A power distribution network operation reliability evaluation method and system considering multi-dimensional uncertainty

Abstract

The present application relates to the technical field of power distribution network evaluation, and more particularly to a power distribution network operation reliability evaluation method and system considering multi-dimensional uncertainty, the method comprising: establishing a renewable energy generation uncertainty model; establishing a demand response interval model of residential users and constructing an operation scenario for reliability evaluation; constructing a hybrid uncertainty power flow algorithm, using an MP model to process the correlation between interval variables, obtaining the operation state response result of the power distribution network under each operation scenario, identifying the minimum cut set of the power distribution network, and determining the hard failure state; calculating the violation probability of node operation constraints and determining the soft failure state; combining the determination results of the hard failure and the soft failure to perform reliability quantitative analysis on the power distribution network and obtain the operation reliability evaluation result. Through the present application, the problem of incomplete characterization of power distribution network operation failure modes under the conditions of high proportion of renewable energy and demand response participation is effectively solved, and the accuracy of power distribution network reliability evaluation is improved.

Description

Technical Field

[0001] This invention relates to the field of distribution network assessment technology, and in particular to a method and system for assessing the operational reliability of distribution networks that takes into account multidimensional uncertainties. Background Technology

[0002] With a high proportion of distributed renewable energy sources such as wind and solar power being integrated into the distribution network, the operating status of the distribution network exhibits significant uncertainties. In actual operation, the fluctuations in renewable energy output and the changes in user load with real-time electricity prices work together to cause the distribution network to operate in a state between "normal operation" and "power outage" under many conditions. That is, no physical faults occur in system components, but operational constraints such as node voltages may be violated. This type of operational risk is usually ignored in traditional reliability assessments, making it difficult for the assessment results to reflect the true operational reliability level of the distribution network under multidimensional uncertainties.

[0003] Existing distribution network reliability assessment methods primarily focus on load point outages caused by component failures, emphasizing the assessment of outage risk through methods such as minimum cut sets, and simply classifying system operating states into "failure" or "non-failure." This assessment approach fails to simultaneously account for operational fluctuations caused by renewable energy output and demand response loads, and also lacks a systematic characterization of the probability of operational constraint violations. Consequently, in distribution networks with a high proportion of renewable energy and extensive demand response participation, the reliability assessment results exhibit significant bias. Summary of the Invention

[0004] This invention provides a method for evaluating the operational reliability of distribution networks that takes into account multidimensional uncertainties, which can effectively solve the problems in the background art.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for assessing the operational reliability of a distribution network that takes into account multidimensional uncertainties, the method comprising: An uncertainty model for renewable energy generation is established, which includes multidimensional noise affine models for wind power generation and photovoltaic power generation. Based on the load price elasticity coefficient matrix of real-time electricity prices, a demand response interval model for residential users is established, and an operational scenario for reliability assessment is constructed. Based on the aforementioned renewable energy uncertainty model and operating scenarios, a hybrid uncertainty power flow algorithm considering source-load correlation is constructed. The MP model is used to process the correlation between interval variables to obtain the operating status response results of the distribution network under various operating scenarios. Based on the operational response results, the minimum cut set of the distribution network is identified, and the unrecoverable operational state caused by component failure is determined to be a hard failure. For non-hard failure operating states, calculate the probability of violation of node operating constraints, and determine the operating risk state caused by violation of operating constraints as a soft failure; Based on the determination results of hard and soft failures, a reliability quantitative analysis of the distribution network is performed to obtain the operational reliability assessment results of the distribution network under multidimensional uncertainty conditions.

[0006] Furthermore, the multidimensional noise affine model includes: ; ; in, A multidimensional affine noise model for wind power generation; The equivalent wind speed variable after introducing a multidimensional noise term; When the wind speed is Theoretical wind power output at that time; To measure wind speed; For wind speed The first derivative of the theoretical wind power at that time; These are the noise term coefficients corresponding to different wind speed disturbance factors; These are mutually independent standard noise variables. =1, 2, 3; For the additional noise term coefficient; The standard noise variable corresponding to the coefficient of the additional noise term; A multidimensional affine noise model for photovoltaic power generation; This represents the maximum test power under standard test conditions. Light intensity under standard test conditions; The power temperature coefficient; This is a reference temperature.

[0007] Furthermore, based on the load price elasticity coefficient matrix of real-time electricity prices, a demand response interval model for residential users is established, and operational scenarios for reliability assessment are constructed, including: Based on the load price elasticity coefficient matrix of real-time electricity price, the load change of residential users under the condition of participating in demand response is calculated to obtain the interval representation of the load of each node; Based on the load range representation and combined with the operating characteristics of the distribution network, the load status under different operating conditions is classified to obtain typical operating scenarios for reliability assessment.

[0008] Furthermore, the formula for calculating the load change of residential users under demand response conditions is as follows: in, The load change matrix for DR users; , ... These represent the load changes at node i at times t=1, 2, ..., T, respectively. , ... The initial loads of node i at times t=1, 2, ..., T are respectively. , ... These represent the changes in electricity prices after users participate in DR at times t=1, 2, ..., t, respectively. , ... These represent the initial electricity prices after users participate in DR at times t=1, 2, ..., t; This is the load price elasticity coefficient matrix.

