Data mining method and system serving micro-grid control system

Abstract

The invention discloses a data mining method and system serving a micro-grid control system, and relates to the technical field of micro-grids, and the method comprises the steps: collecting the related data of a micro-grid, constructing a micro-grid data set S, carrying out the feature extraction, obtaining a voltage stability factor Sv (t), a frequency deviation factor Sf (t) and a harmonic total distortion rate THD, and obtaining a power quality comprehensive score Qelec, the method comprises the following steps: judging whether a micro-grid has a fault or not, if the micro-grid has the fault, carrying out fault positioning, constructing an aging model of an energy storage battery and an inverter, and for a line where a fault point is located and the energy storage battery and the inverter of the fault point, obtaining the residual use time RULb of the energy storage battery and the residual use time RULinv of the inverter, and the Dijkstra shortest path algorithm is used to generate the optimal fault repair path Popt for fault repair, so that the efficiency of micro-grid fault detection and repair is improved, the service life of equipment is prolonged, the maintenance cost is reduced, and the stability of micro-grid operation and the reliability of power supply are improved.

Description

Technical Field

[0001] The present invention relates to the field of microgrid technology, and in particular to a data mining method and system serving a microgrid control system. Background Art

[0002] Microgrid technology is an important part of smart grid. It provides stable and flexible power supply to users in the region through the integrated operation of distributed energy sources such as solar energy, wind energy, energy storage equipment and loads. In the field of microgrid, power quality management and equipment health maintenance are two core research directions to ensure efficient and safe operation of the system. Power quality issues, such as voltage fluctuations, frequency anomalies and harmonic distortion, and equipment health status, such as energy storage battery aging and inverter overload, directly affect the stability and reliability of microgrids. Especially in the modern complex and diverse energy environment, these issues have put forward higher requirements for the operation and maintenance of microgrids.

[0003] In the existing microgrid system, power quality anomaly detection and equipment health prediction are often processed separately. Traditional power quality detection mostly adopts analysis methods based on a single indicator, such as voltage fluctuation amplitude and frequency deviation, which is difficult to fully reflect the power quality problems under multi-factor coupling. The monitoring of equipment health status relies more on regular maintenance and static historical data analysis, which is difficult to reflect the changes in the health status of equipment in real time under power fluctuation environment. This separate monitoring and static analysis method makes it difficult for the system to take optimization measures in time under abnormal conditions. There are deficiencies such as problem discovery delay, difficulty in accurately predicting equipment aging, and low fault recovery efficiency. When power quality problems and equipment health problems are not effectively identified, it may lead to a decrease in microgrid operation efficiency or systemic interruption. For example, frequency fluctuations will accelerate the aging of energy storage batteries and even cause short-circuit faults. Inverter overload may cause power supply interruption due to failure to predict in time. In addition, frequent equipment failures increase maintenance costs and reduce users' electricity satisfaction. In severe cases, it will also affect the stability and security of regional energy supply. These problems not only restrict the application and promotion of microgrids, but also pose a threat to the sustainable development of regional energy supply. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a data mining method and system serving a microgrid control system, which solves the problems in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a data mining system serving a microgrid control system, including a data acquisition module, a feature extraction and anomaly assessment module, a fault location module, an equipment health monitoring module and an intelligent repair module;

[0006] The data acquisition module is used to collect microgrid related data in real time, and construct a microgrid data set S based on the collected microgrid related data;

[0007] The feature extraction and abnormality assessment module is used to perform preprocessing based on the microgrid data set S, and perform summary calculation based on the preprocessed microgrid data set S to obtain the voltage stability factor S v (t), frequency deviation factor S f (t) and total harmonic distortion THD, and according to the voltage stability factor S v (t), frequency deviation factor S f (t) and total harmonic distortion THD to obtain the comprehensive power quality score Q elec , and preset the power score threshold Q normal , the power score threshold Q normal Power quality comprehensive score Q elec Conduct comparative analysis to assess whether the microgrid has a fault. elec >Q normal , it indicates that the microgrid has a fault, and the fault location module is triggered;

[0008] The fault location module is used to set a number of data monitoring points in the microgrid, collect voltage waveform data of the data monitoring points, determine the location of the fault point according to the principle of traveling wave propagation, and obtain the energy storage battery and inverter information of the line where the fault point is located, and generate a fault point device set G;

[0009] The equipment health monitoring module is used to monitor the power quality of the microgrid based on the microgrid data set S and the power quality comprehensive score Q elec , use linear regression algorithm to build energy storage battery aging model and inverter aging model to obtain energy storage battery aging rate and inverter aging rate And calculate the energy storage battery health score SoH b (t) and inverter health score R inv (t), based on the energy storage battery health score SoH b (t) and inverter health score R inv (t), perform comprehensive calculation to obtain the remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv ;

[0010] The intelligent repair module is used to obtain the remaining service life RUL of the energy storage battery of the line where the fault point is located according to the fault point equipment set G b And the remaining use time RUL of the inverter inv, perform device screening, obtain backup devices, and use Dijkstra's shortest path algorithm to obtain the optimal fault repair path P from the backup device starting point to the fault point opt , and based on the optimal fault repair path P opt Perform troubleshooting.

[0011] Preferably, the data acquisition module is used to deploy an intelligent sensor group on the microgrid, collect microgrid related data in real time, and construct a microgrid data set S based on the collected microgrid related data, wherein the microgrid related data includes power quality data and equipment operation data;

[0012] The intelligent sensor group includes a voltage monitor, a current monitor, a frequency detector, a thermal resistance temperature sensor and a power measurement instrument;

[0013] The power quality data includes voltage V, current I and power frequency f;

[0014] The equipment operation data includes the battery state of charge SOC of the energy storage battery, the energy storage battery load rate L SOC , Energy storage battery operating temperature T b , inverter operating temperature T inv , inverter load factor L inv And the line transmission power C current .

[0015] Preferably, the feature extraction and anomaly assessment module comprises a feature extraction unit and an assessment unit;

[0016] The feature extraction unit is used to perform preprocessing based on the microgrid data set S, and obtain a voltage stability factor S based on the preprocessed microgrid data set S. v (t), frequency deviation factor S f (t) and total harmonic distortion THD, wherein the preprocessing includes data cleaning, denoising and data standardization;

[0017] The voltage stability factor S v (t) The method of obtaining is:

[0018]

[0019] Where T represents the sampling time window, V(θ) represents the voltage at time point θ, V(θ+Δt) represents the voltage at time point θ+Δt, θ∈[tT, t], Δt represents the sampling time interval;

[0020] The frequency deviation factor S f (t) The method of obtaining is:

[0021]

[0022] Where f(θ) represents the grid frequency at time θ, f nom Indicates the standard frequency value of the power grid;

[0023] According to the power quality data in the microgrid data set S, the voltage V is decomposed in the frequency domain using fast Fourier transform, and the voltage waveform in the time domain is converted to the frequency domain to obtain the fundamental amplitude H 1 and harmonic amplitude H n , and according to the fundamental amplitude H 1 and harmonic amplitude H n , get the total harmonic distortion THD, where the harmonic amplitude H n The specific acquisition method is as follows:

[0024] H n =n·H 1 ;

[0025] In the formula, H 1 represents the fundamental amplitude, and n represents the harmonic order;

[0026] The total harmonic distortion THD is obtained as follows:

[0027]

[0028] In the formula, H n Indicates harmonic amplitude, n≥2.

