Power distribution network operation risk deduction method, system and medium considering distributed power supply

Abstract

The application discloses a power distribution network operation risk deduction method and system considering distributed power supply, and a medium. The method comprises the following steps: collecting power distribution network operation related data and extracting scene features; generating candidate operation scenes based on the scene features, obtaining typical operation scenes through reduction and merging; constructing a typical scene database, and associating scene triggering conditions to the typical scene database; calculating the matching degree of a current operation section and each typical operation scene, selecting a scene to be deduced, and generating a time sequence curve set of the scene to be deduced with an online condition probability; performing multi-period operation state deduction; calculating risk indexes based on the operation state deduction result; determining a comprehensive risk value according to the online condition probability of the scene to be deduced and the risk index out-of-limit discrimination result, and then determining a comprehensive risk grade; and outputting a risk early warning result and an operation disposal suggestion. The application can improve the accuracy of high-risk low-probability scene identification, future operation state deduction and risk disposal suggestion generation.

Description

Technical Field

[0001] This invention relates to the fields of distribution network operation analysis, distributed power source absorption, operation trend prediction and risk warning, and in particular to a method, system and medium for predicting distribution network operation risks that take into account distributed power sources. Background Technology

[0002] With the large-scale integration of distributed photovoltaic power, decentralized wind power, energy storage devices, electric vehicle charging facilities, and flexible loads, the distribution network is gradually transforming from a traditional unidirectional power supply network into an active distribution network with multi-element interaction among power sources, grid, load, and storage. The output of distributed power sources is affected by weather, time of day, season, and local operating conditions, exhibiting randomness and volatility; the load side is also significantly uncertain due to factors such as concentrated charging of electric vehicles, sudden increases in commercial load, and changes in peak-valley differences in residential loads.

[0003] Under the aforementioned operating conditions, distribution networks are prone to risks such as node voltage exceeding limits, line or transformer overload, reverse power flow in feeders, limited absorption of distributed generation, decreased load recovery capacity after faults, and loss of supply to critical loads. Traditional operation analysis methods are mostly based on single operating sections or deterministic power flow verification, which makes it difficult to reflect the continuous impact of changes in source, load, and storage on the operating status over a future period, and also makes it difficult to make a unified assessment of risks under different operating scenarios.

[0004] Existing methods still have shortcomings in the following aspects: First, the real-time matching between the typical scenario database formed by multi-source data and the current operating section is not close enough; second, the linkage between the output uncertainty and operational risks of distributed photovoltaic, distributed wind power, energy storage devices and other distributed power sources is not sufficiently characterized; third, in the rolling operational status simulation for multiple future time periods, the synchronous updates of network topology, energy storage charge status, distributed power source absorption limitation status and load recovery status are not sufficient; fourth, there is still room for improvement in the unified quantification, ranking and handling suggestions for risks such as voltage over-limit, line overload, load loss and distributed power source absorption limitation. Summary of the Invention

[0005] To address the shortcomings of the existing technologies, this invention provides a system and medium for calculating the operational risks of distribution networks that take into account distributed power sources. This system and medium are used to solve the problems in existing distribution network operational risk analysis, such as the lack of risk weights for scenario reduction, insufficient quantitative matching between current operating sections and typical scenarios, inadequate multi-period simulation and linkage updates, and the lack of scoring and ranking for operational handling suggestions.

[0006] Firstly, a method for extrapolating the operational risks of distribution networks that takes into account distributed generation is provided, including the following steps: S1: Collect relevant data on the operation of the distribution network, wherein the distribution network is a radial or weak-loop distribution network with limited interconnections; S2: Perform data preprocessing on data related to the operation of the distribution network, and extract scenario features based on the preprocessed data. S3: Generate candidate operating scenarios based on scenario features; construct a risk-weighted scenario distance matrix for candidate operating scenarios, and reduce and merge candidate operating scenarios according to offline scenario probability, risk contribution, and scenario coverage to obtain typical operating scenarios; S4: Construct a typical scenario database based on typical operating scenarios, and automatically generate scenario triggering conditions according to the statistical boundary values ​​or state sets of key scenario features in each typical operating scenario, as well as preset logical combination rules, and store the scenario triggering conditions in the typical scenario database. S5: Based on the relevant data of the current operating section of the distribution network, calculate the matching degree between the current operating section and each typical operating scenario, select the top K typical operating scenarios with the highest matching degree as the scenarios to be simulated, and update the online condition probability according to the offline scenario probability and matching degree of each scenario to be simulated, and generate the corresponding source-load-storage time series curve set of the scenarios to be simulated with online condition probability. S6: Based on the source load storage time series curve set of the scenario to be simulated, perform multi-period operation status simulation to obtain the operation status simulation results for each simulation period; S7: Calculate risk indicators based on operational status simulation results; S8: Based on the online conditional probability of the scenario to be simulated and the risk index exceeding the limit judgment results of each simulation period, determine the occurrence probability of each risk type in multiple simulation periods, and determine the comprehensive risk value by weighting and accumulating the occurrence probability, severity of consequences, duration coefficient and importance coefficient of each risk type, and determine the comprehensive risk level based on the comprehensive risk value or key risk index threshold. S9: Output risk warning results and operational handling suggestions based on the comprehensive risk level.

[0007] Furthermore, the data related to the operation of the distribution network includes distribution network topology data, line parameter data, transformer parameter data, node load data, switch status data, protection action data, meteorological data, historical fault data, as well as historical output data, predicted output data, and operating status data of distributed photovoltaic, distributed wind power, energy storage devices, and other distributed power sources.

[0008] Furthermore, in step S2, data preprocessing includes time alignment, outlier removal, missing value completion, topology verification, unit unification, and normalization. Historical and predicted output data of distributed photovoltaic, distributed wind power, and other distributed power sources are divided into time periods and meteorological types to form time series samples of output uncertainty of distributed power sources. For energy storage devices, state samples containing state of charge, rechargeable power, dischargeable power, and remaining available capacity are formed.

[0009] In step S2, the scenario features include one or more of the following: load level, load ramp rate, equipment load rate, weather type, fault type, distributed power output level, distributed power output fluctuation rate, distributed power prediction deviation, distributed power absorption limited status, energy storage charging status, important load ratio, and network topology status.