[0009] Furthermore, based on the aforementioned renewable energy uncertainty model and operating scenarios, a hybrid uncertainty power flow algorithm considering source-load correlation is constructed. The MP model is used to handle the correlation between interval variables, obtaining the operating state response results of the distribution network under various operating scenarios, including: In mixed uncertainty power flow analysis, the uncertainty of renewable energy output and the uncertainty of demand response load are uniformly introduced into the power flow calculation model to form a power flow input that includes probabilistic uncertainty and interval uncertainty; Construct a correlation constraint relationship describing the correlation between interval variables, and process the correlation between interval variables using the MP model based on the correlation constraint; Under the condition of satisfying the correlation constraints, the operating response range of node voltage and branch power of the distribution network under different operating scenarios is solved as the operating state response result.

[0010] Furthermore, the correlation constraint is as follows: ; ; in, For an uncertain input variable vector; and These are the lower and upper bounds of the uncertain input variable, respectively; This is the correlation weight matrix; The center value of the input variable is uncertain; This is a change matrix used to characterize the correlation between interval variables; Let be the radius matrix determined by the upper and lower bounds of the interval; It is a unit vector.

[0011] Furthermore, based on the operational response results, the minimum cut set of the distribution network is identified, and the unrecoverable operational state caused by component failure is determined to be a hard failure, including: The network topology of the distribution network is analyzed using the search tree method to identify the minimum cut set that causes load point power outage; When a component in the minimum cut set fails, causing the corresponding load point to be physically isolated from the power supply side and unable to restore power supply through network reconfiguration, the corresponding operating state is determined to be a hard failure state.

[0012] Furthermore, for non-hard failure operating states, the probability of node operating constraints being violated is calculated, and the operating risk state caused by the violation of operating constraints is determined to be a soft failure, including: Based on the running response results, probability interval samples are generated, and the response intervals of the node running response variables are obtained; The probability-interval samples are classified according to the relationship between their response intervals and preset operating constraint thresholds; Based on the sample classification results, the lower and upper bounds of the cumulative distribution function of the running response variable are constructed, the violation probability of the node running constraints is calculated, and the running state with the violation probability of the running constraints greater than a preset threshold is determined as a soft failure state.

[0013] Furthermore, based on the determination results of hard and soft failures, a reliability quantitative analysis of the distribution network is performed to obtain the operational reliability assessment results of the distribution network under multidimensional uncertainty conditions, including: Based on the hard failure determination results, the duration of load point unavailability caused by component failure is calculated to obtain a traditional reliability index that reflects the impact of topology failure. Based on the determination result of the soft failure, the duration of voltage over-limit corresponding to the violation of node operation constraints is calculated to obtain the voltage over-limit reliability index that reflects the impact of operation risk. The traditional reliability index is integrated with the voltage over-limit reliability index to construct a comprehensive reliability index that includes the system average power outage duration and the system average voltage over-limit duration. The comprehensive reliability index is output as the operational reliability assessment result of the distribution network under multidimensional uncertainty conditions.

[0014] A distribution network operation reliability assessment system that takes into account multidimensional uncertainties, the system comprising: The uncertainty modeling module establishes an uncertainty model for renewable energy power generation, which includes multidimensional noise affine models for wind power generation and photovoltaic power generation. The demand response modeling module establishes a demand response range model for residential users based on the load price elasticity coefficient matrix of real-time electricity prices, and constructs an operational scenario for reliability assessment. The response result acquisition module constructs a hybrid uncertainty power flow algorithm that considers source-load correlation based on the renewable energy uncertainty model and operating scenarios, and uses the MP model to process the correlation between interval variables to obtain the operating status response results of the distribution network under each operating scenario. The hard failure determination module identifies the minimum cut set of the distribution network based on the operation response results and determines that the unrecoverable operating state caused by component failure is a hard failure. The soft failure determination module calculates the probability of violation of node operation constraints for non-hard failure operation states and determines the operation risk state caused by violation of operation constraints as a soft failure. The evaluation result generation module combines the determination results of hard failure and soft failure to perform a quantitative reliability analysis of the distribution network and obtain the operational reliability evaluation results of the distribution network under multidimensional uncertainty conditions.

[0015] The technical solution of this invention can achieve the following technical effects: Compared with existing technologies, this invention introduces a multi-dimensional uncertainty analysis framework in the distribution network reliability assessment process. Under a unified assessment system, it simultaneously considers the impact of renewable energy output fluctuations and demand response load changes on the distribution network's operating status. By distinguishing between unrecoverable power outages caused by component failures and operational risk states caused by violations of operational constraints, it achieves a refined characterization of distribution network operation failure modes. This invention can conduct reliability assessments of operating states under the combined effects of multiple uncertainties, considering the correlation between uncertainties on the source and load sides. This effectively avoids the problem of overly conservative assessment results caused by ignoring correlations. Furthermore, by incorporating the probability of node operational constraint violations into the reliability quantification analysis, the assessment results reflect not only the risk of power outages but also the risk of operational quality degradation. This results in distribution network operation reliability assessments that are more consistent with the actual operating conditions under high-proportion renewable energy access, providing a more reliable technical basis for distribution network planning, operation, and risk management. The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to implement it according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. Attached Figure Description

[0016] 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 recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating a method for assessing the operational reliability of distribution networks that takes into account multidimensional uncertainties; Figure 2 A flowchart illustrating the process of establishing a demand response interval model for residential users and constructing an operational scenario for reliability assessment; Figure 3 A flowchart illustrating the process of obtaining the operational status response results of the power distribution network under various operating scenarios; Figure 4 A flowchart illustrating the process of identifying the minimum cut set of a distribution network and determining that an unrecoverable operating state caused by a component failure is a hard failure. Figure 5 A flowchart illustrating the process of calculating the probability of violation of node operating constraints and determining the operating risk state caused by the violation of operating constraints as a soft failure; Figure 6 A flowchart illustrating the process of obtaining operational reliability assessment results for a power distribution network under multidimensional uncertainty conditions. Detailed Implementation