[0029] Preferably, the evaluation unit is used to determine the voltage stability factor S v (t), frequency deviation factor S f (t) and total harmonic distortion THD, and then calculate and obtain the comprehensive power quality score Q elec , the comprehensive score of power quality Q elec The method of obtaining is:

[0030] Q elec =α 1 ·S v (t)+α 2 ·S f (t)+α 3 THD+C;

[0031] In the formula, α 1 , α 2 and α 3 Represents the voltage stability factor S v (t), frequency deviation factor S f (t) weight coefficient of total harmonic distortion THD, C represents the first correction constant;

[0032] Preset power score threshold Q normal , and the power score threshold Q normal Power quality comprehensive score Q elec Conduct comparative analysis to assess whether the microgrid has a fault. elec ≤Q normal , it indicates that the microgrid has not failed and no intervention is required; if Q elec >Q normal , it indicates that a fault occurs in the microgrid, and the fault location module is triggered to locate the fault of the microgrid.

[0033] Preferably, the fault location module is used to set a number of data monitoring points in the microgrid, and when a fault occurs in the microgrid, collect voltage waveform data of each data monitoring point in real time, wherein the voltage waveform data refers to the waveform change of the voltage signal;

[0034] According to the voltage waveform data obtained at each data monitoring point, the two data monitoring points A and B with the largest change in the amplitude of the voltage traveling wave signal are selected, and the time difference Δt of the voltage traveling wave signal reaching the data monitoring points A and B is obtained. AB , and according to the principle of traveling wave propagation, randomly select any data monitoring point from data monitoring points A and B, and calculate the distance d between any data monitoring point in data monitoring points A and B and the fault point. The specific calculation method is as follows:

[0035]

[0036] In the formula, v h represents the propagation speed of the traveling wave, Δt AB It represents the time difference between the arrival of the traveling wave at data monitoring point A and data monitoring point B;

[0037] According to the microgrid control system, the microgrid topology map is obtained, and the data monitoring point A and the data monitoring point B are marked in the microgrid topology map. Combined with the distance d between any data monitoring point A and B and the fault point, the line where the fault point is located is located in the microgrid topology map and the position of the fault point is marked. The energy storage battery data and inverter data of the line where the fault point is located are collected to generate the fault point device set G.

[0038] Preferably, the equipment health monitoring module includes an equipment aging rate calculation unit and a health status analysis unit;

[0039] The device aging rate calculation unit is used to calculate the device aging rate of the fault point according to the fault point device set G, for the line where the fault point is located and the energy storage battery and inverter of the line where the fault point is located, according to the microgrid data set S and the comprehensive power quality score Q elec, a linear regression algorithm is used to construct an energy storage battery aging model and an inverter aging model. The energy storage battery aging model is specifically expressed as follows:

[0040]

[0041] In the formula, represents the aging rate of the energy storage battery at the current time point t, Q elec (t) represents the comprehensive score of power quality at the current time point t, T b (t) represents the operating temperature of the energy storage battery at the current time point t, SOC(t) represents the battery state of charge of the energy storage battery at the current time point t, k 1 , k 2 and k 3 represents the regression coefficient;

[0042] The inverter aging model is specifically expressed as follows:

[0043]

[0044] In the formula, Indicates the inverter aging rate at the current time point t, L inv (t) represents the inverter load rate at the current time point t, T inv (t) represents the inverter operating temperature at the current time point t, m 1 、m 2 and m 3 represents the regression coefficient;

[0045] Collect historical microgrid operation data and divide it into training set and test set. Use the training set to input the energy storage battery aging model and inverter aging model for training. Use the least squares method to fit the parameters of the energy storage battery aging model and inverter aging model to obtain the regression coefficient k. 1 , k 2 , k 3 、m 1 、m 2 and m 3 , and use the test set to verify the energy storage battery aging model and inverter aging model.

[0046] Preferably, the health status analysis unit is used to obtain the energy storage battery aging rate based on the trained energy storage battery aging model and inverter aging model. and inverter aging rate And calculate the energy storage battery health score SoH b (t) and inverter health score R inv (t), and obtain the remaining usage time RUL of the energy storage battery bAnd the remaining use time RUL of the inverter inv , the energy storage battery health score SoH b (t) and inverter health score R inv (t) is calculated as follows:

[0047]

[0048] In the formula, SoH b (t 0 ) represents the initial time point t 0 The energy storage battery health score at the time, R inv (t 0 ) represents the initial time point t 0 The inverter health score at represents the aging rate of the energy storage battery at time point τ, represents the inverter aging rate at time point τ, τ∈[t 0 , t];

[0049] The remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv The method of obtaining is:

[0050]

[0051] In the formula, SoH b，critical Represents the critical value of the energy storage battery health score, R inv，max Indicates the maximum value of the inverter health score.

[0052] Preferably, the intelligent repair module includes a path planning unit and a fault repair unit;

[0053] The path planning unit is used to obtain the optimal fault repair path P using a path planning algorithm based on the microgrid topology diagram. opt , the optimal fault repair path P opt The specific acquisition method is as follows:

[0054] Based on the health equipment monitoring module and the microgrid data set S, the operating data of the energy storage battery and inverter are obtained, including the remaining service life RUL of the energy storage battery. b , Remaining inverter usage time RUL inv , energy storage battery load rate L SOC , inverter load factor L inv , Energy Storage Battery Health Score SoH b (t) and inverter health score R inv (t);

[0055] The microgrid topology diagram is converted into a microgrid weighted graph F, and the microgrid weighted graph F is specifically expressed as follows:

[0056] F = (N, E);

[0057] Where N represents the nodes in the microgrid, including energy storage batteries, inverters and load nodes, and E represents the edges, including the connection lines between nodes. The load nodes refer to the power consumption ends, that is, the power consumption equipment and power consumption areas that need power supply;

[0058] According to the microgrid weighted graph F, for each energy storage battery and inverter, a preset energy storage battery usage time health threshold T(RUL b ) min and inverter usage time health threshold T(RUL inv ) min , and based on the remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv , combined with the preset energy storage battery usage time health threshold T(RUL b ) min and inverter usage time health threshold T(RUL inv ) min , screen the energy storage battery and inverter, and retain the RUL b >T(RUL b ) min and RUL inv >T(RUL inv ) min Energy storage batteries and inverters as backup equipment;

[0059] The weight of each edge is calculated based on the operating data of the energy storage battery and inverter. The specific calculation process includes:

[0060]

[0061] Where RUL b，i Indicates the remaining usage time of the energy storage battery of node i, RUL inv，i represents the remaining usage time of the inverter at node i, L SOC，i represents the energy storage battery load rate of node i, L inv，i represents the inverter load rate of node i, C remaining，ij represents the remaining transmission capacity of the line between node i and node j, β, γ, δ, ε and ∈ represent weight coefficients;

[0062] The Dijkstra shortest path algorithm is used to search for paths with the microgrid weighted graph G as input, and the path search goal is set to minimize the total path weight. The path with the smallest total path weight is selected as the optimal fault repair path P.opt , the specific form of minimizing the total weight of the path is:

[0063]

[0064] Where W path represents the total weight of path P, e ij represents the edge from node i to node j.