[0010] Furthermore, in step S3, the risk contribution of the i-th candidate running scenario Determined by the following formula: ; in, For a set of risk types, Let the risk value be the risk value of the k-th risk category corresponding to the i-th candidate running scenario. The weight of the k-th type of risk; Risk-weighted scenario distance Determined by the following formula: ; in, Let be the normalized distance between the i-th candidate running scenario and the j-th candidate running scenario on the m-th type of scenario features. and All are weighting coefficients.

[0011] Furthermore, in step S3, the scene coverage Determined by the following formula: ; in, The set of typical operating scenarios that are retained. Let be the offline scenario probability of the i-th candidate running scenario. To cover distance threshold, For indicator functions; Based on the scenario coverage reaching the preset coverage threshold, the top Q% of candidate operating scenarios with the highest risk contribution are retained; when the risk contribution of a candidate operating scenario is higher than the contribution threshold, even if its offline scenario probability is lower than the probability threshold, the candidate operating scenario is still retained as a typical operating scenario.

[0012] Further, in step S3, the offline scenario probability is determined based on at least one of the following: the frequency of occurrence of candidate running scenarios in the historical running time window, the generation probability of prediction error disturbance samples, and the set probability of fault or extreme weather injection samples. The offline scenario probabilities of each candidate running scenario are then normalized to make the sum of the offline scenario probabilities of each candidate running scenario equal to 1.

[0013] Furthermore, in step S4, scenario triggering conditions are automatically generated based on the statistical boundary values ​​of key scenario features within each typical operating scenario and preset logical combination rules, including: For each typical operating scenario, the statistical boundary values ​​of key scenario features are extracted from the candidate operating scenario sample set. The candidate operating scenario sample set includes one or more candidate operating scenario samples that have been reduced and merged in step S3 and belong to the typical operating scenario. When a candidate operating scenario forms a typical operating scenario on its own, the candidate operating scenario constitutes the corresponding candidate operating scenario sample set. When multiple candidate operating scenarios are merged to form a typical operating scenario, the multiple candidate operating scenarios together constitute the corresponding candidate operating scenario sample set. The statistical boundary values ​​include: confidence interval boundaries, quantile boundaries, or mean plus or minus standard deviation intervals of one or more continuous key scenario features among load level, distributed power generation output level, distributed power generation output volatility, load ramp rate, equipment load rate, and energy storage charge status under the typical operating scenario; and state sets or category sets of one or more discrete key scenario features among weather type, fault type, and distributed power generation absorption limitation status. The statistical boundary values ​​are encoded according to preset logical combination rules to generate the scene triggering conditions corresponding to the typical operating scenario; the preset logical combination rules include at least one of "AND" rule, "OR" rule and "NOT" rule, as well as combination rules formed by nesting the above rules; The scenario triggering conditions are associated with the typical operating scenarios and stored in the typical scenario database. The scenario triggering conditions include at least a trigger feature set, a continuous feature threshold range or a discrete feature state set, a logical combination rule, and a trigger confidence level.

[0014] Furthermore, in step S4, each typical operating scenario in the typical scenario database includes a scenario number, scenario name, applicable feeder or transformer area, scenario triggering conditions, candidate operating scenario sample identifier and its scenario feature data, offline scenario probability, source-load-storage time series curve, distributed power output uncertainty sample, network topology status, extrapolation time scale, risk criteria, associated risk indicators, risk labels and handling strategies.

[0015] Furthermore, in step S5, the matching degree between the current operating section and each typical operating scenario is... Calculated using the following formula: ; in, The matching feature set is a subset of key scene features that can be obtained based on relevant data or predicted data of the current running section and is configured with corresponding scene triggering conditions. This represents the real-time or predicted value of the m-th type of matching feature in the currently running cross-section. The triggering condition for the m-th class of matching features corresponding to the s-th typical running scenario is... It is a matching function for continuous feature intervals or a matching function for discrete state sets. These are the weighting coefficients; The top K typical operating scenarios with the highest matching degree are used to form the set of scenarios to be deduced. The online conditional probability of the s-th scenario to be simulated Calculated using the following formula: ; in, Let be the offline scene probability of the s-th scenario to be simulated. To match the sensitivity coefficient, s is the scenario number to be calculated for the online conditional probability, and r is the summation index for traversing the set of scenarios to be calculated.

[0016] Further, in step S6, the multi-time period operation state simulation includes: reading the load value, distributed power output value, distributed power prediction deviation, energy storage charge state, and network topology state in the t-th simulation period; performing power flow calculation; determining whether there are voltage over-limit, line overload, transformer overload, reverse power flow over-limit, fault isolation requirement, or load transfer requirement; updating the network topology, energy storage operation state, distributed power absorption limitation state, and load power supply state based on the determination result; entering the t+1-th simulation period until all time steps in future simulation periods are completed. The energy storage operation state includes energy storage charge state, energy storage charge / discharge state, and energy storage charge / discharge power; in each simulation period, updating the energy storage charge state based on the energy storage charge / discharge state and energy storage charge / discharge power, and using the updated energy storage charge state as the input for the next simulation period.

[0017] Furthermore, when it is determined that there is a need for fault isolation or load transfer, network reconfiguration is performed. The network reconfiguration includes fault section identification, power outage area identification, operation status judgment of sectionalizing switches and tie switches, transfer path search, distributed power source and energy storage support capacity verification, voltage verification after transfer, line capacity verification after transfer, and load recovery result calculation.

[0018] Furthermore, in step S7, the risk indicators include at least one of the following: voltage over-limit risk indicators, line overload risk indicators, transformer overload risk indicators, reverse power flow risk indicators, distributed power source absorption limitation risk indicators, load loss risk indicators, and power supply reliability risk indicators.

[0019] Furthermore, in step S8, the probability of occurrence of the k-th type of risk within the t-th simulation period is... Calculated using the following formula: ; in, For the set of scenarios to be simulated, Let be the online conditional probability of the scenario s to be simulated. Let be the risk index value of the k-th type of risk in the scenario to be simulated during the t-th simulation period. This represents the threshold for exceeding the limit for the k-th type of risk. For indicator functions; The overall risk value is determined using the following cumulative formula across multiple risks and time periods: ; in, For the comprehensive risk value, For a set of risk types, The weight of the k-th type of risk, As for the severity of the consequences, For duration coefficient, The load importance coefficient or equipment importance coefficient; the risk type set consists of one or more of the following: voltage over-limit risk, line overload risk, transformer overload risk, reverse power flow risk, distributed power source absorption limitation risk, load loss risk, and power supply reliability risk.

