Micro-distribution collaborative low-voltage diagnosis method and system based on digital twinning and association mining

Abstract

The invention discloses a distribution and micro collaborative low-voltage diagnosis method and system based on digital twinborn and association mining. The method comprises the following steps: constructing a digital twinborn model of a distribution and micro collaborative power distribution network; analyzing low-voltage spatial-temporal distribution characteristics, screening a governance area, simulating the governance area by using a digital twinborn model, selecting a corresponding governance strategy to adjust the digital twinborn model for simulation again after mining a low-voltage cause, and verifying a governance effect of the selected governance strategy based on a simulation result; and for the governance strategy with the governance effect meeting the requirement, calculating a comprehensive score according to the corresponding simulation result, and taking the governance strategy with the highest comprehensive score as a low-voltage governance strategy. According to the method, the distribution micro-grid collaborative digital twin is constructed, diagnosis and treatment of the low-voltage problem are carried out based on the digital twin and association mining, and the voltage treatment level of the distribution micro-grid is improved.

Description

Technical Field

[0001] The present invention relates to the field of distribution automation, and in particular to a distribution and micro-distribution collaborative low voltage diagnosis method and system based on digital twins and association mining. Background Art

[0002] With the continuous acceleration of power construction, the scale of the power grid is expanding, and the distribution network structure is becoming increasingly complex. Currently, all cities have deployed a new generation of distribution automation master station systems, enabling effective monitoring and control of distribution network operations. With an increasing number of access terminals and automated lines, which are continuously expanding and extending to the low-voltage side, the transparency and observability of the power grid are increasing.

[0003] Digital twins are software-defined technologies that combine perception, computing, modeling, and other information technologies to describe, diagnose, predict, and make decisions about physical spaces. This allows for interactive mapping between physical and digital virtual spaces. Digital twins encompass numerous technical areas, including IoT perception, remote control, data integration, business modeling, and model interoperability. Using digital twins to simulate the real world of physical space significantly reduces trial-and-error costs and improves digital management and operations.

[0004] How to achieve low-voltage management in distribution networks based on digital twins is an issue worthy of study. Summary of the Invention

[0005] The technical problem to be solved by the present invention is: In response to the technical problems existing in the prior art, the present invention provides a distribution micro-cooperative low voltage diagnosis method and system based on digital twins and association mining, which conducts spatiotemporal distribution feature analysis of low voltage in distribution micro-grids based on digital twins, mines the correlation between the causes of low voltage and time and space, identifies the causes of low voltage, and guides the diagnosis and control of low voltage in distribution micro-grids.

[0006] In order to solve the above technical problems, the technical solution proposed by the present invention is:

[0007] A collaborative low voltage diagnosis method for microcontrollers based on digital twins and association mining includes the following steps:

[0008] Obtain physical and environmental data of the target area;

[0009] Build a digital twin model of the micro-distribution and distribution network;

[0010] Obtain historical low voltage data, analyze the spatiotemporal distribution characteristics of low voltage, and screen diagnosis and treatment areas. Use digital twin models to simulate the diagnosis and treatment areas, and use association analysis algorithms to explore the causes of low voltage.

[0011] According to the causes of low voltage, the corresponding treatment strategy is selected to adjust the digital twin model. The adjusted digital twin model is used for simulation to obtain simulation results. The simulation results are compared with historical low voltage data to verify the treatment effect of each selected treatment strategy.

[0012] For the control strategies whose control effects meet the requirements, the comprehensive scores are calculated according to the corresponding simulation results, and the control strategy with the highest comprehensive score is used as the low voltage control strategy.

[0013] Furthermore, the physical data includes one or more of current, voltage, active power, reactive power data, and opening and closing status of switches and circuit breakers, and the environmental data includes one or more of weather, temperature, humidity, and altitude.