[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0020] Example 1: like Figure 1 As shown, this application provides a method for assessing the operational reliability of a distribution network that takes into account multidimensional uncertainties. The method includes: S1: Establish an uncertainty model for renewable energy generation, which includes multidimensional noise affine models for wind power and photovoltaic power generation; Specifically, to accurately characterize the multi-source and complex uncertainties on the generation side of the distribution network under conditions of high-proportion renewable energy access, a multidimensional noise affine modeling method is adopted to uniformly model the output power of wind power and photovoltaic power generation. Unlike existing technologies that only consider single random disturbances or rely on known probability distributions, this invention introduces multiple noise terms to decompose the uncertainty of renewable energy output into several disturbance components with clear physical meanings. This improves the ability to characterize complex uncertainties in actual operating environments while ensuring the simplicity of the model structure. Wind power output is modeled based on the wind speed-power characteristic curve of the wind turbine. A multidimensional noise affine model of wind power generation is constructed by performing a first-order Taylor expansion at the center value of the measured wind speed. This model uses the measured wind speed as a benchmark and introduces major uncertainties such as wind speed prediction error, terrain differences, and wind turbine wake effects as independent noise terms. The influence of various disturbances on the equivalent wind speed offset is adjusted by corresponding sensitivity coefficients. At the same time, to further enhance the model's adaptability to complex operating environments, additional noise terms are introduced to characterize higher-order disturbances or other comprehensive uncertainties that are not clearly modeled. Through the above methods, the wind power output model can not only reflect the impact of wind speed center value changes on power generation, but also characterize the superposition effect of multiple disturbance factors within the same mathematical framework, avoiding the problem of insufficient characterization of uncertainty sources in traditional single-noise models. Considering that photovoltaic power generation output is mainly affected by module temperature and solar irradiance, a multi-dimensional noise affine model with module temperature and irradiance as input variables is adopted to describe photovoltaic power generation. Specifically, based on the measured or predicted center values ​​of module temperature and irradiance, multiple noise terms are introduced to characterize the impact of temperature fluctuations, irradiance prediction errors, and environmental changes on photovoltaic power generation, and corresponding coefficients reflect the sensitivity of various disturbances to power generation. By simultaneously modeling multiple noise factors for temperature and irradiance, the photovoltaic power generation output model can more realistically reflect the output fluctuations caused by the combined effects of multiple environmental factors under complex meteorological conditions.

[0021] S2: Based on the load price elasticity coefficient matrix of real-time electricity price, establish a demand response interval model for residential users and construct an operating scenario for reliability assessment; Specifically, firstly, based on historical electricity consumption data or survey results, a load price elasticity coefficient matrix is ​​constructed to describe the sensitivity of residential users to electricity price changes at different times. The price elasticity coefficient matrix can not only reflect the direct impact of electricity price changes on the load at the same time, but also reflect the characteristics of load shifting between different time periods. When the real-time electricity price changes, the change in the load of residential users at each node is calculated based on the price elasticity coefficient matrix, and combined with the benchmark load level of the corresponding time period, the load change range after residential users participate in demand response is obtained. The load ranges of each time period are selected and combined according to preset rules to form several representative operating scenarios to cover the typical load changes that may occur in the distribution network under the participation of residential users in demand response. By adopting the operating scenario approach, the completeness of the load uncertainty description can be maintained while avoiding the increase in computational complexity caused by exhaustively listing all possible load combinations.

[0022] S3: Based on the renewable energy uncertainty model and operating scenarios, a hybrid uncertainty power flow algorithm considering source-load correlation is constructed. The MP model is used to handle the correlation between interval variables and obtain the operating status response results of the distribution network under various operating scenarios. Specifically, the output power uncertainties of wind and solar power generation are introduced into the power flow model in probabilistic form, while the load change uncertainties after residential users participate in demand response are introduced into the power flow model in interval form. This allows generation-side and load-side uncertainties to be handled within a unified power flow calculation framework. To avoid overly conservative operational state response results due to neglecting the correlations between different uncertain variables during interval uncertainty analysis, a multidimensional parallelepiped model is introduced to model the correlations between interval variables. The range of interval variables is limited by the center value, upper and lower bounds, and correlation matrix, constraining them within the correlation feasible region defined by the multidimensional parallelepiped, thus reasonably reflecting the coupling characteristics between different uncertain factors. In the mixed uncertainty power flow calculation process, for each operational scenario, under the premise of satisfying the power flow balance relationship, the uncertain input variables are solved within the corresponding probability distribution range and interval feasible region to obtain the response results of operational state variables such as voltage at each node and power at each branch of the distribution network. The operational state response results are expressed in interval or probability interval form to characterize the range of changes in the operational state of the distribution network under multidimensional uncertainty conditions.

[0023] S4: Based on the operational response results, identify the minimum cut set of the distribution network and determine that the unrecoverable operational state caused by component failure is a hard failure; Specifically, after obtaining the operational status response results of the distribution network under different operating scenarios, based on the network topology and component connection relationships, fault combinations that may lead to load point power outages are analyzed. A search tree method is used to progressively expand the component fault states and identify the minimum cut set that causes the load point to lose connectivity with the power source. During the search process, starting with single component faults and a few component combination faults, the fault states are expanded layer by layer. When a fault combination is found to cause physical isolation between the load point and the power source for the first time, this fault combination is identified as the minimum cut set of the corresponding load point, and further searches for its supersets are stopped, thus avoiding redundant calculations and ineffective expansion. Based on this, the recoverability of the identified minimum cut sets is assessed. If, after a component fault occurs, power supply to the load point cannot be restored through network reconfiguration, switching operations, or backup path switching, the operating state is determined to be an unrecoverable operating state caused by the component fault, and this operating state is identified as a hard failure. Using the above method, the risk of power outages caused by physical faults in the distribution network can be accurately identified based solely on topology and component fault characteristics, without relying on operational constraint overrun information, thus providing clear hard failure determination results for subsequent failure state analysis.