[0020] Furthermore, in step S8, the comprehensive risk level includes low risk, medium risk, high risk, and severe risk, and the comprehensive risk level is determined based on the range of the comprehensive risk value; when any key risk indicator exceeds the corresponding severe risk threshold, the comprehensive risk level is determined to be severe risk.

[0021] Furthermore, in step S9, the proposed operational handling is determined by calculating a comprehensive score for each candidate handling strategy z. Afterwards, the overall score was determined. Calculated using the following formula: ; in, The normalized overall risk reduction corresponding to candidate disposal strategy z. To normalize execution costs, To normalize the response delay, To normalize executability or equipment availability; when a candidate disposal strategy does not meet equipment operability, regulation capacity, topology constraints, or security constraints, the candidate disposal strategy is eliminated or [the following is omitted as it is not part of the original text]. Set to 0.

[0022] Secondly, a distribution network operation risk simulation system considering distributed generation is provided for executing the aforementioned distribution network operation risk simulation method considering distributed generation. The system includes: The data acquisition module is used to collect data related to the operation of the power distribution network; The data preprocessing module is used to preprocess the operation-related data and extract scenario features based on the preprocessed distribution network operation-related data. The typical scenario database module is used to generate, store, update, and call the typical scenario database. The typical scenario database stores data related to each typical running scenario and the corresponding scenario triggering conditions. The scenario matching module is used to calculate the matching degree between the current operating section and each typical operating scenario based on the relevant data of the current distribution network operating section, and select the top K typical operating scenarios as scenarios to be simulated. The runtime status simulation module is used to perform multi-time period runtime status simulations based on the scenario to be simulated. Run the risk assessment module to calculate the comprehensive score of risk indicators, comprehensive risk value, comprehensive risk level, and candidate disposal strategies; The early warning output module is used to output risk warning results and operational handling suggestions sorted by comprehensive score.

[0023] Thirdly, a distribution network operation risk simulation system considering distributed power sources is provided, including a processor, a memory, and a computer program stored in the memory and executable by the processor. When the processor executes the computer program, it implements the distribution network operation risk simulation method considering distributed power sources as described above.

[0024] Fourthly, a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned method for extrapolating the operational risks of a distribution network considering distributed power sources.

[0025] This invention proposes a method, system, and medium for calculating the operational risks of distribution networks that take into account distributed generation. Compared with existing technologies, this invention has the following advantages: (1) By collecting data from multiple sources, extracting scene features, closing the source of candidate scenes and reducing risk-weighted scenes, a typical scene database that can be called and updated is constructed to provide a scenario-based data foundation for intelligent simulation of distribution network operation risks and retain high-risk, low-probability operation states. (2) The output uncertainty of distributed photovoltaic, distributed wind power, energy storage devices and other distributed power sources is introduced into the whole process of scenario generation, operation status simulation and operation risk assessment, which improves the ability to characterize the operation risk of active distribution network; (3) By quantitative scenario matching and online conditional probability update based on the current operating section, the offline typical scenario library is transformed into a multi-scenario source load storage time series curve set under the current operating section, so as to realize the continuous prediction of future operating status. (4) By classifying the comprehensive risk values ​​and comprehensive risk levels across multiple risk periods, risks such as voltage overruns, line overloads, load loss and limited absorption of distributed power sources are uniformly output as results that can be predicted, sorted and handled. (5) Through a modular system structure, a distribution network operation risk intelligent simulation module is formed, which includes a typical scenario database, online matching, operation status simulation, operation risk assessment and disposal suggestion scoring, to meet the application requirements under various operation scenarios. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a flowchart of a method for extrapolating the operation risk of a distribution network that takes into account distributed power sources, provided in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0029] To address the shortcomings of existing distribution network operation risk analysis, such as the lack of risk weights in scenario reduction, insufficient quantitative matching between current operating sections and typical scenarios, inadequate multi-period simulation and linkage updates, and the lack of scoring and ranking for operational response suggestions, this invention proposes a distribution network operation risk simulation method, system, and medium that considers distributed generation. Through risk-weighted scenario reduction, automatic trigger condition generation, quantitative matching of current operating sections, multi-period linkage simulation of source, load, and storage, and scoring and ranking of response suggestions, a technical solution applicable to risk early warning and operational response is formed. The technical solution of this invention will be described in detail below with reference to specific embodiments.

[0030] like Figure 1 As shown, this embodiment of the invention provides a method for extrapolating the operational risks of a distribution network that takes into account distributed generation, including the following steps:

[0031] S1: Collect data related to the operation of the power distribution network.

[0032] In this embodiment, the distribution network operation-related data includes distribution network topology data, line parameter data, transformer parameter data, node load data, switch status data, protection action data, meteorological data, historical fault data, and historical output data, predicted output data, and operating status data of distributed photovoltaic, distributed wind power, energy storage devices, and other distributed power sources. The distribution network can be a single-feeder, distribution area, local multi-feeder, or regional distribution network, and can adopt a radial or weak-loop network topology with limited interconnections. The distribution network operation-related data comes from one or more of the following systems: SCADA system, distribution automation system, EMS system, DMS system, distribution transformer monitoring terminal, meteorological interface, fault recording device, and power outage management system.

[0033] S2: Perform data preprocessing on data related to the operation of the distribution network, and extract scenario features based on the preprocessed data.

[0034] Specifically, data preprocessing includes time alignment, outlier removal, missing value completion, topology verification, dimensional unification, and normalization. Specifically, historical and predicted output data for distributed photovoltaic, distributed wind power, and other distributed power sources are divided into time periods and meteorological types to form time-series samples of output uncertainty for distributed power sources; for energy storage devices, state samples are formed including state of charge, rechargeable power, dischargeable power, and remaining available capacity.

[0035] Then, scene features are extracted, including one or more of the following: load level, load ramp rate, equipment load rate, weather type, fault type, distributed power output level, distributed power output fluctuation rate, distributed power prediction deviation, distributed power absorption limitation status, energy storage charging status, proportion of important loads, reverse power flow status, and network topology status.