[0014] Furthermore, the steps of constructing a digital twin model of the micro-distribution network include:

[0015] Based on physical data, the distribution network and microgrid are divided by boundaries, the distribution network, microgrid, and the distribution-microgrid boundary are identified, and digital twin entities of the distribution network, microgrid, and distribution-microgrid boundary are established. The digital twin entity of the operating environment is also established based on environmental data, and data measurement mapping relationships are also established;

[0016] Based on the digital twin entities and data measurement mapping relationships, digital twin models of the distribution network, microgrid, distribution-microgrid boundary, and operating environment are constructed;

[0017] According to the data measurement mapping relationship of the digital twin, the data of the corresponding digital twin entity is collected at a specified period and mapped to the digital twin model;

[0018] Carry out digital twin applications and establish a digital twin evaluation index system, and iteratively optimize the digital twin model based on the evaluation results.

[0019] Furthermore, the digital twin application includes:

[0020] Topology verification: Verify the correctness of switch status and topology association based on line power and bus voltage;

[0021] Bad data identification: Identify bad data in measurement data based on collaborative state estimation between the microcontroller and the distribution system;

[0022] Correlation analysis: analyzing the correlation between events;

[0023] Digital simulation: Based on weather data, the load of the distribution microgrid is predicted, the distribution microgrid collaborative power flow calculation is carried out, and the operating status of the distribution microgrid under different meteorological conditions and switching conditions is analyzed.

[0024] Furthermore, when analyzing the spatiotemporal distribution characteristics of low voltage and screening the diagnosis and treatment areas, the following are included:

[0025] According to the historical low voltage data of regions, lines, substations and microgrids, the low voltage characteristic time series of regions, lines, substations and microgrids are constructed, the correlation of the low voltage characteristic time series of each region, line, substation and microgrid is calculated, and the areas with strong correlation or the areas where strongly correlated lines, substations and microgrids are located are selected as diagnosis and treatment areas.

[0026] Further investigation into the causes of low voltage includes:

[0027] The simulation results are obtained and a low-voltage event list is established. The Apriori algorithm is used to mine the low-voltage event list to obtain a typical low-voltage event set. The corresponding low-voltage causes are analyzed for the events in the typical low-voltage event set.

[0028] Furthermore, when calculating the comprehensive score according to the corresponding simulation results, it includes:

[0029] Match the parameter adjustment strategy to the case library and obtain the corresponding governance costs as economic indicators;

[0030] Calculate the average voltage and average three-phase imbalance according to the simulation results, perform weighted summation on the average voltage and average three-phase imbalance to obtain the performance index;

[0031] The economic indicators and performance indicators are weighted and summed to obtain a comprehensive score.

[0032] The present invention also proposes a micro-device collaborative low-voltage diagnosis system based on digital twins and association mining, comprising a microprocessor and a computer-readable storage medium connected to each other, wherein the microprocessor is programmed or configured to execute any one of the micro-device collaborative low-voltage diagnosis methods based on digital twins and association mining.

[0033] The present invention also proposes a computer-readable storage medium, which stores a computer program. The computer program is executed by a microprocessor to implement any step of the micro-cooperative low-voltage diagnosis method based on digital twins and association mining.

[0034] The present invention also proposes a computer program product, which includes a computer program, and the computer program is executed by a microprocessor to implement any step of the micro-cooperative low-voltage diagnosis method based on digital twins and association mining.

[0035] Compared with the prior art, the advantages of the present invention are:

[0036] The present invention takes into account the spatiotemporal correlation characteristics of low voltage, mines the common characteristics of low voltage based on the spatiotemporal correlation characteristics, identifies the causes of occurrence, and verifies the economy and effectiveness of the governance strategy based on data twin simulation, thereby realizing intelligent auxiliary decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 Flowchart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The present invention will be further described below in conjunction with the accompanying drawings and specific preferred embodiments, but the scope of protection of the present invention is not limited thereby.