[0024] S5: For non-hard failure operating states, calculate the probability of violation of node operating constraints and determine the operating risk state caused by violation of operating constraints as a soft failure. Specifically, for operating states not classified as hard failures, the risk of violation of node operating constraints is analyzed based on the operating state response results obtained from the aforementioned mixed uncertainty power flow analysis, in order to identify operating risk states caused by constraint violations. During the analysis, operating constraints such as node voltage are used as evaluation objects. The response results of node operating states under uncertainty conditions are expressed in interval form and compared with preset operating constraint thresholds. Operating states are classified according to the relationship between the response interval and the threshold: when the operating state response interval completely satisfies the operating constraints, the corresponding state is considered a risk-free state; when the response interval completely exceeds the operating constraint threshold, the corresponding state is considered a definite violation state; when the response interval partially satisfies the operating constraints and partially exceeds the threshold, the corresponding state is considered a state with uncertain violation risk. Based on this, the probability of node operating constraint violation is calculated by statistically analyzing the frequency of different types of operating states in the uncertainty sample. Operating states with violation probabilities exceeding the preset risk threshold are classified as operating risk states caused by operating constraint violations, i.e., soft failures. Through this method, potential operating risks in the distribution network can be quantitatively identified even when components have not experienced physical failures, allowing the risk of operating constraint violations to be included in the failure determination scope.

[0025] S6: Combining the results of hard failure and soft failure assessments, a reliability quantitative analysis of the distribution network is performed to obtain the operational reliability assessment results of the distribution network under multidimensional uncertainty conditions.

[0026] Specifically, after obtaining the hard failure judgment results caused by component faults and the soft failure judgment results caused by operational constraint violations, the two types of failure states are uniformly quantified to form the operational reliability assessment results of the distribution network under multidimensional uncertainty conditions. During the quantification process, the load point outage duration corresponding to hard failures is used as an indicator to measure the risk of power outages, while the operational constraint violation duration corresponding to soft failures is used as an indicator to measure operational risk. The results of the two types of failures under different operating scenarios are statistically analyzed and summarized. Subsequently, the two types of failure quantification results are merged according to preset rules to construct a comprehensive reliability index that can simultaneously reflect power outage risk and operational risk. This comprehensive reliability index is then used as the output of the operational reliability assessment results of the distribution network under multidimensional uncertainty conditions. Through the above reliability quantification analysis method, the operational reliability level of the distribution network under uncertainty conditions can be comprehensively reflected within the same assessment framework, making the assessment results closer to actual operating characteristics.

[0027] The technical solution of this invention effectively solves the problem of incomplete characterization of distribution network operation failure modes under conditions of high proportion of renewable energy and demand response participation, and improves the accuracy of distribution network reliability assessment.

[0028] As a preferred embodiment of the above, the multidimensional noise affine model includes: ; ; in, A multidimensional affine noise model for wind power generation; The equivalent wind speed variable after introducing a multidimensional noise term; When the wind speed is Theoretical wind power output at that time; To measure wind speed; For wind speed The first derivative of the theoretical wind power at that time; These are the noise term coefficients corresponding to different wind speed disturbance factors; These are mutually independent standard noise variables. =1, 2, 3; For the additional noise term coefficient; The standard noise variable corresponding to the coefficient of the additional noise term; A multidimensional affine noise model for photovoltaic power generation; This represents the maximum test power under standard test conditions. Light intensity under standard test conditions; The power temperature coefficient; This is a reference temperature.

[0029] Specifically, based on the theoretical wind speed-power characteristic relationship of the wind turbine: in, This refers to the rated wind speed of the fan; This refers to the rated power of the fan. The affine function for wind power is constructed as follows, based on the wind speed-power coefficient: ,in To measure wind speed; , , These are noise terms introduced by forecast error, topographic differences, and wake effects, respectively, and their value ranges are as follows: ; , , This is the corresponding noise term coefficient, reflecting the degree to which the corresponding noise term causes the input wind speed to deviate from the predicted wind speed. Expanding the formula... The polynomial with the noise term, and an additional noise term added. To represent other factors contributing to the wind speed calculation error and higher-order terms of the original noise term, the affine model for wind power with multiple noises is: . The uncertainty of photovoltaic output mainly stems from fluctuations in temperature and light intensity. Therefore, a multi-noise affine model is constructed: (1) Temperature: ,in It is the affine value of the temperature noise term. It measures the ambient temperature. It is the noise term coefficient of the prediction error. The noise term coefficient is the variance between ambient temperature and component temperature. , For the corresponding noise term, the value range is: (2) Light intensity: ,in It is the affine value of the light intensity noise term. It measures the intensity of light. It is the noise term coefficient of the prediction error. For the corresponding noise term, the value range is: Since the photovoltaic output power model is linear, the above temperature affine values ​​are applied... Affine value of light intensity Substituting these values, we obtain the affine function of the photovoltaic output as follows: .

