Novel electric power infrastructure toughness evaluation method and system under extreme disaster
By acquiring and processing power infrastructure data and meteorological data, combining factor analysis and risk prediction models, a set of fault scenarios is generated, which solves the problem of insufficient applicability of power infrastructure evaluation methods under extreme disasters in the existing technology, and improves the evaluation results. accuracy.
Patent Information
- Application Number
- CN202411038367.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-12-03
AI Technical Summary
The prior art has weak applicability to the resilience assessment method for power infrastructure in extreme disasters, resulting in low accuracy of the assessment results.
By obtaining power infrastructure equipment data and historical meteorological data, performing factor analysis and dimensionality reduction processing, inputting risk prediction models for prediction, combining the fault probability model to generate a set of fault scenarios, and evaluating power infrastructure based on resilience evaluation indicators under extreme disasters .
It improves the accuracy of the resilience assessment of power infrastructure in extreme disasters and enhances the reliability and applicability of the assessment results.
Smart Images

Figure CN119067288A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power system resilience assessment, and in particular to a new type of power infrastructure resilience assessment method and system under extreme disasters. Background Art
[0002] As an important part of power generation and transmission in modern society, power infrastructure faces huge challenges in the face of extreme disasters (such as hurricanes, earthquakes, floods, ice and snow disasters, etc.). With the large-scale access of renewable energy in new power systems, the widespread application of power electronic equipment, and the continuous improvement of the intelligence level of power grids, their structure and operating characteristics have undergone profound changes, making the resilience assessment of power infrastructure under extreme disasters more complicated.
[0003] The current existing assessment methods mainly establish a series of indicators for a type of extreme events to quantitatively or qualitatively assess the resilience of new power systems, such as system reliability indicators (such as failure rate, average power outage time, etc.), resilience indicators (such as recovery time, recovery speed, etc.) and adaptability indicators. Or by considering the probability and intensity of disasters and the probability of failure of new power system components, Monte Carlo simulation and other methods are used to assess the performance and risks of the system under different disaster scenarios. The assessment methods obtained by the above methods are less applicable when facing other natural disasters, resulting in low accuracy of the assessment results. Summary of the invention
[0004] In order to solve the above technical problems, an embodiment of the present invention provides a method and system for evaluating the resilience of a new type of power infrastructure under extreme disasters, which can realize the evaluation of the new type of power infrastructure, thereby improving the accuracy of the evaluation results.
[0005] A first aspect of an embodiment of the present invention provides a new method for assessing the resilience of power infrastructure under extreme disasters, the method comprising:
[0006] Obtaining power infrastructure equipment data of the power infrastructure to be assessed and historical meteorological data of the area where the power infrastructure to be assessed is located;
[0007] Determine whether the historical meteorological data and the power infrastructure equipment data can pass the factor analysis. If they can, perform dimensionality reduction processing on the historical meteorological data and the power infrastructure equipment data to obtain the reduced-dimensional data, and input the reduced-dimensional data into the risk prediction model for prediction to obtain the risk prediction results;
[0008] According to the risk prediction results and the constructed failure probability model, the failure probability model is sampled to generate a set of failure scenarios for the power infrastructure to be evaluated. The power infrastructure to be evaluated is evaluated based on the constructed resilience assessment indicators under extreme disasters and the set of failure scenarios to obtain the resilience assessment results.
[0009] In a possible implementation of the first aspect, determining whether the historical meteorological data and the power infrastructure equipment data can pass factor analysis, and if so, performing dimensionality reduction processing on the historical meteorological data and the power infrastructure equipment data to obtain dimensionality reduced data includes:
[0010] Calculate the correlation coefficients between the historical meteorological data and the power infrastructure equipment data, and obtain a correlation coefficient matrix based on the correlation coefficients;
[0011] The covariance matrix is obtained by using the correlation coefficient matrix and the inverse matrix of the correlation coefficient. The KMO test result and the Bartlett sphericity test result are calculated based on the covariance matrix. If the KMO test result is greater than the first preset value and the Bartlett sphericity test result is greater than the second preset value, the factor analysis passes and the historical meteorological data and power infrastructure equipment data are subjected to dimensionality reduction processing to obtain the reduced-dimensional data.