[0039] Example 1

[0040] This embodiment proposes a method for diagnosing low voltage in the power distribution network based on digital twins and association mining. It makes full use of data such as physical models, sensor updates, and operation history to map and construct a digital twin of the power distribution network. Based on the digital twin, it conducts closed-loop management of low voltage spatiotemporal distribution feature analysis, cause mining, strategy generation, and effectiveness evaluation, thereby improving the level of low voltage management in the power distribution network. Figure 1 As shown, the following steps are included:

[0041] S1) obtaining physical data and environmental data of the target area;

[0042] S2) Constructing a digital twin model of the micro-distribution network;

[0043] S3) Obtain historical low voltage data, analyze the spatiotemporal distribution characteristics of low voltage, select diagnosis and treatment areas, simulate the diagnosis and treatment areas using a digital twin model, and use association analysis algorithms to explore the causes of low voltage;

[0044] S4) Selecting corresponding control strategies according to the causes of low voltage to adjust the digital twin model, and using the adjusted digital twin model to perform simulations, obtain simulation results, and compare the simulation results with historical low voltage data to verify the control effect of each selected control strategy;

[0045] S5) For the control strategies whose control effects meet the requirements, a comprehensive score is calculated according to the corresponding simulation results, and the control strategy with the highest comprehensive score is used as the low voltage control strategy.

[0046] The following is a detailed description of each step.

[0047] In this embodiment, through step S1, the distribution and micro-grid area to be digitally twinned is selected, and the real-time physical data and corresponding environmental data of the network topology model of the area to be operated are extracted to construct a digital twin model based on the actual physical area mapping of the distribution and micro-grid collaborative power grid.

[0048] In this embodiment, the physical data includes real-time physical data such as the current, voltage, active power, reactive power data of the distribution area mapped by the electricity collection system and the SCADA system, as well as the opening and closing status of switches and circuit breakers. The environmental data includes the weather, temperature, humidity, altitude, etc. of the operating area.

[0049] In this embodiment, in step S2, a digital twin model is constructed, and simulation calculations are performed based on the data twin to generate distribution microgrid simulation results, which are compared with actual operation data to iteratively optimize the digital twin model. The specific steps include:

[0050] S21) Map digital twin entities. Construct digital twin entities of the distribution network, microgrid, distribution-microgrid boundary, and operating environment. Divide the distribution network and microgrid by their boundaries, identify the distribution network, microgrid, and distribution-microgrid boundary, and establish digital twin entities of the distribution network, microgrid, distribution-microgrid boundary, and operating environment, and establish data measurement mapping relationships.

[0051] S22) Construct a digital twin model. Based on the digital twin entities and data measurement mapping relationships, construct digital twin models of the distribution network, microgrid, distribution-microgrid boundary, and operating environment. The distribution network, microgrid, and distribution-microgrid boundary are modeled as physical entities. The digital twin of the operating environment is constructed as a three-dimensional data model of time, space, and environment.

[0052] S23) Collecting digital twin data in association: According to the digital twin mapping relationship, collect data of the digital twin entity at a certain period and map it to the digital twin model.

[0053] S24) Develop digital twin applications. Specifically, these include topology verification, bad data identification, correlation analysis, and digital deduction. Among them:

[0054] Topology verification verifies the correctness of switch status and topology association based on line power and bus voltage;

[0055] Bad data identification is based on the collaborative state estimation of the micro-device to identify bad data in the measurement data;

[0056] Association analysis is based on algorithms such as apriori to analyze the association between events;

[0057] Digital deduction is based on weather data to predict the load of the distribution microgrid, carry out distribution micro-coordinated power flow calculation, analyze the operating status of the distribution microgrid under different meteorological conditions and switching conditions, and provide auxiliary decision-making functions for subsequent low voltage problem diagnosis and management decisions.

[0058] S25) Iterative optimization of the digital twin. Specifically, a digital twin evaluation index system is established, including acquisition accuracy, acquisition timescale, and bad data ratio. The digital twin model is iteratively optimized based on the evaluation results. Acquisition accuracy includes electrical quantity acquisition accuracy and meteorological data acquisition accuracy; acquisition timescale includes distribution network measurement acquisition timescale, microgrid measurement acquisition timescale, and meteorological data acquisition timescale; bad data includes the measurement bad data ratio and meteorological bad data ratio. Based on the evaluation results, the digital twin is iteratively optimized.

