters in the preliminary model are determined by the least squares method;

[0024] The preliminary model is tested and corrected to determine the residual current prediction model.

[0025] Optionally, after the residual current prediction model is constructed based on the grey system theory, the method further comprises:

[0026] Periodically comparing the residual current predicted by the residual current prediction model with the actual monitored residual current, and adjusting the parameters of the residual current prediction model according to the comparison result.

[0027] Optionally, the residual current prediction model adopts a Transformer model, a convolutional neural network model or a long short-term memory network model.

[0028] According to a second aspect of the present application, a multi-level active power distribution network residual current early warning device is provided, comprising:

[0029] A simulation module is configured to establish a power distribution system simulation model based on the network structure of the power distribution system, and determine the access point of the distributed photovoltaic system in the power distribution system and the position of the distributed photovoltaic system in the power distribution network.

[0030] A data collection and preprocessing module is configured to obtain historical residual current data of different network levels in the network structure of the power distribution system, and preprocess the historical residual current data.

[0031] A residual current mode determination module is configured to determine the residual current mode of different network levels according to the preprocessed historical residual current data and determine the change trend of the residual current through time series.

[0032] A prediction module is configured to screen the associated factors of the residual current in combination with the change trend of the residual current and the residual current mode, construct a residual current prediction model based on the grey system theory, and perform baseline prediction of the residual current according to the change trend of the residual current and the residual current prediction model.

[0033] A preset threshold determination module is configured to set a preset threshold based on the residual current baseline prediction of the residual current prediction model and the power distribution system simulation model.

[0034] An early warning module is configured to trigger early warning when the prediction model predicts that the residual current exceeds the preset threshold.

[0035] The multi-level active power distribution network residual current early warning method provided by the embodiment of the application comprises the following steps: a power distribution system simulation model is established based on the network structure of the power distribution system, and the access point of the distributed photovoltaic system in the power distribution system and the position of the distributed photovoltaic system in the power distribution network are determined, so as to ensure that the simulation model can accurately reflect the operating condition of the actual power distribution system; the historical residual current data of different network levels in the network structure of the power distribution system is acquired, and the historical residual current data is preprocessed, so as to ensure the quality and accuracy of the data used for analysis; the change trend of the residual current is determined according to the preprocessed historical residual current data and through time series, and the correlation factor and change rule analysis are helpful to understand the change mode of the residual current, and then the operating and maintenance strategy of the power distribution system is optimized, and the residual current mode of different network levels is determined; the correlation factor of the residual current is screened in combination with the change trend of the residual current and the residual current mode, a residual current prediction model is constructed based on the grey system theory, the combination of the prediction model and the simulation model can provide decision support for the operation and maintenance personnel, help the operation and maintenance personnel to more effectively allocate resources and arrange maintenance work, improve the operation and maintenance efficiency, and in combination with the historical data and the prediction model, the change of the residual current can be effectively monitored and predicted, and abnormal conditions can be found in time, which is helpful to prevent electrical fires and other related safety accidents, so as to improve the safety and reliability of the entire power distribution system; and the residual current is baseline predicted according to the change trend of the residual current and the residual current prediction model; a preset threshold is set based on the residual current baseline prediction of the residual current prediction model and the power distribution system simulation model; when the prediction model predicts that the residual current exceeds the preset threshold, an early warning is triggered, an early warning signal can be sent before the residual current exceeds the safe range, so that the operation and maintenance personnel can take measures in time to avoid potential safety risks, and the intelligent early warning mechanism can reduce the frequency and intensity of manual inspection, reduce the operation and maintenance cost, and ensure the continuous and stable operation of the system. Through the historical data processing and residual current prediction, the residual current in the power distribution network can be effectively monitored and managed, an alarm is sent when the residual current exceeds the preset threshold, the safety, reliability and economy of the power distribution network can be effectively improved, and the method is especially suitable for the multi-level active power distribution network containing the distributed photovoltaic system. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 is a flowchart of a multi-level active power distribution network residual current early warning method provided by the embodiment of the application;

[0037] Figure 2 is a structural schematic diagram of a multi-level active power distribution network residual current early warning device provided by the embodiment of the application. DETAILED DESCRIPTION

[0038] The application will be described in further detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only intended for explanation of the application and are not limiting of the application. In addition, it should be noted that only the parts related to the application are shown in the drawings for the convenience of description.