[0065] Preferably, the fault repair unit is used to repair the fault according to the optimal fault repair path P opt , perform fault repair operations, where the optimal fault repair path P opt Including the starting equipment, passing lines and terminal loads, the fault repair operation includes initializing the repair task, enabling backup equipment and switching lines, and real-time monitoring of microgrid-related data after the fault repair operation.

[0066] A data mining method for a microgrid control system comprises the following steps:

[0067] Step 1: Collect microgrid related data in real time, and construct a microgrid data set S based on the collected microgrid related data;

[0068] Step 2: Preprocess the microgrid data set S, and perform summary calculation based on the preprocessed microgrid data set S to obtain the voltage stability factor S. v (t), frequency deviation factor S f (t) and total harmonic distortion THD, and according to the voltage stability factor S v (t), frequency deviation factor S f (t) and total harmonic distortion THD to obtain the comprehensive power quality score Q elec , and preset the power score threshold Q normal , the power score threshold Q normal Power quality comprehensive score Q elec Conduct comparative analysis to assess whether the microgrid has a fault. elec >Q normal , it indicates that the microgrid has a fault, and the fault location module is triggered;

[0069] Step 3: Set up several data monitoring points in the microgrid, collect voltage waveform data at the data monitoring points, determine the fault point location based on the traveling wave propagation principle, obtain the energy storage battery and inverter information of the line where the fault point is located, and generate the fault point device set G;

[0070] Step 4: Based on the microgrid data set S and the comprehensive power quality score Q elec, use linear regression algorithm to build energy storage battery aging model and inverter aging model to obtain energy storage battery aging rate and inverter aging rate And calculate the energy storage battery health score SoH b (t) and inverter health score R inv (t), based on the energy storage battery health score SoH b (t) and inverter health score R inv (t), perform comprehensive calculation to obtain the remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv ;

[0071] Step 5: Based on the fault point equipment set G, obtain the remaining service life RUL of the energy storage battery of the line where the fault point is located b And the remaining use time RUL of the inverter inv , perform device screening, obtain backup devices, and use Dijkstra's shortest path algorithm to obtain the optimal fault repair path P from the backup device starting point to the fault point opt , and based on the optimal fault repair path P opt Perform troubleshooting.

[0072] The present invention provides a data mining method and system for serving a microgrid control system, which has the following beneficial effects:

[0073] (1) Through the feature extraction and anomaly assessment module, the voltage stability factor S in the microgrid is collected and analyzed in real time v (t), frequency deviation factor S f (t) Total harmonic distortion rate THD, comprehensive calculation of power quality comprehensive score Q elec And set the power score threshold Q normal , can promptly detect the abnormal state of the microgrid. Compared with the traditional single power quality monitoring method, this method is based on multi-factor comprehensive evaluation, has higher accuracy and real-time performance, and helps to quickly identify abnormal power fluctuations, thereby improving the overall operation stability and power supply quality of the microgrid.

[0074] (2) Through the coordinated work of the fault location module and the equipment health monitoring module, the fault point in the microgrid can be quickly located, and based on the operating data of the energy storage battery and inverter, such as aging rate, health score and remaining usage time, an accurate assessment of the equipment status can be achieved. Compared with the traditional fault handling method, the system introduces equipment health monitoring to dynamically analyze the state changes of the equipment under fault conditions, avoiding secondary faults caused by ignoring the degree of equipment aging. At the same time, the health status information is combined to optimize the equipment usage strategy, effectively extending the service life of the energy storage battery and inverter.

[0075] (3) Through the intelligent repair module, the system can generate the optimal fault repair path P using the path planning algorithm based on the device information of the fault point, combined with the microgrid topology map, energy storage battery operation data, and inverter operation data. opt Compared with the traditional manual repair solution, the system takes into account the remaining use time RUL of the energy storage battery. b , Remaining inverter usage time RUL inv , energy storage battery load rate L SOC , inverter load factor L inv And the line transmission power C current On this basis, the selection of backup equipment and the line switching process are optimized. This method not only shortens the fault repair time, but also reduces the extra burden on backup equipment, improves the microgrid's ability to quickly recover when a fault occurs, and ensures power supply safety and efficient system operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 This is a block diagram of a data mining system serving a microgrid control system according to the present invention.

[0077] Figure 2 The present invention is a flow chart of a data mining method serving a microgrid control system. DETAILED DESCRIPTION

[0078] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0079] Example 1

[0080] See also Figure 1 , the present invention provides a data mining system serving a microgrid control system, including a data acquisition module, a feature extraction and anomaly assessment module, a fault location module, an equipment health monitoring module and an intelligent repair module;

[0081] The data acquisition module is used to collect microgrid related data in real time, and construct a microgrid data set S based on the collected microgrid related data;

[0082] The feature extraction and abnormality assessment module is used to perform preprocessing based on the microgrid data set S, and perform summary calculation based on the preprocessed microgrid data set S to obtain the voltage stability factor S v (t), frequency deviation factor Sf (t) and total harmonic distortion THD, and according to the voltage stability factor S v (t), frequency deviation factor S f (t) and total harmonic distortion THD to obtain the comprehensive power quality score Q elec , and preset the power score threshold Q normal , the power score threshold Q normal Power quality comprehensive score Q elec Conduct comparative analysis to assess whether the microgrid has a fault. elec >Q normal , it indicates that the microgrid has a fault, and the fault location module is triggered;

[0083] The fault location module is used to set a number of data monitoring points in the microgrid, collect voltage waveform data of the data monitoring points, determine the location of the fault point according to the principle of traveling wave propagation, and obtain the energy storage battery and inverter information of the line where the fault point is located, and generate a fault point device set G;

[0084] The equipment health monitoring module is used to monitor the power quality of the microgrid based on the microgrid data set S and the power quality comprehensive score Q elec , use linear regression algorithm to build energy storage battery aging model and inverter aging model to obtain energy storage battery aging rate and inverter aging rate And calculate the energy storage battery health score SoH b (t) and inverter health score R inv (t), based on the energy storage battery health score SoH b (t) and inverter health score R inv (t), perform comprehensive calculation to obtain the remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv ;

[0085] The intelligent repair module is used to obtain the remaining service life RUL of the energy storage battery of the line where the fault point is located according to the fault point equipment set G b And the remaining use time RUL of the inverter inv , perform device screening, obtain backup devices, and use Dijkstra's shortest path algorithm to obtain the optimal fault repair path P from the backup device starting point to the fault point opt , and based on the optimal fault repair path P opt Perform troubleshooting.