[0012] In a possible implementation manner of the first aspect, the risk prediction model is obtained by training based on a radial basis function neural network, including:
[0013] Obtain historical meteorological sample data and power infrastructure equipment sample data;
[0014] Determine whether the historical meteorological sample data and the power infrastructure equipment sample data can pass the factor analysis. If they can, perform dimensionality reduction processing on the historical meteorological data and the power infrastructure equipment data to obtain the reduced-dimensional data. Input the reduced-dimensional data into the initial risk prediction model for prediction to obtain the prediction results. The activation function of the initial risk prediction model is:
[0015]
[0016] In the formula, c i represents the neurons in the hidden layer, σ is the variance of the Gaussian function, x p is the pth sample data, ||x p -c i || is the Euclidean norm;
[0017] According to the prediction results, the error is calculated using the loss function, and the error and the initial value are used as the individual fitness value. The fitness function is used to select the individual fitness value to obtain the optimal fitness value. The optimal fitness value is determined as the weight. The initial risk prediction model is adjusted according to the weight, and the adjusted initial risk prediction model is trained using other historical meteorological sample data until the obtained weight meets the expected value, and the trained risk prediction model is obtained. The expression of the fitness function is:
[0018]
[0019] In the formula, y i represents the prediction result output by the i-th node, o i Represents the true value.
[0020] In a possible implementation manner of the first aspect, the fault probability model includes a power subsystem fault probability model, a natural gas subsystem fault probability model, and a thermal subsystem fault probability model, wherein the power subsystem fault probability model is:
[0021]
[0022] In the formula, represents the total failure probability of the power subsystem, k represents the weight coefficient, represents the failure probability of the jth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and j represents the jth power system component;
[0023] The failure probability model of the natural gas subsystem is:
[0024]
[0025] In the formula, represents the total failure probability of the natural gas subsystem, q represents the weight coefficient, represents the failure probability of the gth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and g represents the jth natural gas system component;
[0026] The failure probability model of the thermal subsystem is:
[0027]
[0028] In the formula, represents the total failure probability of the natural gas subsystem, l represents the weight coefficient, represents the failure probability of the hth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and h represents the hth thermal system component.
[0029] In a possible implementation of the first aspect, the resilience assessment index under extreme disasters constructed includes an optimal load reduction model and a power grid resilience index, wherein the optimal load reduction model is:
[0030]
[0031] In the formula, f j 、f g 、f h represents the reduction in electrical load, natural gas load, and thermal load at the jth, gth, and hth nodes, respectively. jmax 、f gmax 、f hmax Respectively represent the upper limit of the active output of the jth, gth, and hth nodes, f jmin 、f gmin 、f hmin Respectively represent the lower limit of the active output of the jth, gth, and hth nodes;
[0032] The grid resilience indicators are:
[0033]
[0034] Where P f (t) is the probability of load loss, p(t) is the failure probability of the power infrastructure to be evaluated at time t, and f(t) is the ideal load curve under normal operation.
[0035] A second aspect of an embodiment of the present invention provides a new power infrastructure resilience assessment system under extreme disasters, including an acquisition module, a prediction module and an assessment module.
[0036] The acquisition module is used to acquire the power infrastructure equipment data of the power infrastructure to be evaluated and the historical meteorological data of the area where the power infrastructure to be evaluated is located;
[0037] The prediction module is used to determine whether the historical meteorological data and the power infrastructure equipment data can pass the factor analysis. If they can, the historical meteorological data and the power infrastructure equipment data are subjected to dimensionality reduction processing to obtain the reduced-dimensional data, and the reduced-dimensional data are input into the risk prediction model for prediction to obtain the risk prediction results;
[0038] The evaluation module is used to sample the fault probability model according to the risk prediction results and the constructed fault probability model, generate a set of fault scenarios for the power infrastructure to be evaluated, evaluate the power infrastructure to be evaluated based on the constructed resilience assessment indicators under extreme disasters and the set of fault scenarios, and obtain the resilience assessment results.
[0039] In a possible implementation of the second aspect, it is determined whether the historical meteorological data and the power infrastructure equipment data can pass the factor analysis, and if so, the historical meteorological data and the power infrastructure equipment data are subjected to dimensionality reduction processing to obtain the reduced-dimensionality data, including:
[0040] Calculate the correlation coefficients between the historical meteorological data and the power infrastructure equipment data, and obtain a correlation coefficient matrix based on the correlation coefficients;
[0041] The covariance matrix is obtained by using the correlation coefficient matrix and the inverse matrix of the correlation coefficient. The KMO test result and the Bartlett sphericity test result are calculated based on the covariance matrix. If the KMO test result is greater than the first preset value and the Bartlett sphericity test result is greater than the second preset value, the factor analysis passes and the historical meteorological data and power infrastructure equipment data are subjected to dimensionality reduction processing to obtain the reduced-dimensional data.