[0030] As a preferred embodiment of the above, such as Figure 2 As shown, in step S2, based on the load price elasticity coefficient matrix of real-time electricity prices, a demand response interval model for residential users is established, and an operational scenario for reliability assessment is constructed, including: S21: Based on the load price elasticity coefficient matrix of real-time electricity price, calculate the load change of residential users under the condition of participating in demand response, and obtain the interval representation of the load of each node; S22: Based on the load interval representation and combined with the operating characteristics of the distribution network, the load status under different operating conditions is classified to obtain typical operating scenarios for reliability assessment.

[0031] Specifically, firstly, based on the price elasticity relationship between real-time electricity prices and residential user loads, a load price elasticity coefficient matrix is ​​constructed. This matrix characterizes the impact of electricity price changes on residential user electricity consumption behavior over different time periods. When real-time electricity prices change, the load change of residential users under demand response conditions is calculated based on the load price elasticity coefficient matrix. The calculated load change is then superimposed on the baseline load of each node to obtain the load change range of each node under demand response, forming a node load model represented in interval form. This interval representation can simultaneously reflect the differences in user response levels and the uncertainty brought about by electricity price fluctuations, so that the node load is no longer described by a single fixed value. Subsequently, based on the obtained load interval representations of each node, and combined with the operating characteristics of the distribution network, the possible load states under different operating conditions are classified and screened. By reasonably selecting the combination method and value range of the load intervals, several representative typical operating scenarios are formed to describe the different load conditions that the distribution network may experience under demand response participation, thus providing representative operating scenario inputs for subsequent reliability assessment. This embodiment preferably selects three typical scenarios for classification modeling: Scenario 1 (typical summer day, about 90 days): characterized by active DR, high load, and high electricity price. At this time, the main assessment is the undervoltage risk caused by overload; Scenario 2 (typical winter day, about 90 days): characterized by relatively active DR and high load; Scenario 3 (typical transition season day, about 185 days): characterized by inactive DR, light load, but large fluctuations in wind and solar power. At this time, the main assessment is the overvoltage risk caused by excessive output of renewable energy.

[0032] As a preferred embodiment of the above, the formula for calculating the load change of residential users under demand response conditions is as follows: in, The load change matrix for DR users; , ... These represent the load changes at node i at times t=1, 2, ..., T, respectively. , ... The initial loads of node i at times t=1, 2, ..., T are respectively. , ... These represent the changes in electricity prices after users participate in DR at times t=1, 2, ..., t, respectively. , ... These represent the initial electricity prices after users participate in DR at times t=1, 2, ..., t; This is the load price elasticity coefficient matrix.

[0033] Specifically, the load price elasticity coefficient matrix based on real-time electricity prices Calculate the change in user load after node i participates in the demand response; finally, obtain the node load after the response. This value constitutes the input interval variable for subsequent interval power flow calculations.

[0034] As a preferred embodiment of the above, such as Figure 3 As shown, step S3 involves constructing a hybrid uncertainty power flow algorithm considering source-load correlation based on the renewable energy uncertainty model and operating scenarios. The MP model is used to handle the correlation between interval variables, obtaining the operating state response results of the distribution network under various operating scenarios, including: S31: In mixed uncertainty power flow analysis, the uncertainty of renewable energy output and the uncertainty of demand response load are uniformly introduced into the power flow calculation model to form a power flow input that includes probabilistic uncertainty and interval uncertainty; S32: Construct correlation constraints to describe the correlation between interval variables, and use the MP model to process the correlation between interval variables based on the correlation constraints; S33: Under the condition of satisfying the correlation constraint, solve the operating response range of node voltage and branch power of the distribution network under different operating scenarios, and use it as the operating state response result.

[0035] Specifically, in the mixed uncertainty power flow analysis, the uncertainty of renewable energy output and the uncertainty of demand response load are uniformly introduced into the power flow calculation model. The output uncertainty of wind power and photovoltaic power generation is represented in probabilistic form, while the load uncertainty after residential users participate in demand response is represented in interval form. This forms a power flow input model that simultaneously includes probabilistic uncertainty and interval uncertainty, allowing uncertainties on the generation side and the load side to be processed within the same power flow analysis framework. Based on this, to characterize the possible correlations between interval uncertain variables, a correlation constraint relationship describing the correlation between interval variables is constructed. The coupling characteristics between the value ranges of interval variables are characterized by a correlation matrix, and a multidimensional parallelepiped model is introduced to process the correlation between interval variables. This ensures that the values ​​of interval variables are limited to the feasible region defined by the correlation constraint, thereby avoiding the overly conservative operation response results caused by ignoring the correlation. Under the premise of satisfying the correlation constraint conditions, mixed uncertainty power flow solutions are carried out for different operation scenarios to obtain the operation response intervals of distribution network node voltage and branch power under each operation scenario. These operation response intervals are used as the operation state response results of the corresponding operation scenario to reflect the range of changes in the operation state of the distribution network under multidimensional uncertainty conditions.

[0036] As a preferred embodiment of the above, the correlation constraint is: ; ; in, For an uncertain input variable vector; and These are the lower and upper bounds of the uncertain input variable, respectively; This is the correlation weight matrix; The center value of the input variable is uncertain; This is a change matrix used to characterize the correlation between interval variables; Let be the radius matrix determined by the upper and lower bounds of the interval; It is a unit vector.