[0039] The embodiment of the application provides a residual current early warning method for a multi-level active power distribution network, Figure 1 The embodiment of the application provides a residual current early warning method for a multi-level active power distribution network, Figure 1 The residual current early warning method for the multi-level active power distribution network comprises the following steps.

[0040] S110, establishing a power distribution system simulation model based on a network structure of the power distribution system, and determining an access point of a distributed photovoltaic system in the power distribution system and a position of the distributed photovoltaic system in the power distribution network.

[0041] In the network level division and photovoltaic access point identification of the residual current early warning method for the multi-level active power distribution network, the advanced geographic information system technology can be used to realize accurate three-dimensional modeling of the network structure of the power distribution system, to clearly determine the spatial layout of different level nodes and connection modes, and to improve the identification accuracy of the distributed photovoltaic system access point. The process can optimize system design, enhance visual management, improve operation efficiency, support decision making, promote resource management, improve safety, promote data sharing and cooperation, adapt to regulatory requirements and long-term planning support. The above advantages collectively improve the operation efficiency, safety and economy of the power distribution system, and provide strong technical support for sustainable development of the power system.

[0042] S120, acquiring historical residual current data of different network levels in the network structure of the power distribution system, and preprocessing the historical residual current data.

[0043] The Internet of Things sensor technology is used to monitor and collect historical residual current data and environmental variables of each level network in real time, to provide real-time, accurate and high-quality original data for the residual current early warning method for the multi-level active power distribution network. The data comprehensiveness and accuracy are enhanced, resource allocation is optimized, fault diagnosis capability is strengthened, and overall reliability of the early warning system is improved, thereby providing a solid data foundation for realizing intelligent energy management and system optimization.

[0044] S130, determining a residual current mode of different network levels according to the preprocessed historical residual current data and determining a change trend of the residual current through time series.

[0045] S140, screen the correlation factor of the residual current based on the change trend of the residual current and the residual current mode, construct a residual current prediction model based on the grey system theory, and predict the baseline of the residual current based on the change trend of the residual current and the residual current prediction model.

[0046] The grey system theory is to regard random variables as grey variables changing within a certain range and random processes as grey processes changing within a certain range and time zone. The grey system theory converts irregular original data into regular generated sequences through generating functions and grey differential equations, and then establishes a model for prediction and control. Baseline prediction refers to a long-term trend prediction of residual current in different network levels of the power distribution system based on the residual current prediction model. This prediction aims to provide a reference benchmark for comparison with the actual observed residual current data to assess whether the operation of the power distribution system is abnormal.

[0047] The residual current mode includes a normal mode, a sudden change mode, a fluctuation mode, and a noise mode. In the normal mode, the residual current is small and stable. In the sudden change mode, the residual current suddenly increases when there is a fault or abnormal situation in the circuit, such as equipment damage or insulation aging. In the fluctuation mode, the residual current may fluctuate periodically under certain conditions. In the noise mode, the residual current may exhibit certain noise characteristics due to electromagnetic interference, internal device noise, and other factors.

[0048] Observe the change of residual current over time to determine whether it is gradually rising, falling, or fluctuating. A curve graph of residual current over time can be drawn to visually understand its change rule. Calculate the change rate, variance, and other statistical indicators of the residual current to further quantify the degree of change, determine the change trend and mode, and use the residual current prediction model to predict the residual current in the future period. Based on the results of the prediction model and the residual current range during normal equipment operation, determine the baseline value. The baseline value can be a fixed value or a range.

[0049] S150, residual current baseline prediction based on the residual current prediction model and setting a preset threshold value for the power distribution system simulation model.