[0036] S3: Generate candidate operating scenarios based on scenario features. The candidate operating scenarios are derived from at least historical operating time window slices, prediction error disturbance samples, and fault or extreme weather injection samples. Construct a risk-weighted scenario distance matrix for the candidate operating scenarios, and reduce and merge the candidate operating scenarios according to offline scenario probability, risk contribution, and scenario coverage to obtain typical operating scenarios. The scenario coverage is used to characterize the probability coverage of the retained typical operating scenarios over the candidate operating scenarios, and the risk contribution is determined based on the historical operating records corresponding to the candidate operating scenarios or risk indicators obtained from offline pre-simulation based on historical data.

[0037] Specifically, the candidate operating scenarios originate from at least three categories: historical operating scenarios formed by slicing historical SCADA or distribution automation data into preset time windows; forecasted extended scenarios formed by load forecasting, distributed power generation forecasting, and their forecasting error disturbances; and disturbance scenarios formed by offline injection based on fault types, switching actions, meteorological warnings, and extreme weather conditions. To facilitate a unified representation of the candidate operating scenarios, the i-th candidate operating scenario is represented as a scenario feature vector. : ; in, For load level, For load ramp rate, For the output level of distributed power sources, For the output volatility of distributed power sources, For distributed generation prediction bias, This is a state where the absorption capacity of distributed power sources is limited. In the state of energy storage charge, For equipment load rate, It is a reverse current state. Code the fault or weather type. As the proportion of critical loads, The network topology state vector for the i-th candidate operating scenario consists of the opening and closing states of segment switches, tie switches, and other switches involved in adjusting the operation mode of the distribution network.

[0038] Next, a risk-weighted scenario distance matrix is ​​constructed between candidate operating scenarios. The risk-weighted scenario distance integrates at least the load characteristic distance, distributed power output characteristic distance, distributed power absorption limited state distance, energy storage state distance, network topology state distance, and risk contribution distance. Among them, the network topology state distance is used to characterize the differences in the opening and closing states of sectional switches, tie switches, and other operable switches in two candidate operating scenarios. Then, the candidate operating scenarios are reduced and merged according to the offline scenario probability, risk contribution, and scenario coverage to form typical operating scenarios, while retaining high-risk, low-probability scenarios with risk contribution higher than the contribution threshold.

[0039] The network topology state vector of the i-th candidate running scenario is represented as: ; in, The number of switches participating in network topology state coding; when the nth switch is closed, When the nth switch is turned off, .

[0040] Network topology state distance: ; in, The distance between the i-th and j-th candidate running scenarios is the network topology state distance. This distance represents the proportion of different switching states in the two candidate running scenarios. The larger the distance, the greater the difference in network topology.

[0041] In this embodiment, the risk-weighted scenario distance between the i-th candidate running scenario and the j-th candidate running scenario is... Determined by the following formula: ; in, Let be the normalized distance between the i-th candidate running scenario and the j-th candidate running scenario on the m-th type of scenario features. and All are weighting coefficients; The risk contribution of the i-th candidate running scenario.

[0042] Risk contribution of the i-th candidate operating scenario Determined by the following formula: ; in, For a set of risk types, Let the risk value be the risk value of the k-th risk category corresponding to the i-th candidate running scenario. is the weight of the k-th type of risk.

[0043] Offline scene probability The candidate operating scenario is determined based on at least one of the following: the frequency of occurrence of the candidate operating scenario in the historical operating time window, the generation probability of the prediction error disturbance sample, and the set probability of the fault or extreme weather injection sample. The offline scenario probability of each candidate operating scenario is then normalized to make the sum of the probabilities of the offline scenarios equal to 1.

[0044] Scene coverage Determined by the following formula: ; in, The set of typical operating scenarios that are retained. Let be the offline scenario probability of the i-th candidate running scenario. To cover distance threshold, For indicator functions; Based on the premise that the scenario coverage reaches the preset coverage threshold, the top Q% of candidate operating scenarios with the highest risk contribution are retained first, and the value of Q is set according to the actual situation. When the risk contribution of a candidate operating scenario is higher than the contribution threshold, even if its offline scenario probability is lower than the probability threshold, the candidate operating scenario is still retained as a typical operating scenario.

[0045] S4: Construct a typical scenario database based on typical operating scenarios, automatically generate scenario triggering conditions according to the statistical boundary values ​​of key scenario features in each typical operating scenario and preset logical combination rules, and store the scenario triggering conditions in the typical scenario database.

[0046] Specifically, scenario triggering conditions are automatically generated based on the statistical boundary values ​​of key scenario characteristics within each typical operating scenario and preset logical combination rules, including: For each typical operating scenario, the statistical boundary values ​​of key scenario features are extracted from the candidate operating scenario sample set. The candidate operating scenario sample set includes one or more candidate operating scenario samples that have been reduced and merged in step S3 and belong to the typical operating scenario. When a candidate operating scenario alone forms a typical operating scenario, that candidate operating scenario constitutes the corresponding candidate operating scenario sample set. When multiple candidate operating scenarios are merged to form a typical operating scenario, the multiple candidate operating scenarios together constitute the corresponding candidate operating scenario sample set. Specifically, the scenario feature values ​​of all candidate scenario samples under the typical operating scenario are read, and the scene features described in step S2 are used to... Key scenario features for characterizing the typical operating scenario are selected. Statistical boundaries are extracted for continuous key scenario features in load level, distributed power generation output level, distributed power generation output volatility, distributed power generation prediction deviation, load ramp rate, equipment load rate, energy storage charge status, and proportion of important loads. The statistical boundary values ​​can be confidence interval boundaries, quantile boundaries, mean plus or minus standard deviation intervals, or threshold intervals corrected based on expert experience for the feature sample set. For discrete key scenario features in weather type, fault type, distributed power generation absorption limitation state, reverse power flow state, and network topology state, their state sets or category sets are extracted as their statistical boundary values.

[0047] The statistical boundary values ​​are encoded according to preset logical combination rules to generate the scene triggering conditions corresponding to the typical operating scenario; the preset logical combination rules include at least one of "AND" rules, "OR" rules and "NOT" rules, as well as combination rules formed by nesting the above rules; The scenario triggering conditions are associated with the typical operating scenarios and stored in the typical scenario database. The scenario triggering conditions include at least a trigger feature set, a continuous feature threshold range or a discrete feature state set, a logical combination rule, and a trigger confidence score. The trigger confidence score is determined by at least one of the following: historical sample hit rate, statistical boundary stability index, and expert correction coefficient; wherein, the statistical boundary stability index is used to characterize the degree of fluctuation of the key scenario feature threshold range within the scenario in different historical time windows.