[0086] In the embodiment, the data acquisition module collects the power quality data, equipment operation data and environmental data of the microgrid in real time, constructs the microgrid data set S, and provides comprehensive and accurate data support for subsequent analysis. The feature extraction and abnormality assessment module can obtain the voltage stability factor S v (t), frequency deviation factor S f (t) and total harmonic distortion THD, calculate the comprehensive power quality score Q elec , accurately evaluate the power quality status of the microgrid, quickly identify abnormal conditions, the fault location module combines the principle of traveling wave propagation, quickly locates the fault point and obtains related equipment data, generates the fault point equipment set G, and shortens the fault response time. The equipment health monitoring module is based on the aging rate model of the energy storage battery and inverter, dynamically evaluates the equipment health status, predicts the remaining use time of the equipment, and provides a scientific basis for optimizing equipment management and maintenance. The intelligent repair module combines the microgrid topology map and equipment health status data, screens spare equipment, and uses the Dijkstra shortest path algorithm to obtain the optimal fault repair path P opt , ensuring that equipment life and line load capacity are given priority during the repair process, improving repair efficiency and stability. Overall, this system improves the reliability and efficiency of microgrid operation through the collaborative work of power quality anomaly detection, equipment health prediction and intelligent repair, reduces the risk of interruption caused by sudden equipment failures, reduces operation and maintenance costs, and provides important technical support for the intelligent operation and maintenance and sustainable development of microgrids.

[0087] Example 2

[0088] Please refer to Figure 1 Specifically: the data acquisition module is used to deploy an intelligent sensor group on the microgrid, collect microgrid related data in real time, and construct a microgrid data set S based on the collected microgrid related data, wherein the microgrid related data includes power quality data and equipment operation data;

[0089] The intelligent sensor group includes a voltage monitor, a current monitor, a frequency detector, a thermal resistance temperature sensor and a power measurement instrument;

[0090] The power quality data includes voltage V, current I and power frequency f;

[0091] The equipment operation data includes the battery state of charge SOC of the energy storage battery, the energy storage battery load rate L SOC , Energy storage battery operating temperature T b , inverter operating temperature T inv , inverter load factor L inv And the line transmission power C current ;

[0092] The voltage V is obtained by a voltage monitor;

[0093] The current I is obtained by a current monitor;

[0094] The power frequency f is obtained by a frequency detector;

[0095] The battery state of charge SOC of the energy storage battery and the energy storage battery load rate L SOC and the energy storage battery operating temperature T b Obtained through the battery management system BMS;

[0096] The inverter operating temperature T inv Obtained through thermal resistance temperature sensor;

[0097] The inverter load factor L inv and line transmission power C current Obtained through a power meter.

[0098] In the embodiment, by deploying a smart sensor group on the microgrid, the system can obtain power quality data in real time, such as voltage V, current I, power frequency f, and equipment operation data, such as the battery state of charge SOC of the energy storage battery, the energy storage battery load rate L SOC , Energy storage battery operating temperature T b , inverter operating temperature T inv , inverter load factor L inv And the line transmission power C current ,This precise collection and integration of multi-source data builds a comprehensive microgrid data set S, ,provides high-quality basic data support for subsequent data analysis and ,optimization scheduling. Compared with the traditional single parameter monitoring method, ,the intelligent sensor group can capture the dynamic changes in the operation of the ,microgrid in real time, efficiently and multi-dimensionally, ensuring that the system can ,respond to grid anomalies or equipment operation deviations in a timely manner. ,Through comprehensive data collection, the system can more accurately evaluate power quality ,problems and equipment health status, effectively improve the monitoring accuracy and ,operation stability of the microgrid, laying the foundation for the intelligent management and ,optimization of power supply, thereby reducing the ,occurrence rate of failures and maintenance costs, and enhancing the reliability and ,sustainability of the system.

[0099] Example 3

[0100] Please refer to Figure 1 ,Specifically: the feature extraction and anomaly assessment module includes a feature extraction unit and an assessment unit;

[0101] The feature extraction unit is used to perform preprocessing based on the microgrid data set S, and obtain a voltage stability factor S based on the preprocessed microgrid data set S. v (t), frequency deviation factor S f(t) and total harmonic distortion THD, wherein the preprocessing includes data cleaning, denoising and data standardization;

[0102] The voltage stability factor S v (t) The method of obtaining is:

[0103]

[0104] Where T represents the sampling time window, V(θ) represents the voltage at time point θ, V(θ+Δt) represents the voltage at time point θ+Δt, θ∈[tT,t], Δt represents the sampling time interval, S v The larger the value of (t), the more severe the voltage fluctuation is and the worse the stability of the power grid is.

[0105] The frequency deviation factor S f (t) The method of obtaining is:

[0106]

[0107] Where f(θ) represents the grid frequency at time θ, f nom Indicates the standard frequency value of the power grid, the frequency deviation factor S f The closer (t) is to zero, the more stable the grid frequency is;

[0108] The standard frequency value of the power grid f nom Obtained through the National Grid Operation Standard Table;

[0109] According to the power quality data in the microgrid data set S, the voltage V is decomposed in the frequency domain using fast Fourier transform, and the voltage waveform in the time domain is converted to the frequency domain to obtain the fundamental amplitude H 1 and harmonic amplitude H n , and according to the fundamental amplitude H 1 and harmonic amplitude H n , get the total harmonic distortion THD, where the harmonic amplitude H n The specific acquisition method is as follows:

[0110] H n =n·H 1 ;

[0111] In the formula, H 1 represents the fundamental amplitude, and n represents the harmonic order;

[0112] The fundamental amplitude H 1 Refers to the lowest frequency sinusoidal component in the voltage signal, that is, the nominal frequency of the power system;

[0113] The harmonic amplitude H nRefers to the sinusoidal component of the frequency component that is an integer multiple of the fundamental frequency. It is a waveform distortion component introduced by the use of nonlinear loads and power electronic equipment. The harmonic amplitude is an integer multiple of the fundamental amplitude.

[0114] The total harmonic distortion THD is obtained as follows:

[0115]

[0116] In the formula, H n Indicates harmonic amplitude, n≥2.

[0117] The evaluation unit is used to determine the voltage stability factor S v (t), frequency deviation factor S f (t) and total harmonic distortion THD, and then calculate and obtain the comprehensive power quality score Q elec , the comprehensive score of power quality Q elec The method of obtaining is:

[0118] Q elec =α 1 ·S v (t)+α 2 ·S f (t)+α 3 THD+C;

[0119] In the formula, α 1 , α 2 and α 3 Represents the voltage stability factor S v (t), frequency deviation factor S f (t) Weight coefficient of total harmonic distortion rate THD, C represents the first correction constant, where the value of the weight coefficient is set by the customer according to the actual situation, 0<α 1 <1, 0<α 2 <1, 0<α 3 <1,α 1 +α 2 +α 3 =1;

[0120] Preset power score threshold Q normal , and the power score threshold Q normal Power quality comprehensive score Q elec Conduct comparative analysis to assess whether the microgrid has a fault. elec ≤Q normal , it indicates that the microgrid has not failed and no intervention is required; if Q elec >Q normal , it indicates that a fault occurs in the microgrid. At this time, the fault location module is triggered to locate the fault of the microgrid.