[0050] According to the determined baseline value, in combination with the prediction of the future change trend of the residual current, a preset threshold is set. Generally, the threshold can be set by adding a reasonable deviation range to the baseline value. The change of the residual current under different working conditions is simulated by using the power distribution system simulation model, and the response and safety of the system under different threshold settings are observed. According to the simulation results, the preset threshold is adjusted and optimized. If the simulation results show that the residual current is likely to exceed the threshold under certain working conditions, resulting in false alarm or equipment shutdown, the threshold can be appropriately increased. If it is found that the threshold is set too high and cannot timely discover abnormal conditions, the threshold can be reduced. When setting the preset threshold, the safety of the equipment and the actual operation demand should be fully considered. The threshold cannot be set too high, so as to ignore potential safety hazards; nor can it be set too low, resulting in frequent false alarms and equipment shutdown, affecting normal production and life. At the same time, the particularity of different equipment and different systems should also be considered, and individualized threshold setting schemes should be developed according to the actual situation.

[0051] S160, when the prediction model predicts that the residual current exceeds the preset threshold, triggering an early warning.

[0052] Specifically, based on the network structure of the power distribution system, a power distribution system simulation model is built by using professional simulation software, the access point of the distributed photovoltaic system in the power distribution system and the position of the distributed photovoltaic system in the power distribution network are determined, the historical residual current data of different network levels is obtained from the network structure of the power distribution system, the historical residual current data is preprocessed, including data cleaning, denoising, missing value processing, etc., the correlation analysis method is used to screen the associated factors of the residual current according to the preprocessed historical residual current data, the time series analysis method is used to determine the change trend of the residual current, the change trend of the residual current and the associated factors are determined, the residual current mode of different network levels is determined, the grey system theory is used, the change trend of the residual current and the residual current mode are combined, the residual current prediction model is constructed, the baseline of the residual current is predicted according to the residual current prediction model, the baseline prediction result of the power distribution system simulation model and the residual current prediction model is set as the preset threshold, when the prediction model predicts that the residual current exceeds the preset threshold, the early warning mechanism is triggered, and the relevant personnel are timely informed to handle.

[0053] The multi-level active power distribution network residual current early warning method provided by the embodiment of the application comprises the following steps: a power distribution system simulation model is established based on the network structure of the power distribution system, and the access point of the distributed photovoltaic system in the power distribution system and the position of the distributed photovoltaic system in the power distribution network are determined, so as to ensure that the simulation model can accurately reflect the operating condition of the actual power distribution system; the historical residual current data of different network levels in the network structure of the power distribution system is obtained, and the historical residual current data is preprocessed, so as to ensure the quality and accuracy of the data used for analysis; the change trend of the residual current is determined according to the preprocessed historical residual current data and through time series, and the correlation factor and change rule analysis are helpful to understand the change mode of the residual current, and then the operating and maintenance strategy of the power distribution system is optimized, and the residual current mode of different network levels is determined; the correlation factor of the residual current is screened in combination with the change trend of the residual current and the residual current mode, a residual current prediction model is constructed based on the grey system theory, the combination of the prediction model and the simulation model can provide decision support for the operation and maintenance personnel, help the operation and maintenance personnel to more effectively allocate resources and arrange maintenance work, improve the operation and maintenance efficiency, and in combination with the historical data and the prediction model, the change of the residual current can be effectively monitored and predicted, and abnormal conditions can be found in time, which is helpful to prevent electrical fires and other related safety accidents, so as to improve the safety and reliability of the entire power distribution system; and the residual current is baseline predicted according to the change trend of the residual current and the residual current prediction model; a preset threshold is set based on the residual current baseline prediction of the residual current prediction model and the power distribution system simulation model; when the prediction model predicts that the residual current exceeds the preset threshold, an early warning is triggered, an early warning signal can be sent before the residual current exceeds the safe range, so that the operation and maintenance personnel can take measures in time to avoid potential safety risks, and the intelligent early warning mechanism can reduce the frequency and intensity of manual inspection, reduce the operation and maintenance cost, and ensure the continuous and stable operation of the system. Through the historical data processing and residual current prediction, the residual current in the power distribution network can be effectively monitored and managed, an alarm is sent when the residual current exceeds the preset threshold, the safety, reliability and economy of the power distribution network can be effectively improved, and the method is especially suitable for the multi-level active power distribution network containing the distributed photovoltaic system.