[0042] In a possible implementation manner of the second aspect, the risk prediction model is obtained by training based on a radial basis function neural network, including:
[0043] Obtain historical meteorological sample data and power infrastructure equipment sample data;
[0044] Determine whether the historical meteorological sample data and the power infrastructure equipment sample data can pass the factor analysis. If they can, perform dimensionality reduction processing on the historical meteorological data and the power infrastructure equipment data to obtain the reduced-dimensional data. Input the reduced-dimensional data into the initial risk prediction model for prediction to obtain the prediction results. The activation function of the initial risk prediction model is:
[0045]
[0046] In the formula, c i represents the neurons in the hidden layer, σ is the variance of the Gaussian function, x p is the pth sample data, ||x p -c i || is the Euclidean norm;
[0047] According to the prediction results, the error is calculated using the loss function, and the error and the initial value are used as the individual fitness value. The fitness function is used to select the individual fitness value to obtain the optimal fitness value. The optimal fitness value is determined as the weight. The initial risk prediction model is adjusted according to the weight, and the adjusted initial risk prediction model is trained using other historical meteorological sample data until the obtained weight meets the expected value, and the trained risk prediction model is obtained. The expression of the fitness function is:
[0048]
[0049] In the formula, y i represents the prediction result output by the i-th node, o i Represents the true value.
[0050] In a possible implementation of the second aspect, the fault probability model includes a power subsystem fault probability model, a natural gas subsystem fault probability model, and a thermal subsystem fault probability model, wherein the power subsystem fault probability model is:
[0051]
[0052] In the formula, represents the total failure probability of the power subsystem, k represents the weight coefficient, represents the failure probability of the jth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and j represents the jth power system component;
[0053] The failure probability model of the natural gas subsystem is:
[0054]
[0055] In the formula, represents the total failure probability of the natural gas subsystem, q represents the weight coefficient, represents the failure probability of the gth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and g represents the jth natural gas system component;
[0056] The failure probability model of the thermal subsystem is:
[0057]
[0058] In the formula, represents the total failure probability of the natural gas subsystem, l represents the weight coefficient, represents the failure probability of the hth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and h represents the hth thermal system component.
[0059] In a possible implementation of the second aspect, the resilience assessment index under extreme disasters constructed includes an optimal load reduction model and a power grid resilience index, wherein the optimal load reduction model is:
[0060]
[0061] In the formula, f j 、f g 、f h represents the reduction in electrical load, natural gas load, and thermal load at the jth, gth, and hth nodes, respectively. jmax 、f gmax 、fhmax Respectively represent the upper limit of the active output of the jth, gth, and hth nodes, f jmin 、f gmin 、f hmin Respectively represent the lower limit of the active output of the jth, gth, and hth nodes;
[0062] The grid resilience indicators are:
[0063]
[0064] Where P f (t) is the probability of load loss, p(t) is the failure probability of the power infrastructure to be evaluated at time t, and f(t) is the ideal load curve under normal operation.
[0065] The method for assessing the resilience of new power infrastructure under extreme disasters provided by an embodiment of the present invention obtains the power infrastructure equipment data of the power infrastructure to be assessed and the historical meteorological data of the area where the power infrastructure to be assessed is located; determines whether the historical meteorological data and the power infrastructure equipment data can pass the factor analysis, and if so, performs dimensionality reduction processing on the historical meteorological data and the power infrastructure equipment data to obtain the reduced-dimensional data, and inputs the reduced-dimensional data into the risk prediction model for prediction to obtain the risk prediction result; based on the risk prediction result and the constructed fault probability model, samples the fault probability model to generate a set of fault scenarios for the power infrastructure to be assessed, and assesses the power infrastructure to be assessed based on the constructed resilience assessment index and fault scenario set under extreme disasters to obtain the resilience assessment result. Through the above method, it is possible to evaluate the new power infrastructure, thereby improving the accuracy of the evaluation results. BRIEF DESCRIPTION OF THE DRAWINGS
[0066] Figure 1 : A schematic diagram of a resilience assessment method flow chart of an embodiment of a new type of power infrastructure resilience assessment method under extreme disasters provided by the present invention;
[0067] Figure 2 : RBF neural network structure diagram of an embodiment of a new power infrastructure resilience assessment method under extreme disasters provided by the present invention;
[0068] Figure 3 : A system block diagram of an embodiment of a method for assessing the resilience of a new type of power infrastructure under extreme disasters provided by the present invention. DETAILED DESCRIPTION
[0069] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0070] Please refer to Figure 1 , which is a flow chart of an embodiment of a method for evaluating the resilience of a new type of power infrastructure under extreme disasters provided by an embodiment of the present invention, including steps S101 to S103, each of which is specifically as follows:
[0071] S101. Obtain power infrastructure equipment data of the power infrastructure to be evaluated and historical meteorological data of the area where the power infrastructure to be evaluated is located.