[0048] For example, in a distributed power generation high-output, low-load scenario, the scenario triggering condition can be set as follows: the distributed power generation output level is higher than a first output threshold, the feeder load level is lower than a first load threshold, and the weather type meets the conditions of sunny or partly cloudy. In a distributed power generation low-output scenario during peak load, the scenario triggering condition can be set as follows: the feeder load level is higher than a second load threshold, and the distributed power generation output level is lower than a second output threshold. In a scenario of concentrated charging of electric vehicles or a sudden increase in load, the scenario triggering condition can be set as follows: the load ramp rate exceeds a preset ramp rate threshold, or the electric vehicle charging power exceeds a preset charging power threshold. In a scenario of fault disturbance or extreme weather, the scenario triggering condition can be set as follows: a fault state, protection action, switch tripping, or weather warning state is detected.

[0049] Each typical operating scenario in the typical scenario database includes a scenario number, scenario name, applicable feeder or transformer area, scenario triggering conditions, candidate operating scenario sample identifier and its scenario characteristic data, offline scenario probability, source-load-storage time series curve, distributed power output uncertainty sample, network topology status, projection time scale, risk criteria, associated risk indicators, risk labels and handling strategies.

[0050] S5: Based on relevant data from the current operating sections of the distribution network (current operating section data, predicted output of distributed generation, load forecast results, energy storage availability status, and weather forecast results), calculate the matching degree between the current operating section and each typical operating scenario. Select the top K typical operating scenarios with the highest matching degree as the scenarios to be simulated. Update the online conditional probability based on the offline scenario probability and matching degree of each scenario to be simulated, generating the corresponding set of source-load-storage time-series curves for the scenarios to be simulated with online conditional probabilities. It should be noted that when K=1, the set of source-load-storage time-series curves for the scenarios to be simulated only includes single-scenario source-load-storage time-series curves.

[0051] Specifically, the matching process is as follows: real-time feature values ​​are extracted from the current operating section, and the load matching degree, distributed power output matching degree, energy storage status matching degree, topology status matching degree, meteorological status matching degree, and fault status matching degree are calculated respectively. Then, the matching degree between the current operating section and each typical operating scenario is calculated using the following formula. : ; in, The matching feature set is a subset of key scene features that can be obtained based on relevant data or predicted data of the current running section and is configured with corresponding scene triggering conditions. This represents the real-time or predicted value of the m-th type of matching feature in the currently running cross-section. The triggering condition for the m-th class of matching features corresponding to the s-th typical running scenario is... These are the weighting coefficients; The function is either a continuous feature interval matching function or a discrete state set matching function. The continuous feature interval matching function is determined based on the distance between the real-time value or the predicted value and the feature threshold interval. The discrete state set matching function is determined based on whether the real-time state or the predicted state belongs to the state set in the triggering condition. When multiple typical operating scenarios meet the triggering conditions, the system will determine the matching degree. The scenarios are ranked by trigger confidence, probability of occurrence, and risk contribution, and the top K typical operating scenarios are selected to form the set of scenarios to be simulated. The online conditional probability is updated according to the following formula. : ; in, Let be the offline scene probability of the s-th scenario to be simulated. To match the sensitivity coefficient, s is the scenario number to be calculated for the online conditional probability, and r is the summation index for traversing the set of scenarios to be calculated.

[0052] S6: Based on the source-load-storage time-series curve set of the scenario to be simulated, perform multi-period operation status simulation. Perform power flow calculation at each time step according to the preset simulation time step, and synchronously update the network topology, energy storage operation status, distributed power consumption limitation status and load power supply status to obtain the operation status quantities in the future simulation period, including node voltage, line power flow, transformer load rate, reverse power flow, power outage load and load recovery status.

[0053] Specifically, the multi-period operation status simulation includes: assuming the future simulation period is divided into multiple time steps, denoted as t=1,2,…,T; in one embodiment, the future simulation period is 4 hours to 24 hours, and the simulation time step Δt is 15 minutes or 1 hour; in the t-th simulation period, load values, distributed generation output values, distributed generation prediction deviations, energy storage operation status, and network topology status are read, and power flow calculation is performed. The power flow calculation can employ at least one of AC power flow, DistFlow power flow, or linearized power flow models to obtain node voltage, branch power flow, line load rate, transformer load rate, network loss, and reverse power flow; and determining whether there is a voltage overload. The simulation process considers various factors, including: power supply limitations, line overload, transformer overload, reverse power flow exceeding limits, fault isolation requirements, and load transfer requirements. Based on the judgment results and under the premise of meeting equipment operability constraints, the network topology, energy storage operating status, distributed power source absorption limitation status, and load power supply status are updated, and the updated status variables are used as inputs for the next simulation period. In the (t+1)th simulation period, for distributed power sources and energy storage devices, the output status, absorption limitation status, and energy storage operating status of distributed photovoltaic, distributed wind power, and other distributed power sources are updated at each time step (simulation period) before simulation. This process continues until all time steps within the future simulation period are completed, forming the simulation results for the operating status within the future simulation period. The energy storage operating status describes the operating characteristics of the energy storage device during the multi-period simulation process, including the energy storage state of charge, energy storage charge / discharge status, and energy storage charge / discharge power. The energy storage state of charge characterizes the remaining energy level of the energy storage, and the energy storage charge / discharge status characterizes the current operating mode of the energy storage.

[0054] When power flow calculations indicate voltage overload, line overload, transformer overload, reverse power flow overload, fault isolation requirements, or load transfer requirements, the system determines whether switching actions or network reconfiguration are necessary. If necessary, the network topology is updated based on tie switches, sectionalizing switches, and transferable paths, and the operating status is recalculated; if not, the current topology and operating mode are maintained.

[0055] In one implementation, when the system is in a scenario of high output and low load of distributed power sources, if the simulation results show that the node voltage exceeds the upper limit, reverse power flow occurs at the feeder outlet, the duration of reverse power flow exceeds the set threshold, or the proportion of distributed power source absorption is limited exceeds the set threshold, then it is determined that there are risks related to distributed power source absorption and voltage exceeding the limit.

[0056] In another implementation, when the system is in a scenario of low output of distributed power supply during peak load or centralized charging of electric vehicles, if the simulation results show that the line load rate or transformer load rate exceeds the limit, the node voltage is lower than the lower limit, and the power supply margin of important loads is insufficient, it is determined that there is a risk of operation during peak load or local load surge.