[0037] Specifically, to effectively constrain the correlation between uncertain variables in the mixed uncertainty power flow analysis process, a correlation constraint based on a multidimensional parallelepiped model is introduced. The range of values ​​for the uncertain input variable vector is first limited by its upper and lower bounds; that is, each component of the variable vector lies between its corresponding lower and upper bounds, thus ensuring that the values ​​of the uncertain variables do not exceed the physical or operational allowable range. Based on this, a linear constraint relationship including a correlation weight matrix, a change matrix, and a radius matrix is ​​constructed to describe the correlation between uncertain input variables. The uncertain input variable vector uses its center value as a reference, and the correlation weight matrix weights the degree of correlation between different variables. The change matrix characterizes the correlation structure between variables in the interval, the radius matrix is ​​determined by the upper and lower bounds of the interval, and it limits the magnitude of the variable's deviation from the center value. The unit vector is used to unify the constraint scale, ensuring that while the uncertain input variables satisfy the interval boundary conditions, their value combinations are limited to the feasible region defined by the correlation constraints. Through this correlation constraint method, the correlation between variables in the interval can be reasonably characterized while maintaining the integrity of the interval uncertainty description, avoiding overly conservative uncertainty analysis results due to neglecting correlation.

[0038] Specifically ; ; ; in , Let these be the upper and lower limits of the interval variable. This represents the weight of the interval variable.

[0039] As a preferred embodiment of the above, such as Figure 4 As shown, step S4, based on the operational response results, identifies the minimum cut set of the distribution network and determines that the unrecoverable operational state caused by component failure is a hard failure, including: S41: The network topology of the distribution network is analyzed using the search tree method to identify the minimum cut set that causes load point power failure; S42: When a component in the minimum cut set fails, causing the corresponding load point to be physically isolated from the power supply side and unable to restore power supply through network reconfiguration, the corresponding operating state is determined to be a hard failure state.

[0040] Specifically, after obtaining the operational status response results of the distribution network under different operating scenarios, a search tree method is used to systematically analyze component fault combinations that may lead to load point power outages, based on the network topology and component connection relationships of the distribution network. During the search process, a single component fault is used as the initial node, and fault combinations are expanded layer by layer according to the topological connection relationships between components. The impact of each fault combination on the power supply connectivity of the load point is judged in real time. When a fault combination first causes the load point to lose electrical connectivity with the power source, this fault combination is identified as the minimum cut set of the corresponding load point, and further searches for higher-order fault states containing this combination are terminated, thereby improving the efficiency of minimum cut set identification. After identifying the minimum cut set, the power supply recoverability when a component in the cut set fails is judged. When a component fault causes physical isolation between the load point and the power source, and power supply cannot be restored through network reconfiguration, switching operations, or backup power path switching, the corresponding operating state is determined as an unrecoverable operating state caused by a component fault and identified as a hard failure. For hard failures, the search tree method is used to identify the minimum cut set and calculate the annual outage rate. and shutdown time At each load point, the two elements of the second-order cut set are connected in parallel, and all cut sets are approximately connected in series. Taking the permanent failure rate and repair time of each cut set element, and using the series and parallel formulas, the unavailability rate of the corresponding load point is calculated. in and For the permanent failure rate and repair time of components; , These represent the annual failure rate and downtime of the parallel system, respectively.

[0041] As a preferred embodiment of the above, such as Figure 5 As shown, step S5, for non-hard failure operating states, calculates the probability of node operating constraints being violated, and determines the operating risk state caused by the violation of operating constraints as a soft failure, including: S51: Based on the running response results, generate probability interval samples and obtain the response interval of the node running response variables; S52: Classify the probability-interval samples according to the relationship between their response interval and the preset operating constraint threshold; S53: Based on the sample classification results, construct the lower and upper bounds of the cumulative distribution function of the running response variable, calculate the probability of violation of the running constraints of the node, and determine the running state with the probability of violation of the running constraints greater than the preset threshold as a soft failure state.

[0042] Specifically, after excluding hard failure operating states caused by component faults, the risk of node operating constraint violations is further analyzed based on the operating state response results obtained from hybrid uncertainty power flow analysis. First, probability-interval samples are generated according to the probabilistic and interval characteristics of uncertain input variables. Using operating response variables such as node voltage as the object, the corresponding operating response interval for each sample is determined, thus forming a probability-interval sample set that reflects the results of uncertainty propagation. Subsequently, the generated probability-interval samples are classified according to the relationship between their operating response intervals and preset operating constraint thresholds. When the operating response interval is completely within the allowable range of operating constraints, the corresponding sample is classified as a satisfied sample; when the operating response interval completely exceeds the operating constraint threshold, the corresponding sample is classified as a non-satisfied sample; when the operating response interval simultaneously contains values ​​that satisfy and violate operating constraints, the corresponding sample is classified as an ambiguous sample. After completing sample classification, based on the proportions of satisfied, unsatisfied, and ambiguous samples in the total sample size, a lower and upper bound is constructed for the cumulative distribution function of the operational response variable. The probability of violation of node operational constraints is then calculated. When the probability of violation exceeds a preset risk threshold, the corresponding operational state is classified as an operational risk state caused by constraint violation, i.e., a soft failure. This method allows for the quantitative identification of potential operational risks in the distribution network even without physical faults in components, enabling the risk of operational constraint violation to be incorporated into the failure determination process in probabilistic form.

[0043] As a preferred embodiment of the above, such as Figure 6 As shown, in step S6, combining the results of hard and soft failure determinations, a reliability quantification analysis of the distribution network is performed to obtain the operational reliability assessment results of the distribution network under multidimensional uncertainty conditions, including: S61: Based on the hard failure determination results, calculate the duration of load point unavailability caused by component failure, and obtain the traditional reliability index that reflects the impact of topology failure. S62: Based on the soft failure determination results, calculate the voltage over-limit duration corresponding to the violation of node operation constraints, and obtain the voltage over-limit reliability index that reflects the impact of operation risk. S63: Integrate traditional reliability indicators with voltage over-limit reliability indicators to construct a comprehensive reliability indicator that includes the system average power outage duration and the system average voltage over-limit duration. S64: Output the comprehensive reliability index as the result of the operational reliability assessment of the distribution network under multidimensional uncertainty conditions.