[0121] In the embodiment, the microgrid data set S is preprocessed, and the voltage stability factor S is extracted according to the preprocessed microgrid data set S. v (t), frequency deviation factor S f (t) and total harmonic distortion THD, which can comprehensively reflect the dynamic characteristics of power quality. In addition, the fundamental amplitude H of the voltage V is extracted using fast Fourier transform. 1 and harmonic amplitude H n , to calculate the total harmonic distortion THD, which can effectively capture the harmonic distortion caused by nonlinear loads and power electronic equipment. The evaluation unit calculates the voltage stability factor S by summarizing v (t), frequency deviation factor S f (t) and total harmonic distortion THD to generate a comprehensive power quality score Q elec The power quality comprehensive score Q elec With the preset power score threshold Q normal By conducting comparative analysis, it is possible to quickly determine whether there is an abnormality in the microgrid. Compared with the traditional detection method that relies only on a single indicator, the multi-factor collaborative evaluation of the feature extraction and anomaly assessment module can improve the detection accuracy of power quality problems. At the same time, the unified scoring system simplifies the analysis process of complex power data. In the event of a microgrid failure, the fault location module is triggered in time to avoid the expansion of abnormal situations. Overall, this module not only improves the monitoring capability of the microgrid operating status, but also realizes the rapid identification and early warning of abnormal problems, providing key support for subsequent fault location and repair, and improving the stability and operation efficiency of the microgrid system.

[0122] Example 4

[0123] Please refer to Figure 1 , specifically: the fault location module is used to set a number of data monitoring points in the microgrid, and when a fault occurs in the microgrid, collect voltage waveform data of each data monitoring point in real time, wherein the voltage waveform data refers to the waveform change of the voltage signal;

[0124] According to the voltage waveform data obtained at each data monitoring point, the two data monitoring points A and B with the largest change in the amplitude of the voltage traveling wave signal are selected, and the time difference Δt of the voltage traveling wave signal reaching the data monitoring points A and B is obtained. AB , and according to the principle of traveling wave propagation, randomly select any data monitoring point from data monitoring points A and B, and calculate the distance d between any data monitoring point in data monitoring points A and B and the fault point. The specific calculation method is as follows:

[0125]

[0126] In the formula, vh represents the propagation speed of the traveling wave, Δt AB It represents the time difference between the arrival of the traveling wave at data monitoring point A and data monitoring point B;

[0127] According to the microgrid control system, the microgrid topology map is obtained, and the data monitoring point A and the data monitoring point B are marked in the microgrid topology map. Combined with the distance d between any data monitoring point A and B and the fault point, the line where the fault point is located is located in the microgrid topology map and the position of the fault point is marked. The energy storage battery data and inverter data of the line where the fault point is located are collected to generate the fault point device set G.

[0128] In the embodiment, through the real-time monitoring and precise analysis of the fault location module, the positioning efficiency and accuracy of the microgrid when a fault occurs can be effectively improved. First, the fault location module can quickly capture the propagation characteristics of the traveling wave signal when a fault occurs by setting multiple data monitoring points and collecting voltage waveform data, so as to select key monitoring points through amplitude changes, and calculate the distance between the fault point and the monitoring point based on the traveling wave propagation principle. Compared with the traditional manual troubleshooting and single-point monitoring methods, this method not only improves the speed of fault location, but also reduces the misjudgment rate. Secondly, by combining with the microgrid topology map, the fault point can be accurately mapped to the specific line location, and the operation data of related line equipment, such as energy storage batteries and inverters, can be further obtained to generate the fault point equipment set G. This multi-dimensional analysis method combining power waveform data and equipment operation information can comprehensively evaluate the impact range of the fault and provide accurate data support for subsequent health monitoring and repair path planning. Finally, the application of this module shortens the fault response time, improves the reliability and stability of the microgrid operation, and reduces the wider range of power outages and equipment damage caused by positioning delays.

[0129] Example 5

[0130] Please refer to Figure 1 ,Specifically: the equipment health monitoring module includes an equipment aging rate calculation unit and a health status analysis unit;

[0131] The device aging rate calculation unit is used to calculate the device aging rate of the fault point according to the fault point device set G, for the line where the fault point is located and the energy storage battery and inverter of the line where the fault point is located, according to the microgrid data set S and the comprehensive power quality score Q elec , a linear regression algorithm is used to construct an energy storage battery aging model and an inverter aging model. The energy storage battery aging model is specifically expressed as follows:

[0132]

[0133] In the formula, represents the aging rate of the energy storage battery at the current time point t, Qelec (t) represents the comprehensive score of power quality at the current time point t, T b (t) represents the operating temperature of the energy storage battery at the current time point t, SOC(t) represents the battery state of charge of the energy storage battery at the current time point t, k 1 , k 2 and k 3 represents the regression coefficient;

[0134] The inverter aging model is specifically expressed as follows:

[0135]

[0136] In the formula, Indicates the inverter aging rate at the current time point t, L inv (t) represents the inverter load rate at the current time point t, T inv (t) represents the inverter operating temperature at the current time point t, m 1 、m 2 and m 3 represents the regression coefficient;

[0137] Collect historical microgrid operation data and divide it into training set and test set. Use the training set to input the energy storage battery aging model and inverter aging model for training. Use the least squares method to fit the parameters of the energy storage battery aging model and inverter aging model to obtain the regression coefficient k. 1 , k 2 , k 3 、m 1 、m 2 and m 3 , and use the test set to verify the energy storage battery aging model and inverter aging model.

[0138] The health status analysis unit is used to obtain the energy storage battery aging rate based on the trained energy storage battery aging model and inverter aging model. and inverter aging rate And calculate the energy storage battery health score SoH b (t) and inverter health score R inv (t), and obtain the remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv , the energy storage battery health score SoH b (t) and inverter health score R inv (t) is calculated as follows:

[0139]

[0140] In the formula, SoHb (t 0 ) represents the initial time point t 0 The energy storage battery health score, R inv (t 0 ) represents the initial time point t 0 The inverter health score at represents the aging rate of the energy storage battery at time point τ, represents the inverter aging rate at time point τ, τ∈[t 0 , t];

[0141] The remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv The method of obtaining is:

[0142]

[0143] In the formula, SoH b，critical Represents the critical value of the energy storage battery health score, R inv，max Indicates the maximum value of the inverter health score.

[0144] The energy storage battery health score critical value SoH b，critical The maximum value of the inverter health score R is obtained from the manufacturer's standard data. inv，max Set to 1.

[0145] In the embodiment, the operation safety of the microgrid and the accuracy of equipment maintenance are improved through the collaborative work of the equipment aging rate calculation unit and the health status analysis unit. First, the aging model of the energy storage battery and the inverter is constructed through the linear regression algorithm, which can dynamically quantify the aging rate of the equipment under different operating conditions. This dynamic analysis method breaks through the limitations of traditional static evaluation, so that the changes in the health status of the equipment can be captured in real time. Secondly, through the health status analysis unit, the aging rate of the energy storage battery is used to calculate the aging rate of the energy storage battery. and inverter aging rate Calculating the Battery Health Score (SoH) b (t) and inverter health score R inv (t), and further calculate the remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv, which provides a reliable basis for the precise maintenance and scheduling of energy storage batteries and inverters. This dynamic monitoring method based on health scores and remaining life effectively avoids sudden failures caused by the deterioration of equipment health. In addition, the scheduling strategy of the microgrid is optimized in combination with equipment health data. In power fluctuation scenarios, equipment in good health can be given priority to reduce equipment load and extend equipment service life. In summary, this module can effectively reduce the operation and maintenance costs of microgrids and improve power supply stability through precise health monitoring and life assessment, and provide timely and scientific data support for fault repair. It is the key guarantee for efficient and safe operation of microgrids.