[0059] This embodiment analyzes the spatiotemporal distribution characteristics of low voltage based on data twins and explores the causes of low voltage in step S3, including the following steps:

[0060] S31) Analyze the spatiotemporal distribution characteristics of low voltage and select diagnosis and treatment areas. Specifically, the regional correlation of low voltage distribution is mined based on correlation association mining, including:

[0061] S311) obtaining a spatiotemporal history curve of the distribution microgrid, marking a bus with a bus voltage lower than a specified per-unit value as a low-voltage bus based on the per-unit voltage level of each node, multiple low-voltage buses forming a low-voltage area, and the substation and microgrid where low voltage occurs as a low-voltage substation and low-voltage microgrid, and counting the time when the low voltage occurs;

[0062] S312) constructing a low voltage characteristic time series of the region, line, substation, and microgrid according to the historical low voltage data of the region, line, substation, and microgrid, specifically, in the order of region, line, substation, and microgrid, according to the time dimensions of day, month, and quarter, at 96 points a day, i.e., one point every 15 minutes, to construct a low voltage curve sequence, and construct a low voltage characteristic time series of the region, line, and substation;

[0063] S313) Calculate the correlation of the low voltage characteristic time series of each region, line, substation, and microgrid, and select the strongly correlated regions or the regions where the strongly correlated lines, substations, and microgrids are located as the diagnosis and treatment areas.

[0064] When calculating the correlation of the low voltage characteristic time series of each region, line, substation, and microgrid, the Pearson correlation coefficient of the low voltage time series between regions, lines, and substations is calculated. The Pearson correlation coefficient calculation formula is as follows:

[0065]

[0066] The correlation coefficient ranges from -1 to 1. |r| indicates the degree of correlation between two variables. r > 0 indicates positive correlation, r < 0 indicates negative correlation, and r = 0 indicates no correlation. In this embodiment, correlations are divided into four categories based on the absolute value of the correlation coefficient r: 1) |r| < 0.3 is considered a very weak correlation; 2) 0.3 ≤ |r| < 0.5 is considered a weak correlation; 3) 0.5 ≤ |r| < 0.8 is considered a significant correlation; and 4) |r| ≥ 0.8 is considered a strong correlation. Lines, substations, and microgrids with strong positive correlations are prioritized for low voltage problem diagnosis and management.

[0067] S32) mining the causes of low voltage, specifically mining the causes of low voltage based on event correlation mining, including:

[0068] S321) Obtain simulation results and create a low voltage event list, specifically:

[0069] First, we build analysis dimensions, including relevant regions, low voltage / overvoltage events, low voltage / overvoltage occurrence periods, dates / holidays, low voltage / overvoltage severity, whether three-phase imbalance occurs, whether taps are appropriate, whether the line is too long, whether the wire diameter is too thin, and whether there are large-scale load fluctuations.

[0070] Then, we obtain the simulation data provided by the data twin. Specifically, we use the data twin to calculate the distribution network power flow, one point every 15 minutes. We calculate 96 power flows for all devices in the data twin and label the situation of each device.

[0071] Low voltage / overvoltage events are marked with 0, 1, and 2, where 0 indicates no low voltage / overvoltage occurs, 1 indicates low voltage occurs, and 2 indicates overvoltage occurs.

[0072] Ordinary dates / holidays are marked with 0 and 1, 0 for ordinary dates and 1 for holidays;

[0073] The low voltage / overvoltage time period is marked as 0 from 0:00 to 8:00, 1 from 9:00 to 17:00, and 2 from 18:00 to 24:00.

[0074] The severity of low voltage / overvoltage is marked with 0 or 1, where 0 is normal and 1 is severe.