[0054] Optionally, the network structure of the power distribution system comprises network levels, nodes of the network levels and connection modes; the network levels comprise a main power distribution layer, a secondary power distribution layer and an end user access layer.

[0055] The three points in the main power distribution layer, the secondary power distribution layer and the end user access layer level can be used to determine the access point of the distributed photovoltaic system and the specific position of the distributed photovoltaic system in the power distribution network, which is helpful for subsequent data processing and analysis, ensures the accuracy and integrity of the data, improves the efficiency and accuracy of system management, provides a clear framework for subsequent data collection and preprocessing, facilitates the identification and positioning of potential problem points, and provides a basis for system optimization.

[0056] Optionally, the historical residual current data includes residual current data of different grid levels under different working conditions without photovoltaic access, residual current data of different grid levels under different working conditions with photovoltaic access, configuration parameters and operation data of the distributed photovoltaic access point, and environmental data; wherein the environmental data includes environmental temperature and environmental humidity.

[0057] The correlation factor includes at least one of the operating state of the power grid equipment, environmental factors, load conditions, power grid parameters, power grid wiring methods and line states.

[0058] Specifically, the historical residual current data of each level grid and environmental variables such as temperature and humidity are monitored in real time by using Internet of Things sensor technology, which ensures the real-time and accuracy of the data and provides high-quality raw data for subsequent analysis.

[0059] Optionally, the historical residual current data of different grid levels in the grid structure of the power distribution system is obtained, and the historical residual current data is preprocessed, including:

[0060] The historical residual current data is removed from outliers and noise, and data dimension reduction and feature extraction are performed, and normalization or standardization processing is performed.

[0061] The normalization or standardization processing is used to eliminate the influence of dimension, improve the convergence speed of the model, and improve the performance of the model. By preprocessing the data, the error and bias of subsequent analysis are reduced, and the normalization or standardization processing helps to improve the generalization ability of the model, which can ensure the quality and consistency of the data and improve the reliability of data analysis.

[0062] Optionally, the historical residual current data is removed from outliers and noise, and data dimension reduction and feature extraction are performed, and normalization or standardization processing is performed, including:

[0063] Machine learning algorithms are used to identify and remove outliers and noise, and principal component analysis is used for data dimension reduction and feature extraction.

[0064] Among them, introducing machine learning algorithms and advanced data processing technologies such as principal component analysis into the data preprocessing process can automatically identify and eliminate outliers and noise, and perform data dimensionality reduction and feature extraction, thereby improving data quality and processing efficiency. This method can improve data quality, reduce manual intervention, improve processing efficiency, strengthen data standardization, reduce data dimensions, highlight key variables, improve model performance, enhance model generalization capabilities, support complex analysis and maintain system stability, and can ensure the accuracy, reliability and efficiency of the early warning system.

[0065] Optionally, the residual current correlation factors can be screened based on the residual current change trend and residual current mode, including:

[0066] The correlation factors were screened through grey correlation analysis or principal component analysis.

[0067] Specifically, it is used to extract features that may affect the change of residual current, identify the main factors affecting the residual current, screen the related factors, analyze the changing trend of the residual current, and identify the residual current patterns at different grid levels.

[0068] Optionally, a residual current prediction model is constructed based on grey system theory, including:

[0069] Establish a preliminary forecasting model using grey system theory;

[0070] The parameters in the preliminary model were determined by the least squares method;

[0071] The preliminary model is tested and revised to determine the residual current prediction model.