[0146] Example 6

[0147] Please refer to Figure 1 ,Specifically: the intelligent repair module includes a path planning unit and a fault repair unit;

[0148] The path planning unit is used to obtain the optimal fault repair path P using a path planning algorithm based on the microgrid topology diagram. opt , the optimal fault repair path P opt The specific acquisition method is as follows:

[0149] Based on the health equipment monitoring module and the microgrid data set S, the operating data of the energy storage battery and inverter are obtained, including the remaining service life RUL of the energy storage battery. b , Remaining inverter usage time RUL inv , energy storage battery load rate L SOC , inverter load factor L inv , Energy Storage Battery Health Score SoH b (t) and inverter health score R inv (t);

[0150] The microgrid topology diagram is converted into a microgrid weighted graph F, and the microgrid weighted graph F is specifically expressed as follows:

[0151] F = (N, E);

[0152] Where N represents the nodes in the microgrid, including energy storage batteries, inverters and load nodes, and E represents the edges, including the connection lines between nodes. The load nodes refer to the power consumption ends, that is, the power consumption equipment and power consumption areas that need power supply;

[0153] According to the microgrid weighted graph F, for each energy storage battery and inverter, a preset energy storage battery usage time health threshold T(RUL b ) min and inverter usage time health threshold T(RUL inv ) min , and based on the remaining usage time RUL of the energy storage batteryb And the remaining use time RUL of the inverter inv , combined with the preset energy storage battery usage time health threshold T(RUL b ) min and inverter usage time health threshold T(RUL inv ) min , screen the energy storage battery and inverter, and retain the RUL b >T(RUL b ) min and RUL inv >T(RUL inv ) min Energy storage batteries and inverters as backup equipment;

[0154] The weight of each edge is calculated based on the operating data of the energy storage battery and inverter. The specific calculation process includes:

[0155]

[0156] Where RUL b，i Indicates the remaining usage time of the energy storage battery of node i, RUL inv，i represents the remaining usage time of the inverter at node i, L SOC，i represents the energy storage battery load rate of node i, L inv，i represents the inverter load rate of node i, C remaining，ij represents the remaining transmission capacity of the line between node i and node j, where the remaining transmission power C remaining，ij By using the maximum transmission power of the line C current，max Subtract the line transmission power C current Obtain, β, γ, δ, ε and ∈ represent weight coefficients, and the data of weight coefficients are set by the customer according to the specific situation, 0<β<1, 0<γ<1, 0<δ<1, 0<ε<1, 0<∈<1, β+γ+δ+ε+∈=1;

[0157] The Dijkstra shortest path algorithm is used to search for paths with the microgrid weighted graph G as input, and the path search goal is set to minimize the total path weight. The path with the smallest total path weight is selected as the optimal fault repair path P. opt , the specific form of minimizing the total weight of the path is:

[0158]

[0159] Where W path represents the total weight of path P, e ij represents the edge from node i to node j.

[0160] The fault repair unit is used to repair the fault according to the optimal fault repair path P opt , perform fault repair operations, where the optimal fault repair path P opt Including the starting equipment, passing lines and terminal loads, the fault repair operation includes initializing the repair task, enabling backup equipment and switching lines, and real-time monitoring of microgrid-related data after the fault repair operation.

[0161] In the embodiment, through the collaborative work of the path planning unit and the fault repair unit, the present solution can quickly generate the optimal fault repair path P after a microgrid fault occurs. opt , and efficiently perform repair tasks, effectively improving the operational reliability and repair efficiency of the microgrid. The path planning unit combines the health equipment monitoring module and the microgrid data set S to obtain the remaining use time RUL of the energy storage battery b , Remaining inverter usage time RUL inv , energy storage battery load rate L SOC , inverter load factor L inv , Energy Storage Battery Health Score SoH b (t) and inverter health score R inv (t), and generate a microgrid weighted graph F based on the microgrid topology graph. By selecting energy storage devices and inverters in good health as backup devices, the weight of each edge is calculated, and factors such as device health status, load rate, and line remaining transmission capacity are comprehensively considered to ensure the scientificity and efficiency of the repair path selection. Finally, the Dijkstra shortest path algorithm is used to minimize the total path weight and select the optimal fault repair path P from the backup device to the fault point. opt , the fault repair unit is based on the optimal fault repair path P opt Perform repair operations and monitor the fault repair effect in real time to ensure that the system returns to normal operation. Compared with traditional fault repair solutions, this solution dynamically optimizes the repair path through a path planning algorithm, fully considers the health status of the equipment and the line load, avoids repeated faults, improves repair efficiency, reduces the additional burden on standby equipment, and reduces the impact of fault repair on the overall operation of the power grid, ensuring the safety and stability of the microgrid.

[0162] Please refer to Figure 2 ,Specifically: A data mining method serving a microgrid control system includes the following steps,

[0163] Step 1: Collect microgrid related data in real time, and construct a microgrid data set S based on the collected microgrid related data;

[0164] Step 2: Preprocess the microgrid data set S, and perform summary calculation based on the preprocessed microgrid data set S to obtain the voltage stability factor S. v (t), frequency deviation factor S f (t) and total harmonic distortion THD, and according to the voltage stability factor S v (t), frequency deviation factor S f (t) and total harmonic distortion THD to obtain the comprehensive power quality score Q elec , and preset the power score threshold Q normal , the power score threshold Q normal Power quality comprehensive score Q elec Conduct comparative analysis to assess whether the microgrid has a fault. elec >Q normal , it indicates that the microgrid has a fault, and the fault location module is triggered;

[0165] Step 3: Set up several data monitoring points in the microgrid, collect voltage waveform data at the data monitoring points, determine the fault point location based on the traveling wave propagation principle, obtain the energy storage battery and inverter information of the line where the fault point is located, and generate the fault point device set G;

[0166] Step 4: Based on the microgrid data set S and the comprehensive power quality score Q elec , use linear regression algorithm to build energy storage battery aging model and inverter aging model to obtain energy storage battery aging rate and inverter aging rate And calculate the energy storage battery health score SoH b (t) and inverter health score R inv (t), based on the energy storage battery health score SoH b (t) and inverter health score R inv (t), perform comprehensive calculation to obtain the remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv ;

[0167] Step 5: Based on the fault point equipment set G, obtain the remaining service life RUL of the energy storage battery of the line where the fault point is located b And the remaining use time RUL of the inverter inv , perform device screening, obtain backup devices, and use Dijkstra's shortest path algorithm to obtain the optimal fault repair path P from the backup device starting point to the fault point opt , and based on the optimal fault repair path P opt Perform troubleshooting.