[0072] In this embodiment, the resilience of the new power infrastructure is evaluated based on the power infrastructure equipment data and meteorological disaster parameters. The power infrastructure equipment data of the power infrastructure to be evaluated and the historical meteorological data of the area where the power infrastructure to be evaluated is located are first obtained. The power infrastructure equipment data includes the equipment usage time and equipment information, and the equipment information includes the capacity of the substation equipment, the length of the line, etc.
[0073] It should be noted that the new power infrastructure refers to the power system that deeply integrates new digital technologies with traditional technologies, that is, the power infrastructure. The power infrastructure equipment data and historical meteorological data are selected from the preset time period as needed. This application gives priority to data within 1 year.
[0074] S102. Determine whether the historical meteorological data and the power infrastructure equipment data can pass the factor analysis. If so, perform dimensionality reduction processing on the historical meteorological data and the power infrastructure equipment data to obtain the reduced-dimensional data, and input the reduced-dimensional data into the risk prediction model for prediction to obtain the risk prediction result.
[0075] In this embodiment, extreme weather forecasting is performed by using a risk prediction model constructed by analyzing meteorological data and fault probability data. Factor analysis can aggregate numerous complex variables into a few independent common factors with minimal or no loss of original data information. These common factors can reflect the main information of the original numerous variables, reducing the number of variables while reflecting the intrinsic connections between the variables.
[0076] Therefore, by exploring the correlation coefficient matrix between historical meteorological data and power infrastructure equipment data, the data is grouped according to the size of the correlation, so that the correlation between data in the same group is higher and the correlation between data in different groups is lower, and the new variables representing the basic structure of each group of data are called common factors. It is judged whether the historical meteorological data and power infrastructure equipment data can pass the factor analysis. If they pass, the historical meteorological data and power infrastructure equipment data are subjected to dimensionality reduction processing to obtain the reduced-dimensional data, and the reduced-dimensional data is input into the risk prediction model for prediction to obtain the risk prediction results.
[0077] In some embodiments, in step S102, "determining whether the historical meteorological data and the power infrastructure equipment data can pass the factor analysis, and if so, performing dimensionality reduction processing on the historical meteorological data and the power infrastructure equipment data to obtain the reduced dimensionality data" includes but is not limited to the following steps:
[0078] Calculate the correlation coefficients between the historical meteorological data and the power infrastructure equipment data, and obtain a correlation coefficient matrix based on the correlation coefficients;
[0079] The covariance matrix is obtained by using the correlation coefficient matrix and the inverse matrix of the correlation coefficient. The KMO test result and the Bartlett sphericity test result are calculated based on the covariance matrix. If the KMO test result is greater than the first preset value and the Bartlett sphericity test result is greater than the second preset value, the factor analysis passes and the historical meteorological data and power infrastructure equipment data are subjected to dimensionality reduction processing to obtain the reduced-dimensional data.
[0080] In this embodiment, to determine whether the original data are suitable for factor analysis, first calculate the correlation coefficients between the various data in the historical meteorological data and the power infrastructure equipment data, and obtain the correlation coefficient matrix according to the various correlation coefficients, and then use KMO and Bart l ett's test to determine whether it is suitable for factor analysis. The closer the KMO value is to 1, the better it is. Generally, it is set to be suitable for factor analysis if it is greater than 0.5. The closer the Bart l ett's test value is to 0, the more suitable it is for factor analysis. Then, the principal component analysis method is used to estimate the factor load matrix, and then the variance contribution and variance contribution rate of the common factor are calculated to obtain the common factor. The rotation component matrix is calculated according to the common factor, and the sampling regression method is used to calculate the factor to obtain the coefficient, and the final common factor is obtained. The data corresponding to the common factor is the data after dimensionality reduction.