[0072] Among them, the grey system theory is suitable for dealing with incomplete information and suitable for complex systems such as power distribution networks. Based on the grey theory, a current prediction model is constructed. Data collection and preprocessing: Collect historical data of residual current, including residual current values under different working conditions, environmental temperature, humidity and other related factors. Clean and preprocess the data to remove outliers and missing values, and ensure the accuracy and reliability of the data. Establish a grey system model: Treat the residual current system as a grey system containing known and unknown information. According to the trend and pattern of residual current, determine the input (independent variable) and output (dependent variable) of the system, i.e. the factors affecting the residual current as input and the residual current value as output. Use the modeling method in the grey system theory, such as GM(1,1) model, to establish a residual current prediction model. GM(1,1) model is a commonly used grey prediction model, which generates and processes the original data once to weaken randomness and enhance regularity, and then establishes a differential equation model for prediction. Model solving and parameter estimation: Estimate the parameters in the model by least squares method and other methods. Solve the model to get the predicted value of the residual current. Model testing and correction: Test the established grey prediction model, common testing methods include residual test, correlation test and posterior difference test. If the accuracy of the model does not meet the requirements, the model can be modified and optimized by adding correction terms, adjusting model parameters, etc. Result analysis and application: According to the predicted residual current value, analyze and apply it in combination with the actual situation. For example, you can determine whether the residual current is abnormal and take appropriate measures in advance; or combine the prediction results with other safety monitoring systems to improve the overall safety performance.

[0073] Optionally, after constructing the residual current prediction model based on the grey system theory, it further includes:

[0074] Periodically compare the residual current predicted by the residual current preset model with the actual monitored residual current, and adjust the parameters of the residual current prediction model according to the comparison results.

[0075] Among them, the prediction results are evaluated and compared with the actual monitoring data to evaluate the accuracy of the prediction model, and the model parameters are adjusted regularly to adapt to system changes, which not only improves the accuracy and reliability of the prediction model, but also enhances decision support, improves the stability and resource utilization efficiency of the system, and also supports long-term planning of the power system and improves user trust.

[0076] Optionally, the residual current prediction model uses a Transformer model, a convolutional neural network model or a long short-term memory network model.

[0077] The transformer model has the advantages of strong ability to process long-distance dependencies, strong parallel computing ability, high scalability and flexibility, and can more accurately predict the change trend of the residual current, improve the accuracy and timeliness of the early warning, and can generate the early warning result faster due to high calculation efficiency, thereby providing real-time protection for the safe operation of the power distribution network. The convolutional neural network model has the advantages of local perception and weight sharing, strong feature extraction capability, and sensitivity to image and sequence data, can extract useful features from the residual current data, improve the prediction accuracy, and is sensitive to abnormal changes in the power distribution network, so that potential faults can be discovered and warned in time. The long short-term memory network model has the advantages of strong ability to process sequence data, sensitivity to time series data and strong robustness, and can accurately predict the future change trend of the residual current, thereby providing reliable protection for the safe operation of the power distribution network, and is sensitive to abnormal changes in the time series data, so that potential faults can be discovered and warned in time.

[0078] The embodiment of the present application also provides a multi-level active power distribution network residual current early warning device, Figure 2 It is a structural schematic diagram of a multi-level active power distribution network residual current early warning device provided by the embodiment of the present application, referring to Figure 2 The multi-level active power distribution network residual current early warning device comprises:

[0079] The simulation module 210 is configured to establish a power distribution system simulation model based on the grid structure of the power distribution system, and determine the access point of the distributed photovoltaic system in the power distribution system and the position of the distributed photovoltaic system in the power distribution network.

[0080] The data collection and preprocessing module 220 is configured to obtain historical residual current data of different grid levels in the grid structure of the power distribution system, and preprocess the historical residual current data.

[0081] The residual current mode determination module 230 is configured to determine the change trend of the residual current according to the preprocessed historical residual current data, and determine the residual current mode of different grid levels through time series.