[0075] Whether three-phase imbalance occurs is marked with 0 or 1, 0 is normal and 1 is serious;

[0076] Whether the tap is reasonable is marked with 0 or 1, 0 means reasonable and 1 means unreasonable;

[0077] Whether the line is too long is marked with 0 or 1, 0 means it is reasonable, and 1 means the line diameter is too long;

[0078] Whether the wire diameter is too thin or not is marked with 0 or 1, 0 means it is reasonable and 1 means the wire diameter is too thin;

[0079] Whether the large-scale load fluctuation is reasonable is marked with 0 or 1, where 0 means no load fluctuation and 1 means load fluctuation occurs.

[0080] S322) performing event mining on the low voltage event list using the Apriori algorithm to obtain a typical low voltage event set;

[0081] The Apriori association algorithm is a frequent itemset algorithm used in data mining to discover association rules. Let I = {I1, I2, …Ip, …, Im} be a set of m distinct items. Ip (p = 1, 2, …, m) is called an item, or simply an item. The set of items I is called an itemset, or simply an itemset. Any nonempty subset X of I that contains k items is called a k-itemset.

[0082] The transaction set of association mining is denoted as D, D = {T1, T2, ..., Tk, ..., Tn}, Tk (k = 1, 2, ..., n) is a non-empty subset of the item set I, called a transaction. Each transaction has one and only one identifier, called TID (TransactionID)

[0083] When using the Apriori algorithm for association rule mining, the association strength is controlled by three parameters: support, confidence, and lift:

[0084] (1) Support

[0085] The number of transactions in transaction set D that contain item set X is called the support number of item set X, recorded as occur(X). Then the support of item set X is defined as:

[0086] supp(X)=occur(X) / count(D)

[0087] If X,Y∈I and X∩Y is empty, the support of the association rule R:X→Y is denoted as supp(R):

[0088] supp(R)=support(X→Y)=P(X,Y)=occur(X and Y) / count(D)

[0089] Indicates the probability of X and Y occurring simultaneously. This parameter is used to eliminate association rules that appear infrequently and are meaningless.

[0090] (2) Confidence

[0091] The confidence of the association rule R:X→Y is defined as:

[0092] conf(R)=supp(X∪Y) / supp(X)=P(Y|X)

[0093] It indicates the possibility that Y will occur at the same time when X occurs. This parameter ensures the reliability of the mined association rules.

[0094] Based on the Apriori algorithm, low voltage association mining analysis is carried out to calculate support and confidence respectively, and typical low voltage events are obtained, including but not limited to:

[0095] One or more substations experience low voltage / overvoltage at multiple / single moments;

[0096] One or more substations experience low voltage at a specific time (9-17 o'clock or 18-24 o'clock);

[0097] Low voltage / overvoltage occurs on the lines in the substation at the same time;

[0098] Low voltage at the end of the line;

[0099] S323) Analyze the causes of low voltage corresponding to the typical low voltage events, including:

[0100] Low voltage / overvoltage occurs in one or more substations. Possible causes include unreasonable tap position, unreasonable configuration of reactive power compensation device, unreasonable wire diameter, etc.

[0101] Possible reasons for low voltage in one or more substations at a specific time are that the lines are too long or the distributed generation stops generating power at night;

[0102] The possible reasons for the simultaneous occurrence of low voltage / overvoltage on the substation lines are that the lines are too long and the load distribution is unreasonable;

[0103] Possible reasons for low voltage at the end of the line are that the line is too long, the wire diameter is unreasonable, etc.

[0104] In this embodiment, the cause of low voltage is verified based on the data twin in step S4. Specifically, the corresponding parameters of the digital twin are adjusted according to the low voltage cause mining results, and the distribution micro-grid collaborative power flow analysis calculation is performed to verify the effect of the parameter adjustment on low voltage control. The distribution micro-grid low voltage control strategy is generated in combination with the corresponding adjustment strategy. When the corresponding parameters of the digital twin are adjusted according to the low voltage cause mining results, the following situations are included:

[0105] Adjust the main transformer tap position according to the power flow sensitivity and evaluate the impact of adjusting the main transformer tap position on low voltage.