[0081] In some embodiments, the “risk prediction model is obtained by training based on a radial basis function neural network” in step S102 includes but is not limited to the following steps:
[0082] Obtain historical meteorological sample data and power infrastructure equipment sample data;
[0083] Determine whether the historical meteorological sample data and the power infrastructure equipment sample data can pass the factor analysis. If they can, perform dimensionality reduction processing on the historical meteorological data and the power infrastructure equipment data to obtain the reduced-dimensional data. Input the reduced-dimensional data into the initial risk prediction model for prediction to obtain the prediction results. The activation function of the initial risk prediction model is:
[0084]
[0085] In the formula, c i represents the neurons in the hidden layer, σ is the variance of the Gaussian function, x p is the pth sample data, ||x p -c i || is the Euclidean norm;
[0086] According to the prediction results, the error is calculated using the loss function, and the error and the initial value are used as the individual fitness value. The fitness function is used to select the individual fitness value to obtain the optimal fitness value. The optimal fitness value is determined as the weight. The initial risk prediction model is adjusted according to the weight, and the adjusted initial risk prediction model is trained using other historical meteorological sample data until the obtained weight meets the expected value, and the trained risk prediction model is obtained. The expression of the fitness function is:
[0087]
[0088] In the formula, y i represents the prediction result output by the i-th node, o i Represents the true value.
[0089] In this embodiment, the RBF neural network has good generalization ability, and has the advantages of simple model and fast training speed. Figure 2 As shown in the figure, the RBF neural network consists of an input layer, a hidden layer and an output layer. There is no weight connection from the input layer to the hidden layer, but there is a weight connection from the hidden layer to the output layer. The neurons are radial basis functions. There are two types of parameters that need to be adjusted in the RBF neural network. One is the center of the radial basis function, and the other is the weight from the hidden layer to the output layer. During training, historical meteorological sample data and power infrastructure equipment sample data are obtained;
[0090] Determine whether the historical meteorological sample data and the power infrastructure equipment sample data can pass the factor analysis. If they can, perform dimensionality reduction processing on the historical meteorological data and the power infrastructure equipment data to obtain the reduced-dimensional data. Input the reduced-dimensional data into the initial risk prediction model for prediction to obtain the prediction results. The activation function of the initial risk prediction model is:
[0091]
[0092] In the formula, c i represents the neurons in the hidden layer, σ is the variance of the Gaussian function, x p is the pth sample data, ||x p -c i || is the Euclidean norm;
[0093] According to the prediction results, the error is calculated using the loss function, and the error and the initial value are used as the individual fitness value. The fitness function is used to select the individual fitness value to obtain the optimal fitness value. The optimal fitness value is determined as the weight. The initial risk prediction model is adjusted according to the weight, and the adjusted initial risk prediction model is trained using other historical meteorological sample data until the obtained weight meets the expected value, and the trained risk prediction model is obtained. The expression of the fitness function is:
[0094]
[0095] In the formula, y i represents the prediction result output by the i-th node, o i Represents the true value.
[0096] S103. According to the risk prediction results and the constructed fault probability model, the fault probability model is sampled to generate a set of fault scenarios for the power infrastructure to be evaluated. The power infrastructure to be evaluated is evaluated based on the constructed resilience assessment indicators under extreme disasters and the set of fault scenarios to obtain a resilience assessment result.
[0097] In this embodiment, according to the risk prediction results and the constructed fault probability model, the fault probability model is sampled to generate a set of fault scenarios for the power infrastructure to be evaluated. The power infrastructure to be evaluated is evaluated based on the constructed resilience assessment indicators under extreme disasters and the set of fault scenarios to obtain a resilience assessment result.
[0098] In some embodiments, the “failure probability model” in step S103 includes but is not limited to the following steps:
[0099] The fault probability model includes the power subsystem fault probability model, the natural gas subsystem fault probability model and the thermal subsystem fault probability model. The power subsystem fault probability model is:
[0100]
[0101] In the formula, represents the total failure probability of the power subsystem, k represents the weight coefficient, represents the failure probability of the jth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and j represents the jth power system component;
[0102] The failure probability model of the natural gas subsystem is:
[0103]
[0104] In the formula, represents the total failure probability of the natural gas subsystem, q represents the weight coefficient, represents the failure probability of the gth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and g represents the jth natural gas system component;
[0105] The failure probability model of the thermal subsystem is:
[0106]
[0107] In the formula, represents the total failure probability of the natural gas subsystem, l represents the weight coefficient, represents the failure probability of the hth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and h represents the hth thermal system component.