[0168] In the embodiment, by collecting microgrid related data in real time and constructing a microgrid data set S, a comprehensive diagnosis and optimal scheduling of power quality and equipment health status is achieved, and in the power quality assessment, the voltage stability factor S is used v (t), frequency deviation factor S f (t) and total harmonic distortion THD, conduct multi-factor joint analysis to obtain the comprehensive power quality score Q elec , accurately detect abnormal conditions of microgrids, use the principle of traveling wave propagation to quickly determine the location of fault points for fault location problems, and obtain equipment information at fault points in combination with the microgrid topology map, laying the foundation for subsequent health status analysis and path planning. Through the aging model and health status prediction of energy storage batteries and inverters, dynamically obtain the remaining service life RUL of energy storage batteries b And the remaining use time RUL of the inverter inv , and in fault repair, use Dijkstra's shortest path algorithm to generate the optimal fault repair path P opt ,This method not only improves the efficiency of microgrid fault detection and ,repair, but also effectively prolongs the equipment life, reduces the equipment maintenance ,cost, and improves the stability and power supply reliability of the microgrid ,and meets the needs of efficient and reliable operation of modern ,smart grid.

[0169] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A data mining system serving a microgrid control system, characterized in that: It includes data acquisition module, feature extraction and anomaly assessment module, fault location module, equipment health monitoring module and intelligent repair module; The data acquisition module is used to collect microgrid related data in real time, and construct a microgrid data set S based on the collected microgrid related data; The feature extraction and abnormality assessment module is used to perform preprocessing based on the microgrid data set S, and perform summary calculation based on the preprocessed microgrid data set S to obtain the voltage stability factor S v (t), frequency deviation factor S f (t) and total harmonic distortion THD, and according to the voltage stability factor S v (t), frequency deviation factor S f (t) and total harmonic distortion THD to obtain the comprehensive power quality score Q elec , and preset the power score threshold Q normal , the power score threshold Q normal Power quality comprehensive score Q elec Conduct comparative analysis to assess whether the microgrid has a fault. elec >Q normal , it indicates that the microgrid has a fault, and the fault location module is triggered; The fault location module is used to set a number of data monitoring points in the microgrid, collect voltage waveform data of the data monitoring points, determine the location of the fault point according to the principle of traveling wave propagation, and obtain the energy storage battery and inverter information of the line where the fault point is located, and generate a fault point device set G; The equipment health monitoring module is used to monitor the power quality of the microgrid based on the microgrid data set S and the power quality comprehensive score Q elec , use linear regression algorithm to build energy storage battery aging model and inverter aging model to obtain energy storage battery aging rate and inverter aging rate And calculate the energy storage battery health score SoH b (t) and inverter health score R inv (t), based on the energy storage battery health score SoH b (t) and inverter health score R inv (t), perform comprehensive calculation to obtain the remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv ; The intelligent repair module is used to obtain the remaining service life RUL of the energy storage battery of the line where the fault point is located according to the fault point equipment set G b And the remaining use time RUL of the inverter inv , perform device screening, obtain backup devices, and use Dijkstra's shortest path algorithm to obtain the optimal fault repair path P from the backup device starting point to the fault point opt , and based on the optimal fault repair path P opt Perform troubleshooting.

2. A data mining system serving a microgrid control system according to claim 1, characterized in that: The data acquisition module is used to deploy an intelligent sensor group on the microgrid, collect microgrid related data in real time, and construct a microgrid data set S based on the collected microgrid related data, wherein the microgrid related data includes power quality data and equipment operation data; The intelligent sensor group includes a voltage monitor, a current monitor, a frequency detector, a thermal resistance temperature sensor and a power measuring instrument; The power quality data includes voltage V, current I and power frequency f; The equipment operation data includes the battery state of charge SOC of the energy storage battery, the energy storage battery load rate L SOC , Energy storage battery operating temperature T b , inverter operating temperature T inv , inverter load factor L inv And the line transmission power C current .

3. A data mining system serving a microgrid control system according to claim 2, characterized in that: The feature extraction and anomaly assessment module includes a feature extraction unit and an assessment unit; The feature extraction unit is used to perform preprocessing based on the microgrid data set S, and obtain a voltage stability factor S based on the preprocessed microgrid data set S. v (t), frequency deviation factor S f (t) and total harmonic distortion THD, wherein the preprocessing includes data cleaning, denoising and data standardization; The voltage stability factor S v (t) The method of obtaining is: Where T represents the sampling time window, V(θ) represents the voltage at time point θ, V(θ+Δt) represents the voltage at time point θ+Δt, θ∈[tT, t], Δt represents the sampling time interval; The frequency deviation factor S f (t) The method of obtaining is: Where f(θ) represents the grid frequency at time θ, f nom Indicates the standard frequency value of the power grid; According to the power quality data in the microgrid data set S, the voltage V is decomposed in the frequency domain using fast Fourier transform, and the voltage waveform in the time domain is converted to the frequency domain to obtain the fundamental amplitude H1 and the harmonic amplitude H n , and based on the fundamental amplitude H1 and harmonic amplitude H n , get the total harmonic distortion THD, where the harmonic amplitude H n The specific acquisition method is as follows: H n =n·H1; In the formula, H1 represents the fundamental amplitude, and n represents the harmonic order; The total harmonic distortion THD is obtained as follows: In the formula, H n Indicates harmonic amplitude, n≥2.

4. A data mining system serving a microgrid control system according to claim 3, characterized in that: The evaluation unit is used to determine the voltage stability factor S v (t), frequency deviation factor S f (t) and total harmonic distortion THD, and then calculate and obtain the comprehensive power quality score Q elec , the comprehensive score of power quality Q elec The method of obtaining is: Q elec =α1·S v (t)+α2·S f (t)+α3·THD+C; Where α1, α2 and α3 represent the voltage stability factor S v (t), frequency deviation factor S f (t) weight coefficient of total harmonic distortion THD, C represents the first correction constant; Preset power score threshold Q normal , and the power score threshold Q normal Power quality comprehensive score Q elec Conduct comparative analysis to assess whether the microgrid has a fault. elec ≤Q normal , it indicates that the microgrid has not failed and no intervention is required; if Q elec >Q normal , it indicates that a fault occurs in the microgrid. At this time, the fault location module is triggered to locate the fault of the microgrid.

5. A data mining system serving a microgrid control system according to claim 4, characterized in that: The fault location module is used to set a number of data monitoring points in the microgrid, and when a fault occurs in the microgrid, collect voltage waveform data of each data monitoring point in real time, wherein the voltage waveform data refers to the waveform change of the voltage signal; According to the voltage waveform data obtained at each data monitoring point, the two data monitoring points A and B with the largest change in the amplitude of the voltage traveling wave signal are selected, and the time difference Δt of the voltage traveling wave signal reaching the data monitoring points A and B is obtained. AB , and according to the principle of traveling wave propagation, randomly select any data monitoring point from data monitoring points A and B, and calculate the distance d between any data monitoring point in data monitoring points A and B and the fault point. The specific calculation method is as follows: In the formula, v h represents the propagation speed of the traveling wave, Δt AB It represents the time difference between the arrival of the traveling wave at data monitoring point A and data monitoring point B; According to the microgrid control system, the microgrid topology map is obtained, and the data monitoring point A and the data monitoring point B are marked in the microgrid topology map. Combined with the distance d between any data monitoring point A and B and the fault point, the line where the fault point is located is located in the microgrid topology map and the position of the fault point is marked. The energy storage battery data and inverter information of the line where the fault point is located are collected to generate the fault point device set G.