[0082] The prediction module 240 is configured to combine the change trend of the residual current and the residual current mode to filter the associated factors of the residual current, construct a residual current prediction model based on the grey system theory, and perform baseline prediction on the residual current according to the change trend of the residual current and the residual current prediction model.

[0083] The preset threshold determination module 250 is configured to set a preset threshold based on the residual current baseline prediction of the residual current prediction model and the power distribution system simulation model.

[0084] The early warning module 260 is configured to trigger early warning when the prediction model predicts that the residual current exceeds the preset threshold.

[0085] The technical scheme of the embodiment is used to establish a power distribution system simulation model based on the network structure of the power distribution system, and determine the access point of the distributed photovoltaic system in the power distribution system and the position of the distributed photovoltaic system in the power distribution network through the simulation module; the data collection and preprocessing module is used to obtain historical residual current data of different network levels in the network structure of the power distribution system, and preprocess the historical residual current data; the residual current mode determination module is used to determine the residual current mode of different network levels according to the preprocessed historical residual current data and by determining the change trend of the residual current through time series; the prediction module is used to screen the associated factors of the residual current in combination with the change trend of the residual current and the residual current mode, construct a residual current prediction model based on the gray system theory, and perform baseline prediction on the residual current according to the change trend of the residual current and the residual current prediction model; the preset threshold determination module is used to set a preset threshold based on the residual current baseline prediction of the residual current prediction model and the power distribution system simulation model; and the early warning module is used to trigger early warning when the prediction model predicts that the residual current exceeds the preset threshold. Through the processing of historical data and the prediction of residual current, the residual current in the power distribution network can be effectively monitored and managed, and an alarm is issued to remind when the residual current exceeds the preset threshold, which can effectively improve the safety, reliability and economy of the power distribution network, and is especially suitable for multi-level active power distribution networks containing distributed photovoltaic systems.

[0086] Optionally, the network structure of the power distribution system includes network levels, nodes of the network levels, and connection modes; the network levels include a main power distribution layer, a secondary power distribution layer, and an end user access layer.

[0087] Optionally, the data collection and preprocessing module 220 is specifically used for:

[0088] The historical residual current data is de-noised to remove outliers, and data dimension reduction and feature extraction are performed, and normalization or standardization processing is performed.

[0089] Optionally, the data collection and preprocessing module 220 is specifically used for:

[0090] Anomaly values and noises are identified and removed by using a machine learning algorithm, and data dimension reduction and feature extraction are performed by using principal component analysis.

[0091] Optionally, the residual current mode determination module 230 is specifically used for:

[0092] The associated factors are screened by gray correlation analysis or principal component analysis.

[0093] Optionally, the prediction module 240 is specifically used for:

[0094] A preliminary prediction model is established by using the gray system theory;

[0095] The parameters in the preliminary model are determined by a least square method;

[0096] The preliminary model is tested and corrected to determine the residual current prediction model.

[0097] Optionally, the prediction module 240 further comprises:

[0098] The residual current predicted by the residual current preset model is compared with the actually monitored residual current at regular intervals, and the parameters of the residual current prediction model are adjusted according to the comparison result.

[0099] In some embodiments, the residual current prediction model adopts a Transformer model, a convolutional neural network model or a long short-term memory network model.

[0100] It should be understood that the various forms of flow shown above can be used to reorder, add or delete steps. For example, the steps described in the present application can be executed in parallel, sequentially or in a different order, as long as the desired results of the technical solutions of the present application can be achieved, which is not limited herein.

[0101] The above detailed description does not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A multi-level active distribution network residual current early warning method, characterized in that, The method comprises the following steps: establishing a power distribution system simulation model based on the grid structure of the power distribution system, and determining the access point of the distributed photovoltaic system in the power distribution system and the location of the distributed photovoltaic system in the power distribution network; obtaining historical residual current data of different grid levels in the grid structure of the power distribution system, and preprocessing the historical residual current data; determining the residual current mode of different grid levels according to the preprocessed historical residual current data and the time sequence to determine the change trend of the residual current; screening the associated factors of the residual current in combination with the change trend of the residual current and the residual current mode, constructing a residual current prediction model based on the grey system theory, and performing baseline prediction of the residual current according to the change trend of the residual current and the residual current prediction model; setting a preset threshold based on the residual current baseline prediction of the residual current prediction model and the power distribution system simulation model; triggering a warning when the prediction model predicts that the residual current exceeds the preset threshold.