[0044] Specifically, after obtaining the determination results of hard and soft failures, a unified reliability quantification analysis is performed on the two types of failure states to form the operational reliability assessment results of the distribution network under multidimensional uncertainty conditions. In the hard failure quantification process, based on the minimum cut set analysis results, the unrecoverable power outages caused by permanent component failures at each load point under different operating scenarios are statistically analyzed. The outage duration of the load point is calculated according to the permanent failure rate and repair time of components in the cut set. By weighting and summing the outage times of different cut sets, a system average outage duration index reflecting the impact of topology faults is obtained. In the soft failure quantification process, based on the calculation results of the node operating constraint violation probability, the node voltage over-limit states are statistically analyzed. The duration corresponding to the node operating constraint violation is taken as the voltage over-limit duration, and a voltage over-limit reliability index is constructed in conjunction with the operating constraint violation probability to reflect the impact of operational risks on system reliability. On this basis, the system average outage duration index and the voltage over-limit reliability index are comprehensively processed according to a preset fusion relationship to construct a comprehensive reliability index that simultaneously includes the system average outage duration and the system average voltage over-limit duration. This comprehensive reliability index is then output as the operational reliability assessment result of the distribution network under multidimensional uncertainty conditions. By introducing the aforementioned quantitative method based on outage time and over-limit duration, the risk of component failure and the risk of violation of operating constraints can be reflected simultaneously within the same assessment framework. This ensures that the assessment results are consistent with the reliability calculation formula established in the document and are more in line with the actual operating characteristics of the distribution network.

[0045] Example 2: Based on the same inventive concept as the distribution network operation reliability assessment method considering multidimensional uncertainty in the foregoing embodiments, the present invention also provides a distribution network operation reliability assessment system considering multidimensional uncertainty, comprising: The uncertainty modeling module establishes uncertainty models for renewable energy power generation, including multidimensional noise affine models for wind power and photovoltaic power generation. The demand response modeling module establishes a demand response range model for residential users based on the load price elasticity coefficient matrix of real-time electricity prices, and constructs an operational scenario for reliability assessment. The response result acquisition module constructs a hybrid uncertainty power flow algorithm that considers source-load correlation based on the renewable energy uncertainty model and operating scenarios. It uses the MP model to process the correlation between interval variables and obtains the operating status response results of the distribution network under various operating scenarios. The hard failure determination module identifies the minimum cut set of the distribution network based on the operation response results and determines that the unrecoverable operating state caused by component failure is a hard failure. The soft failure determination module calculates the probability of violation of node operation constraints for non-hard failure operation states and determines the operation risk state caused by violation of operation constraints as a soft failure. The evaluation result generation module combines the judgment results of hard failure and soft failure to perform a quantitative analysis of the reliability of the distribution network and obtain the operational reliability evaluation results of the distribution network under multidimensional uncertainty conditions.

[0046] The evaluation system described above in this invention can effectively realize a method for evaluating the reliability of power distribution network operation that takes into account multidimensional uncertainties. The technical effects it can achieve are as described in the above embodiments, and will not be repeated here.

[0047] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of the application as defined herein, and are to be considered as covering any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Thus, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for assessing the operational reliability of a distribution network that takes into account multidimensional uncertainties, characterized in that, The method includes: An uncertainty model for renewable energy generation is established, which includes multidimensional noise affine models for wind power generation and photovoltaic power generation. Based on the load price elasticity coefficient matrix of real-time electricity prices, a demand response interval model for residential users is established, and an operational scenario for reliability assessment is constructed. Based on the aforementioned renewable energy uncertainty model and operating scenarios, a hybrid uncertainty power flow algorithm considering source-load correlation is constructed. The MP model is used to process the correlation between interval variables to obtain the operating status response results of the distribution network under various operating scenarios. Based on the operational response results, the minimum cut set of the distribution network is identified, and the unrecoverable operational state caused by component failure is determined to be a hard failure. For non-hard failure operating states, calculate the probability of violation of node operating constraints, and determine the operating risk state caused by violation of operating constraints as a soft failure; Based on the determination results of hard and soft failures, a reliability quantitative analysis of the distribution network is performed to obtain the operational reliability assessment results of the distribution network under multidimensional uncertainty conditions.

2. The method for assessing the operational reliability of a distribution network considering multidimensional uncertainties according to claim 1, characterized in that, The multidimensional noise affine model includes: ； ； in, A multidimensional affine noise model for wind power generation; The equivalent wind speed variable after introducing a multidimensional noise term; When the wind speed is Theoretical wind power output at that time; To measure wind speed; For wind speed The first derivative of the theoretical wind power at that time; These are the noise term coefficients corresponding to different wind speed disturbance factors; These are mutually independent standard noise variables. =1, 2, 3; For the additional noise term coefficient; The standard noise variable corresponding to the coefficient of the additional noise term; A multidimensional affine noise model for photovoltaic power generation; This represents the maximum test power under standard test conditions. Light intensity under standard test conditions; The power temperature coefficient; This is a reference temperature.