[0106] According to the reactive power optimization method, the impact of configuring reactive power compensation devices on low voltage is evaluated.

[0107] Adjust the line diameter and evaluate the impact of the adjusted line diameter on low voltage based on the power flow calculation results.

[0108] For newly built substations, evaluate their impact on low voltage based on the power flow calculation results.

[0109] In this embodiment, the low voltage causes and control strategies are generated in step S5. Specifically, based on the digital twin simulation results, evaluation index weights are constructed from the two dimensions of economy and control effect. The comprehensive scores of various strategies are calculated and ranked. The microgrid low voltage control strategy is preferably selected. When calculating the comprehensive scores according to the corresponding simulation results, the following steps are included:

[0110] S51) Match the parameter adjustment strategy to the case library to obtain the corresponding governance cost as an economic indicator. Governance economics refers to the governance cost corresponding to each governance measure. Specific governance costs can be evaluated by matching actual governance measures from the distribution network construction and transformation case library. Economics, from small to large, include adjusting taps, configuring reactive power compensation devices, changing line diameters, and building new substations.

[0111] S52) Calculate the average voltage and average three-phase imbalance according to the simulation results, perform weighted summation on the average voltage and average three-phase imbalance, and obtain the effectiveness index. The effectiveness is divided into two dimensions: the degree of improvement of low voltage / overvoltage and the effect of three-phase imbalance control after using the control strategy. Use digital twins to calculate the node average voltage and average three-phase imbalance index at 96 points in the distribution network for 24 hours a day. Determine the weights of the average voltage and average three-phase imbalance index based on the hierarchical analysis method, and calculate the comprehensive control effect by weight;

[0112] S53) Weighted summation of economic indicators and effectiveness indicators is performed to obtain a comprehensive score. Based on the analytic hierarchy process, the weights of economic and effectiveness are determined, and weighted scores are performed on the governance economic indicators and governance effectiveness. The comprehensive scores of the governance measures are obtained and ranked to obtain the recommended governance strategy.

[0113] Example 2

[0114] This embodiment also proposes a micro-device collaborative low-voltage diagnosis system based on digital twins and association mining, including a microprocessor and a computer-readable storage medium connected to each other, and the microprocessor is programmed or configured to execute the micro-device collaborative low-voltage diagnosis method based on digital twins and association mining described in Example 1.

[0115] This embodiment also proposes a computer-readable storage medium, which stores a computer program. The computer program is executed by a microprocessor to implement the steps of the micro-cooperative low-voltage diagnosis method based on digital twins and association mining described in Example 1.

[0116] This embodiment also proposes a computer program product, which includes a computer program. The computer program is executed by a microprocessor to implement the steps of the micro-cooperative low-voltage diagnosis method based on digital twins and association mining described in Example 1.

[0117] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed above with reference to the preferred embodiment, it is not intended to limit the present invention. Therefore, any simple modifications, equivalent variations, and modifications to the above embodiment that do not depart from the technical solution of the present invention and are based on the technical essence of the present invention shall fall within the scope of protection of the technical solution of the present invention.

Claims

1. A micro-device collaborative low voltage diagnosis method based on digital twins and association mining, characterized by: The following steps are involved: Obtain physical and environmental data of the target area; Build a digital twin model of the micro-distribution and distribution network; Obtain historical low voltage data, analyze the spatiotemporal distribution characteristics of low voltage, and screen diagnosis and treatment areas. Use digital twin models to simulate the diagnosis and treatment areas, and use association analysis algorithms to explore the causes of low voltage. According to the causes of low voltage, the corresponding treatment strategy is selected to adjust the digital twin model. The adjusted digital twin model is used for simulation to obtain simulation results. The simulation results are compared with historical low voltage data to verify the treatment effect of each selected treatment strategy. For the control strategies whose control effects meet the requirements, the comprehensive scores are calculated according to the corresponding simulation results, and the control strategy with the highest comprehensive score is used as the low voltage control strategy.