[0057] In another implementation, when the system is under fault disturbance or extreme weather conditions, the system performs fault isolation and network reconfiguration simulations. The network reconfiguration simulation includes: identifying faulty sections, identifying power outage areas, determining the operability of sectionalizing switches and tie switches, searching for feasible power transfer paths, verifying voltage constraints and line capacity constraints after power transfer, verifying the support capacity of distributed power sources and energy storage devices for local loads, and calculating recoverable load, unrecoverable load, and estimated recovery time.

[0058] S7: Calculate risk indicators based on the operational status simulation results.

[0059] Specifically, the risk indicators include at least one of the following: voltage over-limit risk indicators, line overload risk indicators, transformer overload risk indicators, reverse power flow risk indicators, distributed power source absorption limitation risk indicators, load loss risk indicators, and power supply reliability risk indicators.

[0060] More specifically, the voltage over-limit risk index refers to the magnitude and duration of the over-limit voltage at nodes; the line overload risk index refers to the line load rate and overload duration; the transformer overload risk index refers to the transformer load rate and overload duration; the reverse power flow risk index refers to the feeder reverse power flow and reverse power flow duration; the distributed generation absorption limitation risk index refers to the proportion and capacity of the distributed generation absorption limitation; the load loss risk index refers to the amount of power outage, recoverable load, and unrecoverable load; and the power supply reliability risk index refers to the interruption time of important loads.

[0061] S8: Based on the online conditional probability of the scenario to be simulated and the risk index exceeding the limit judgment results of each simulation period, determine the occurrence probability of each risk type in multiple simulation periods, and determine the comprehensive risk value by weighting and accumulating the occurrence probability, severity of consequences, duration coefficient and importance coefficient of each risk type, and determine the comprehensive risk level based on the comprehensive risk value or key risk index threshold.

[0062] Specifically, the probability of occurrence of the k-th type of risk within the t-th projection period. Calculated using the following formula: ; in, For the set of scenarios to be simulated, Let be the online conditional probability of the scenario s to be simulated. Let be the risk index value of the k-th type of risk in the scenario to be simulated during the t-th simulation period. This represents the threshold for exceeding the limit for the k-th type of risk. Let S5 be the indicator function; thus, the online conditional probability obtained from S5 forms a closed loop with the comprehensive risk value from S8. The overall risk value is determined using the following cumulative formula across multiple risks and time periods: ; in, For the comprehensive risk value, For a set of risk types, The weight of the k-th type of risk, For the severity of the consequences of the k-th type of risk, Let be the duration coefficient of the k-th type of risk. The risk type set consists of one or more of the following: voltage overload risk, line overload risk, transformer overload risk, reverse power flow risk, distributed power source absorption limitation risk, load loss risk, and power supply reliability risk. The severity of the risk consequences is determined by the overload amplitude, overload degree, reverse power flow power, distributed power source absorption limitation capacity, power loss load amount, or important load interruption time.

[0063] In this embodiment, the comprehensive risk level includes low risk, medium risk, high risk, and severe risk, and the comprehensive risk level is determined based on the range of comprehensive risk values. When any key risk indicator exceeds the corresponding severe risk threshold, the comprehensive risk level is determined to be severe risk. When multiple risks occur simultaneously, the system sorts them according to comprehensive risk level, degree of impact of important loads, duration of exceeding limits, and equipment importance to form a risk list.

[0064] S9: Output risk warning results and operational handling suggestions based on the comprehensive risk level. The risk warning results include risk scenario, risk type, risk location, risk level, risk occurrence period, and risk development trend. The operational handling suggestions calculate a comprehensive score based on the normalized reduction amount of the comprehensive risk value, normalized execution cost, normalized response latency, and equipment executable constraints of the candidate handling strategies, and output the priority according to the score from high to low.

[0065] Specifically, the proposed operational handling is achieved by calculating a comprehensive score for each candidate handling strategy z. Afterwards, the overall score was determined. Calculated using the following formula: ; in, The normalized overall risk reduction corresponding to candidate disposal strategy z. To normalize execution costs, To normalize the response delay, To normalize executability or equipment availability; when a candidate disposal strategy does not meet equipment operability, regulation capacity, topology constraints, or security constraints, the candidate disposal strategy is eliminated or [the following is omitted as it is not part of the original text]. Set to 0.

[0066] When the risk type is voltage over-limit risk, candidate handling strategies include reactive power compensation adjustment, energy storage charging, distributed power output adjustment, or operation mode adjustment; when the risk type is line overload or transformer overload risk, candidate handling strategies include load transfer, network reconfiguration, energy storage discharge, or load adjustment; when the risk type is load loss risk, candidate handling strategies include distributed power support, energy storage support, load restoration, or emergency repair resource allocation; when the risk type is distributed power absorption limitation risk, candidate handling strategies include energy storage absorption, reactive power and voltage regulation, or distributed power output optimization.

[0067] The above embodiments provide a method for intelligent simulation of distribution network operation risks. Through multi-source data acquisition, scenario feature extraction, candidate scenario source closure, and risk-weighted scenario reduction, a callable and updatable database of typical scenarios is constructed. This provides a scenario-based data foundation for intelligent simulation of distribution network operation risks while retaining high-risk, low-probability operating states. The method incorporates the output uncertainty of distributed photovoltaic, decentralized wind power, energy storage devices, and other distributed power sources into the entire process of scenario generation, operation state simulation, and operation risk assessment, improving the ability to characterize active distribution network operation risks. Furthermore, it achieves this through quantitative scenario matching based on the current operating section. The system integrates online conditional probability updates, transforming the offline typical scenario library into a set of multi-scenario source-load-storage time-series curves under the current operating section, enabling continuous prediction of future operating states. Through multi-risk, multi-time-period comprehensive risk values ​​and comprehensive risk level classification, risks such as voltage overruns, line overloads, load loss, and limited distributed power absorption are uniformly output as predictable, sortable, and manageable results. Through a modular system structure, an intelligent prediction module for distribution network operation risks is formed, which includes a typical scenario database, online matching, operation status prediction, operation risk assessment, and handling suggestion scoring, meeting the application requirements under various operating scenarios.