3. The method for assessing the operational reliability of a distribution network considering multidimensional uncertainties according to claim 1, characterized in that, Based on the load price elasticity coefficient matrix of real-time electricity prices, a demand response interval model for residential users is established, and operational scenarios for reliability assessment are constructed, including: Based on the load price elasticity coefficient matrix of real-time electricity price, the load change of residential users under the condition of participating in demand response is calculated to obtain the interval representation of the load of each node; Based on the load range representation and combined with the operating characteristics of the distribution network, the load status under different operating conditions is classified to obtain typical operating scenarios for reliability assessment.

4. The method for assessing the operational reliability of a distribution network considering multidimensional uncertainties according to claim 3, characterized in that, The formula for calculating the load change of residential users under demand response conditions is: in, The load change matrix for DR users; , ... These represent the load changes at node i at times t=1, 2, ..., T, respectively. , ... The initial loads of node i at times t=1, 2, ..., T are respectively. , ... These represent the changes in electricity prices after users participate in DR at times t=1, 2, ..., t, respectively. , ... These represent the initial electricity prices after users participate in DR at times t=1, 2, ..., t; This is the load price elasticity coefficient matrix.

5. The method for assessing the operational reliability of a distribution network considering multidimensional uncertainties according to claim 1, characterized in that, Based on the aforementioned renewable energy uncertainty model and operating scenarios, a hybrid uncertainty power flow algorithm considering source-load correlation is constructed. The MP model is used to handle the correlation between interval variables, obtaining the operating state response results of the distribution network under various operating scenarios, including: In mixed uncertainty power flow analysis, the uncertainty of renewable energy output and the uncertainty of demand response load are uniformly introduced into the power flow calculation model to form a power flow input that includes probabilistic uncertainty and interval uncertainty; Construct a correlation constraint relationship describing the correlation between interval variables, and process the correlation between interval variables using the MP model based on the correlation constraint; Under the condition of satisfying the correlation constraints, the operating response range of node voltage and branch power of the distribution network under different operating scenarios is solved as the operating state response result.

6. The method for assessing the operational reliability of a distribution network considering multidimensional uncertainties according to claim 5, characterized in that, The correlation constraint is: ； ； in, For an uncertain input variable vector; and These are the lower and upper bounds of the uncertain input variable, respectively; This is the correlation weight matrix; The center value of the input variable is uncertain; This is a change matrix used to characterize the correlation between interval variables; Let be the radius matrix determined by the upper and lower bounds of the interval; It is a unit vector.

7. The method for assessing the operational reliability of a distribution network considering multidimensional uncertainties according to claim 1, characterized in that, Based on the operational response results, the minimum cut set of the distribution network is identified, and the unrecoverable operational state caused by component failure is determined to be a hard failure, including: The network topology of the distribution network is analyzed using the search tree method to identify the minimum cut set that causes load point power outage; When a component in the minimum cut set fails, causing the corresponding load point to be physically isolated from the power supply side and unable to restore power supply through network reconfiguration, the corresponding operating state is determined to be a hard failure state.

8. The method for assessing the operational reliability of a distribution network considering multidimensional uncertainties according to claim 1, characterized in that, For non-hard failure operating states, calculate the probability of node operational constraints being violated, and determine the operational risk state caused by operational constraint violation as a soft failure, including: Based on the running response results, probability interval samples are generated, and the response intervals of the node running response variables are obtained; The probability-interval samples are classified according to the relationship between their response intervals and preset operating constraint thresholds; Based on the sample classification results, the lower and upper bounds of the cumulative distribution function of the running response variable are constructed, the violation probability of the node running constraints is calculated, and the running state with the violation probability of the running constraints greater than a preset threshold is determined as a soft failure state.

9. The method for assessing the operational reliability of a distribution network considering multidimensional uncertainties according to claim 1, characterized in that, Based on the determination results of hard and soft failures, a reliability quantitative analysis of the distribution network is performed to obtain the operational reliability assessment results of the distribution network under multidimensional uncertainty conditions, including: Based on the hard failure determination results, the duration of load point unavailability caused by component failure is calculated to obtain a traditional reliability index that reflects the impact of topology failure. Based on the determination result of the soft failure, the duration of voltage over-limit corresponding to the violation of node operation constraints is calculated to obtain the voltage over-limit reliability index that reflects the impact of operation risk. The traditional reliability index is integrated with the voltage over-limit reliability index to construct a comprehensive reliability index that includes the system average power outage duration and the system average voltage over-limit duration. The comprehensive reliability index is output as the operational reliability assessment result of the distribution network under multidimensional uncertainty conditions.

10. A distribution network operation reliability assessment system that takes into account multidimensional uncertainties, characterized in that, The system includes: The uncertainty modeling module establishes an uncertainty model for renewable energy power generation, which includes multidimensional noise affine models for wind power generation and photovoltaic power generation. The demand response modeling module establishes a demand response range model for residential users based on the load price elasticity coefficient matrix of real-time electricity prices, and constructs an operational scenario for reliability assessment. The response result acquisition module constructs a hybrid uncertainty power flow algorithm that considers source-load correlation based on the renewable energy uncertainty model and operating scenarios, and uses the MP model to process the correlation between interval variables to obtain the operating status response results of the distribution network under each operating scenario. The hard failure determination module identifies the minimum cut set of the distribution network based on the operation response results and determines that the unrecoverable operating state caused by component failure is a hard failure. The soft failure determination module calculates the probability of violation of node operation constraints for non-hard failure operation states and determines the operation risk state caused by violation of operation constraints as a soft failure. The evaluation result generation module combines the determination results of hard failure and soft failure to perform a quantitative reliability analysis of the distribution network and obtain the operational reliability evaluation results of the distribution network under multidimensional uncertainty conditions.

Images