2. The method for low voltage diagnosis based on digital twin and association mining according to claim 1 is characterized in that: The physical data includes one or more of current, voltage, active power, reactive power data, and the opening and closing status of switches and circuit breakers; the environmental data includes one or more of weather, temperature, humidity, and altitude.

3. The method for low voltage diagnosis based on digital twin and association mining according to claim 1 is characterized in that: The steps to build a digital twin model of the micro-distribution network include: Based on physical data, the distribution network and microgrid are divided by boundaries, the distribution network, microgrid, and the distribution-microgrid boundary are identified, and digital twin entities of the distribution network, microgrid, and distribution-microgrid boundary are established. The digital twin entity of the operating environment is also established based on environmental data, and data measurement mapping relationships are also established; Based on the digital twin entities and data measurement mapping relationships, digital twin models of the distribution network, microgrid, distribution-microgrid boundary, and operating environment are constructed; According to the data measurement mapping relationship of the digital twin, the data of the corresponding digital twin entity is collected at a specified period and mapped to the digital twin model; Carry out digital twin applications and establish a digital twin evaluation index system, and iteratively optimize the digital twin model based on the evaluation results.

4. The method for low voltage diagnosis based on digital twin and association mining according to claim 3 is characterized in that: The digital twin applications include: Topology verification: Verify the correctness of switch status and topology association based on line power and bus voltage; Bad data identification: Identify bad data in measurement data based on collaborative state estimation between the microcontroller and the distribution system; Correlation analysis: analyzing the correlation between events; Digital simulation: Based on weather data, the load of the distribution microgrid is predicted, the distribution microgrid collaborative power flow calculation is carried out, and the operating status of the distribution microgrid under different meteorological conditions and switching conditions is analyzed.

5. The method for low voltage diagnosis based on digital twin and association mining according to claim 1 is characterized in that: When analyzing the spatiotemporal distribution characteristics of low voltage and screening diagnosis and control areas, the following should be considered: According to the historical low voltage data of regions, lines, substations and microgrids, the low voltage characteristic time series of regions, lines, substations and microgrids are constructed, the correlation of the low voltage characteristic time series of each region, line, substation and microgrid is calculated, and the areas with strong correlation or the areas where strongly correlated lines, substations and microgrids are located are selected as diagnosis and treatment areas.

6. The method for low voltage diagnosis based on digital twin and association mining according to claim 1 is characterized in that: When using the association analysis algorithm to explore the causes of low voltage, it includes: The simulation results are obtained and a low-voltage event list is established. The Apriori algorithm is used to mine the low-voltage event list to obtain a typical low-voltage event set. The corresponding low-voltage causes are analyzed for the events in the typical low-voltage event set.

7. The method for low voltage diagnosis based on digital twin and association mining according to claim 1 is characterized in that: When calculating the comprehensive score according to the corresponding simulation results, it includes: Match the parameter adjustment strategy to the case library and obtain the corresponding governance costs as economic indicators; Calculate the average voltage and average three-phase imbalance according to the simulation results, perform weighted summation on the average voltage and average three-phase imbalance to obtain the performance index; The economic indicators and performance indicators are weighted and summed to obtain a comprehensive score.

8. A micro-device collaborative low voltage diagnosis system based on digital twins and association mining, characterized by: It includes a microprocessor and a computer-readable storage medium connected to each other, and the microprocessor is programmed or configured to execute the micro-cooperative low-voltage diagnosis method based on digital twins and association mining according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which is executed by a microprocessor to implement the steps of the micro-cooperative low-voltage diagnosis method based on digital twins and association mining according to any one of claims 1 to 7.

10. A computer program product, characterized in that The computer program product includes a computer program, which is executed by a microprocessor to implement the steps of the micro-cooperative low-voltage diagnosis method based on digital twins and association mining as described in any one of claims 1 to 7.

Images