6. A data mining system serving a microgrid control system according to claim 5, characterized in that: The equipment health monitoring module includes an equipment aging rate calculation unit and a health status analysis unit; The device aging rate calculation unit is used to calculate the device aging rate of the fault point according to the fault point device set G, for the line where the fault point is located and the energy storage battery and inverter of the line where the fault point is located, according to the microgrid data set S and the comprehensive power quality score Q elec , a linear regression algorithm is used to construct an energy storage battery aging model and an inverter aging model. The energy storage battery aging model is specifically expressed as follows: In the formula, represents the aging rate of the energy storage battery at the current time point t, Q elec (t) represents the comprehensive score of power quality at the current time point t, T b (t) represents the operating temperature of the energy storage battery at the current time point t, SOC(t) represents the battery state of charge of the energy storage battery at the current time point t, k1, k2 and k3 represent regression coefficients; The inverter aging model is specifically expressed as follows: In the formula, Indicates the inverter aging rate at the current time point t, L inv (t) represents the inverter load rate at the current time point t, T inv (t) represents the inverter operating temperature at the current time point t, m1, m2 and m3 represent regression coefficients; Historical microgrid operation data are collected and divided into training set and test set. The training set is used to input the energy storage battery aging model and the inverter aging model for training. The least squares method is used to fit the parameters of the energy storage battery aging model and the inverter aging model to obtain the regression coefficients k1, k2, k3, m1, m2 and m3. The test set is used to verify the energy storage battery aging model and the inverter aging model.

7. A data mining system serving a microgrid control system according to claim 6, characterized in that: The health status analysis unit is used to obtain the energy storage battery aging rate based on the trained energy storage battery aging model and inverter aging model. and inverter aging rate And calculate the energy storage battery health score SoH b (t) and inverter health score R inv (t), and obtain the remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv , the energy storage battery health score SoH b (t) and inverter health score R inv (t) is calculated as follows: In the formula, SoH b (t0) represents the energy storage battery health score at the initial time point t0, R inv (t0) represents the inverter health score at the initial time point t0, represents the aging rate of the energy storage battery at time point τ, represents the inverter aging rate at time point τ, τ∈[t0,t]; The remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv The method of obtaining is: In the formula, SoH b，critical Represents the critical value of the energy storage battery health score, R inv，max Indicates the maximum value of the inverter health score.

8. A data mining system serving a microgrid control system according to claim 7, characterized in that: The intelligent repair module includes a path planning unit and a fault repair unit; The path planning unit is used to obtain the optimal fault repair path P using a path planning algorithm based on the microgrid topology diagram. opt , the optimal fault repair path P opt The specific acquisition method is as follows: Based on the health equipment monitoring module and the microgrid data set S, the operating data of the energy storage battery and inverter are obtained, including the remaining service life RUL of the energy storage battery. b , Inverter remaining use time RUL inv , energy storage battery load rate L SOC , inverter load factor L inv , Energy Storage Battery Health Score SoH b (t) and inverter health score R inv (t); The microgrid topology diagram is converted into a microgrid weighted graph F, and the microgrid weighted graph F is specifically expressed as follows: F = (N, E); Where N represents the nodes in the microgrid, including energy storage batteries, inverters and load nodes, and E represents the edges, including the connection lines between nodes. The load nodes refer to the power consumption ends, that is, the power consumption equipment and power consumption areas that need power supply; According to the microgrid weighted graph F, for each energy storage battery and inverter, a preset energy storage battery usage time health threshold T(RUL b ) min and inverter usage time health threshold T(RUL inv ) min , and based on the remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv , combined with the preset energy storage battery usage time health threshold T(RUL b ) min and inverter usage time health threshold T(RUL inv ) min , screen the energy storage battery and inverter, and retain the RUL b >T(RUL b ) min and RUL inv >T(RUL inv ) min Energy storage batteries and inverters as backup equipment; The weight of each edge is calculated based on the operating data of the energy storage battery and inverter. The specific calculation process includes: Where RUL b，i Indicates the remaining usage time of the energy storage battery of node i, RUL inv，i represents the remaining usage time of the inverter at node i, L SOC，i represents the energy storage battery load rate of node i, L inv，i represents the inverter load rate of node i, C remaining，ij represents the remaining transmission capacity of the line between node i and node j, β, γ, δ, ε and ∈ represent weight coefficients; The Dijkstra shortest path algorithm is used to search for paths with the microgrid weighted graph G as input, and the path search goal is set to minimize the total path weight. The path with the smallest total path weight is selected as the optimal fault repair path P. opt , the specific form of minimizing the total weight of the path is: Where W path represents the total weight of path P, e ij represents the edge from node i to node j.

9. A data mining system serving a microgrid control system according to claim 8, characterized in that: The fault repair unit is used to repair the fault according to the optimal fault repair path P opt , perform fault repair operations, where the optimal fault repair path P opt Including the starting equipment, passing lines and terminal loads, the fault repair operation includes initializing the repair task, enabling backup equipment and switching lines, and real-time monitoring of microgrid-related data after the fault repair operation.

10. A data mining method serving a microgrid control system, used to implement a data mining system serving a microgrid control system as claimed in any one of claims 1 to 9, characterized in that: The following steps are included: Step 1: Collect microgrid related data in real time, and construct a microgrid data set S based on the collected microgrid related data; Step 2: Preprocess the microgrid data set S, and perform summary calculation based on the preprocessed microgrid data set S to obtain the voltage stability factor S. v (t), frequency deviation factor S f (t) and total harmonic distortion THD, and according to the voltage stability factor S v (t), frequency deviation factor S f (t) and total harmonic distortion THD to obtain the comprehensive power quality score Q elec , and preset the power score threshold Q normal , the power score threshold Q normal Power quality comprehensive score Q elec Conduct comparative analysis to assess whether the microgrid has a fault. elec >Q normal , it indicates that the microgrid has a fault, and the fault location module is triggered; Step 3: Set up several data monitoring points in the microgrid, collect voltage waveform data at the data monitoring points, determine the fault point location based on the traveling wave propagation principle, obtain the energy storage battery and inverter information of the line where the fault point is located, and generate the fault point device set G; Step 4: Based on the microgrid data set S and the comprehensive power quality score Q elec , use linear regression algorithm to build energy storage battery aging model and inverter aging model to obtain energy storage battery aging rate and inverter aging rate And calculate the energy storage battery health score SoH b (t) and inverter health score R inv (t), based on the energy storage battery health score SoH b (t) and inverter health score R inv (t), perform comprehensive calculation to obtain the remaining usage time RUL of the energy storage battery b And the remaining use time RUL of the inverter inv ; Step 5: Based on the fault point equipment set G, obtain the remaining service life RUL of the energy storage battery of the line where the fault point is located b And the remaining use time RUL of the inverter inv , perform device screening, obtain backup devices, and use Dijkstra's shortest path algorithm to obtain the optimal fault repair path P from the backup device starting point to the fault point opt , and based on the optimal fault repair path P opt Perform troubleshooting.

Images