[0068] This invention also provides a distribution network operation risk simulation system considering distributed generation, used to execute the distribution network operation risk simulation method considering distributed generation as described above. The system includes: The data acquisition module is used to collect data related to the operation of the power distribution network; The data preprocessing module is used to preprocess the operation-related data, extract scenario features based on the preprocessed distribution network operation-related data, and form a time series sample of distributed power generation output uncertainty. The typical scenario database module is used to generate, store, update, and call the typical scenario database. The typical scenario database stores data related to each typical running scenario and the corresponding scenario triggering conditions. The scenario matching module is used to calculate the matching degree between the current operating section and each typical operating scenario based on the relevant data of the current distribution network operating section, select the top K typical operating scenarios as scenarios to be simulated, and generate the corresponding multi-scenario source-load-storage time series curve set with online conditional probability. The runtime status simulation module is used to perform multi-time period runtime status simulations based on the scenario to be simulated. Run the risk assessment module to calculate the comprehensive score of risk indicators, comprehensive risk value, comprehensive risk level, and candidate disposal strategies; The early warning output module is used to output risk warning results and operational handling suggestions sorted by comprehensive score; The visualization or human-computer interaction module is used to display the risk location, risk level, risk development trend and operational handling suggestions, and to receive instructions to adjust scenario parameters, risk thresholds or handling strategies.

[0069] This invention also provides a distribution network operation risk simulation system that takes into account distributed power sources, including a processor, a memory, and a computer program stored in the memory and executable by the processor. When the processor executes the computer program, it implements the distribution network operation risk simulation method that takes into account distributed power sources as described above.

[0070] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the aforementioned method for calculating the operational risks of a distribution network considering distributed power sources.

[0071] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0072] It should be understood that the functional unit modules in the various embodiments of the present invention can be concentrated in one processing unit, or each unit module can exist physically separately, or two or more unit modules can be integrated into one unit module, and can be implemented in hardware or software.

[0073] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0074] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0075] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0076] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0077] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for extrapolating the operational risks of a distribution network that takes into account distributed generation, characterized in that, Includes the following steps: S1: Collect relevant data on the operation of the distribution network, wherein the distribution network is a radial or weak-loop distribution network with limited interconnections; S2: Perform data preprocessing on data related to the operation of the distribution network, and extract scenario features based on the preprocessed data. S3: Generate candidate operating scenarios based on scenario features; construct a risk-weighted scenario distance matrix for candidate operating scenarios, and reduce and merge candidate operating scenarios according to offline scenario probability, risk contribution, and scenario coverage to obtain typical operating scenarios; S4: Construct a typical scenario database based on typical operating scenarios, and automatically generate scenario triggering conditions according to the statistical boundary values ​​of key scenario features in each typical operating scenario and preset logical combination rules, and store the scenario triggering conditions in the typical scenario database. S5: Based on the relevant data of the current operating section of the distribution network, calculate the matching degree between the current operating section and each typical operating scenario, select the top K typical operating scenarios with the highest matching degree as the scenarios to be simulated, and update the online condition probability according to the offline scenario probability and matching degree of each scenario to be simulated, and generate the corresponding source-load-storage time series curve set of the scenarios to be simulated with online condition probability. S6: Based on the source load storage time series curve set of the scenario to be simulated, perform multi-period operation status simulation to obtain the operation status simulation results for each simulation period; S7: Calculate risk indicators based on operational status simulation results; S8: Based on the online conditional probability of the scenario to be simulated and the risk index exceeding the limit judgment results of each simulation period, determine the occurrence probability of each risk type in multiple simulation periods, and determine the comprehensive risk value by weighting and accumulating the occurrence probability, severity of consequences, duration coefficient and importance coefficient of each risk type. Determine the comprehensive risk level based on the comprehensive risk value or the threshold of key risk indicators. S9: Output risk warning results and operational handling suggestions based on the comprehensive risk level.

2. The method for extrapolating the operation risk of distribution networks considering distributed generation as described in claim 1, characterized in that, In step S2, the scenario features include one or more of the following: load level, load ramp rate, equipment load rate, weather type, fault type, distributed power output level, distributed power output fluctuation rate, distributed power prediction deviation, distributed power absorption limited status, energy storage charging status, important load ratio, and network topology status.

3. The method for extrapolating the operation risk of distribution networks considering distributed generation as described in claim 1, characterized in that, In step S3, the risk contribution of the i-th candidate running scenario Determined by the following formula: ； in, For a set of risk types, Let the risk value be the risk value of the k-th risk category corresponding to the i-th candidate running scenario. The weight of the k-th type of risk; Risk-weighted scenario distance Determined by the following formula: ； in, Let be the normalized distance between the i-th candidate running scenario and the j-th candidate running scenario on the m-th type of scenario features. and All are weighting coefficients.

4. The method for extrapolating the operation risk of distribution networks considering distributed generation as described in claim 3, characterized in that, In step S3, the scene coverage Determined by the following formula: ； in, The set of typical operating scenarios that are retained. Let be the offline scenario probability of the i-th candidate running scenario. To cover distance threshold, For indicator functions; Based on the scenario coverage reaching the preset coverage threshold, the top Q% of candidate operating scenarios with the highest risk contribution are retained; when the risk contribution of a candidate operating scenario is higher than the contribution threshold, even if its offline scenario probability is lower than the probability threshold, the candidate operating scenario is still retained as a typical operating scenario.

5. The method for extrapolating the operation risk of distribution networks considering distributed generation as described in claim 1, characterized in that, In step S3, the offline scenario probability is determined based on at least one of the following: the frequency of occurrence of candidate running scenarios in the historical running time window, the generation probability of prediction error disturbance samples, and the set probability of fault or extreme weather injection samples. The offline scenario probability of each candidate running scenario is then normalized to make the sum of the offline scenario probabilities of each candidate running scenario equal to 1.

6. The method for extrapolating the operation risk of a distribution network considering distributed generation as described in claim 1, characterized in that, In step S4, scenario triggering conditions are automatically generated based on the statistical boundary values ​​of key scenario features in each typical operating scenario and preset logical combination rules, including: For each typical operating scenario, the statistical boundary values ​​of key scenario features are extracted from the candidate operating scenario sample set. The statistical boundary values ​​include: the confidence interval boundary, quantile boundary, or mean plus or minus standard deviation interval of one or more continuous key scenario features in the typical operating scenario, such as load level, distributed power output level, distributed power output fluctuation rate, load ramp rate, equipment load rate, and energy storage charge state; and the state set or category set of one or more discrete key scenario features in the weather type, fault type, and distributed power absorption limitation state. The statistical boundary values ​​are encoded according to preset logical combination rules to generate the scene triggering conditions corresponding to the typical operating scenario; the preset logical combination rules include at least one of "AND" rule, "OR" rule and "NOT" rule, as well as combination rules formed by nesting the above rules; The scenario triggering conditions are associated with the typical operating scenarios and stored in the typical scenario database. The scenario triggering conditions include at least a trigger feature set, a continuous feature threshold range or a discrete feature state set, a logical combination rule, and a trigger confidence level.