2. The multi-level active power distribution network residual current warning method according to claim 1, wherein: the grid structure of the power distribution system comprises grid levels and nodes and connection modes of the grid levels; the grid levels comprise a main power distribution layer, a secondary power distribution layer, and a terminal user access layer.

3. The multi-level active power distribution network residual current warning method according to claim 1, wherein: the historical residual current data comprises residual current data of different grid levels under different working conditions without photovoltaic access, residual current data of different grid levels under different working conditions with photovoltaic access, configuration parameters and operation data of the distributed photovoltaic access point, and environmental data; the environmental data comprises environmental temperature and environmental humidity; the associated factors comprise at least one of the operating state of the power grid equipment, environmental factors, load conditions, power grid parameters, power grid connection modes, and line states.

4. The multi-tier active distribution network residual current pre-warning method according to claim 1, characterized in that, Obtaining historical residual current data of different grid levels in the grid structure of the power distribution system and preprocessing the historical residual current data comprises: removing outliers and noise from the historical residual current data, performing data dimensionality reduction and feature extraction, and performing normalization or standardization processing.

5. The multi-tier active distribution network residual current pre-warning method according to claim 4, characterized in that, Removing outliers and noise from the historical residual current data, performing data dimensionality reduction and feature extraction, and performing normalization or standardization processing comprises: using a machine learning algorithm to identify and remove outliers and noise, and using principal component analysis to perform data dimensionality reduction and feature extraction.

6. The multi-tier active distribution network residual current pre-warning method according to claim 1, characterized in that, Screening the associated factors of the residual current in combination with the change trend of the residual current and the residual current mode comprises: screening the associated factors through grey correlation analysis or principal component analysis.

7. The multi-tier active distribution network residual current pre-warning method according to claim 1, characterized in that, Constructing a residual current prediction model based on the grey system theory comprises: establishing a preliminary prediction model using the grey system theory; determining the parameters in the preliminary model through the least squares method; testing and correcting the preliminary model to determine the residual current prediction model.

8. The multi-tier active distribution network residual current pre-warning method according to claim 1, characterized in that, After constructing the residual current prediction model based on the grey system theory, the method further comprises: periodically comparing the residual current predicted by the residual current preset model with the actual monitored residual current, and adjusting the parameters of the residual current prediction model according to the comparison result.

9. The multi-tier active power distribution network residual current pre-warning method according to claim 1, characterized in that, The residual current prediction model adopts a Transformer model, a convolutional neural network model or a long short-term memory network model.

10. A multi-level active distribution network residual current early warning device, characterized in that, Comprise: The simulation module is used for establishing a power distribution system simulation model based on the grid structure of the power distribution system, and determining the access point of the distributed photovoltaic system in the power distribution system and the position of the distributed photovoltaic system in the power distribution network; The data collection and preprocessing module is used for obtaining historical residual current data of different grid levels in the grid structure of the power distribution system, and preprocessing the historical residual current data; The residual current mode determination module is used for determining the residual current mode of different grid levels according to the preprocessed historical residual current data and determining the change trend of the residual current through time series; The prediction module is used for screening the associated factors of the residual current in combination with the change trend of the residual current and the residual current mode, constructing a residual current prediction model based on the grey system theory, and performing baseline prediction on the residual current according to the change trend of the residual current and the residual current prediction model; The preset threshold determination module is used for setting a preset threshold based on the residual current baseline prediction of the residual current prediction model and the power distribution system simulation model; The early warning module is used for triggering early warning when the prediction model predicts that the residual current exceeds the preset threshold.

Images