7. The method for extrapolating the operation risk of distribution networks considering distributed generation as described in claim 1, characterized in that, In step S4, each typical operating scenario in the typical scenario database includes a scenario number, scenario name, applicable feeder or transformer area, scenario triggering conditions, candidate operating scenario sample identifier and its scenario feature data, offline scenario probability, source-load-storage time series curve, distributed power output uncertainty sample, network topology status, extrapolation time scale, risk criteria, associated risk indicators, risk labels and handling strategies.

8. The method for extrapolating the operation risk of distribution networks considering distributed generation as described in claim 1, characterized in that, In step S5, the matching degree between the current operating section and each typical operating scenario is determined. Calculated using the following formula: ； in, The matching feature set is a subset of key scene features that can be obtained based on relevant data or predicted data of the current running section and is configured with corresponding scene triggering conditions. This represents the real-time or predicted value of the m-th type of matching feature in the currently running cross-section. Let m be the trigger condition for the m-th class of matching features corresponding to the s-th typical running scenario. It is a matching function for continuous feature intervals or a matching function for discrete state sets. These are the weighting coefficients; The top K typical operating scenarios with the highest matching degree are used to form the set of scenarios to be deduced. The online conditional probability of the s-th scenario to be simulated Calculated using the following formula: ； in, Let be the offline scene probability of the s-th scenario to be simulated. To match the sensitivity coefficient.

9. The method for extrapolating the operation risk of a distribution network considering distributed generation as described in claim 1, characterized in that, In step S6, the multi-time period operation status simulation includes: reading the load value, distributed power output value, distributed power prediction deviation, energy storage charge status, and network topology status in the t-th simulation period; performing power flow calculation; determining whether there are voltage over-limit, line overload, transformer overload, reverse power flow over-limit, fault isolation requirement, or load transfer requirement; updating the network topology, energy storage operation status, distributed power absorption limitation status, and load power supply status based on the judgment results to obtain node voltage, line power flow, transformer load rate, reverse power flow, power outage load, and load recovery status; entering the t+1-th simulation period until all time steps in future simulation periods are completed.

10. The method for extrapolating the operation risk of a distribution network considering distributed generation as described in claim 9, characterized in that, When it is determined that there is a need for fault isolation or load transfer, network reconfiguration is performed.

11. The method for extrapolating the operation risk of a distribution network considering distributed generation as described in claim 1, characterized in that, In step S7, the risk indicators include at least one of the following: voltage over-limit risk indicators, line overload risk indicators, transformer overload risk indicators, reverse power flow risk indicators, distributed power source absorption limitation risk indicators, load loss risk indicators, and power supply reliability risk indicators.

12. The method for extrapolating the operation risk of a distribution network considering distributed generation as described in claim 1, characterized in that, In step S8, the probability of occurrence of the k-th type of risk during the t-th simulation period. Calculated using the following formula: ； in, For the set of scenarios to be simulated, Let be the online conditional probability of the scenario s to be simulated. Let be the risk index value of the k-th type of risk in the scenario to be simulated during the t-th simulation period. This represents the threshold for exceeding the limit for the k-th type of risk. For indicator functions; The overall risk value is determined using the following cumulative formula across multiple risks and time periods: ； in, For the comprehensive risk value, For a set of risk types, The weight of the k-th type of risk, As for the severity of the consequences, For duration coefficient, The load importance coefficient or equipment importance coefficient; the risk type set consists of one or more of the following: voltage over-limit risk, line overload risk, transformer overload risk, reverse power flow risk, distributed power source absorption limitation risk, load loss risk, and power supply reliability risk.

13. The method for extrapolating the operation risk of a distribution network considering distributed generation as described in claim 1, characterized in that, In step S8, the comprehensive risk level includes low risk, medium risk, high risk, and severe risk, and the comprehensive risk level is determined based on the range of the comprehensive risk value. When any key risk indicator exceeds the corresponding severe risk threshold, the overall risk level will be determined as severe risk.

14. The method for extrapolating the operation risk of a distribution network considering distributed generation as described in claim 1, characterized in that, In step S9, the proposed operational handling is determined by calculating a comprehensive score for each candidate handling strategy z. Afterwards, the overall score was determined. Calculated using the following formula: ； in, The normalized overall risk reduction corresponding to candidate disposal strategy z. To normalize execution costs, To normalize the response delay, To normalize executability or equipment availability; when a candidate disposal strategy does not meet equipment operability, regulation capacity, topology constraints, or security constraints, the candidate disposal strategy is eliminated or [the following is omitted as it is not part of the original text]. Set to 0.

15. A distribution network operation risk simulation system considering distributed generation, characterized in that, The system is used to perform the distribution network operation risk estimation method considering distributed generation as described in any one of claims 1 to 14, the system comprising: The data acquisition module is used to collect data related to the operation of the power distribution network; The data preprocessing module is used to preprocess the operation-related data and extract scenario features based on the preprocessed distribution network operation-related data. The typical scenario database module is used to generate, store, update, and call the typical scenario database. The typical scenario database stores data related to each typical running scenario and the corresponding scenario triggering conditions. The scenario matching module is used to calculate the matching degree between the current operating section and each typical operating scenario based on the relevant data of the current distribution network operating section, and select the top K typical operating scenarios as scenarios to be simulated. The runtime status simulation module is used to perform multi-time period runtime status simulations based on the scenario to be simulated. Run the risk assessment module to calculate the comprehensive score of risk indicators, comprehensive risk value, comprehensive risk level, and candidate disposal strategies; The early warning output module is used to output risk warning results and operational handling suggestions sorted by comprehensive score.

16. A distribution network operation risk simulation system considering distributed generation, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein the processor executes the computer program to implement the distribution network operation risk simulation method considering distributed power sources as described in any one of claims 1 to 14.

17. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for calculating the operation risk of a distribution network taking into account distributed power sources as described in any one of claims 1 to 14.

Images