[0108] In this embodiment, based on risk assessment prediction, various meteorological data under real-time extreme weather are determined, and the failure probability of the new power infrastructure caused by the changes in the new power infrastructure caused by extreme weather is determined. The failure probability model includes a power subsystem failure probability model, a natural gas subsystem failure probability model, and a thermal subsystem probability model, wherein the power subsystem failure probability model is:
[0109]
[0110] In the formula, represents the total failure probability of the power subsystem, k represents the weight coefficient, represents the failure probability of the jth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and j represents the jth power system component;
[0111] The failure probability model of the natural gas subsystem is:
[0112]
[0113] In the formula, represents the total failure probability of the natural gas subsystem, q represents the weight coefficient, represents the failure probability of the gth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and g represents the jth natural gas system component;
[0114] The failure probability model of the thermal subsystem is:
[0115]
[0116] In the formula, represents the total failure probability of the natural gas subsystem, l represents the weight coefficient, represents the failure probability of the hth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and h represents the hth thermal system component.
[0117] In some embodiments, the “toughness evaluation index” in step S103 includes but is not limited to the following steps:
[0118] The resilience assessment index under extreme disasters constructed includes an optimal load reduction model and a power grid resilience index, wherein the optimal load reduction model is:
[0119]
[0120]
[0121] In the formula, f j 、f g 、f h represents the reduction in electrical load, natural gas load, and thermal load at the jth, gth, and hth nodes, respectively. jmax 、f gmax 、f hmax Respectively represent the upper limit of the active output of the jth, gth, and hth nodes, f jmin 、f gmin 、f hmin Respectively represent the lower limit of the active output of the jth, gth, and hth nodes;
[0122] The grid resilience indicators are:
[0123]
[0124] Where P f (t) is the probability of load loss, p(t) is the failure probability of the power infrastructure to be evaluated at time t, and f(t) is the ideal load curve under normal operation.
[0125] In this embodiment, the present application decouples the IES optimization problem into the load reduction optimization of each energy subnet, and the optimal load reduction model is:
[0126]
[0127] In the formula, f j 、f g 、f hrepresents the reduction in electrical load, natural gas load, and thermal load at the jth, gth, and hth nodes, respectively. jmax 、f gmax 、f hmax Respectively represent the upper limit of the active output of the jth, gth, and hth nodes, f jmin 、f gmin 、f hmin Respectively represent the lower limit of the active output of the jth, gth, and hth nodes;
[0128] The grid resilience indicators are:
[0129]
[0130] Where P f (t) is the probability of load loss, p(t) is the failure probability of the power infrastructure to be evaluated at time t, and f(t) is the ideal load curve under normal operation.
[0131] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.
[0132] In some embodiments, Figure 3 As shown, it shows a block diagram of a new power infrastructure resilience assessment system 300 under extreme disasters provided by an embodiment of the present application, including: an acquisition module 301, a prediction module 302 and an assessment module 303, wherein:
[0133] The acquisition module is used to acquire the power infrastructure equipment data of the power infrastructure to be evaluated and the historical meteorological data of the area where the power infrastructure to be evaluated is located;
[0134] The prediction module is used to determine whether the historical meteorological data and the power infrastructure equipment data can pass the factor analysis. If they can, the historical meteorological data and the power infrastructure equipment data are subjected to dimensionality reduction processing to obtain the reduced-dimensional data, and the reduced-dimensional data are input into the risk prediction model for prediction to obtain the risk prediction results;
[0135] The evaluation module is used to sample the fault probability model according to the risk prediction results and the constructed fault probability model, generate a set of fault scenarios for the power infrastructure to be evaluated, evaluate the power infrastructure to be evaluated based on the constructed resilience assessment indicators under extreme disasters and the set of fault scenarios, and obtain the resilience assessment results.
[0136] The specific implementation of the new power infrastructure resilience assessment system under extreme disasters is basically the same as the specific implementation example of the new power infrastructure resilience assessment method under extreme disasters mentioned above, and will not be repeated here.
[0137] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.
[0138] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0139] The specific embodiments described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.
Claims
1. A new method for assessing the resilience of power infrastructure under extreme disasters, characterized in that: include: Obtaining power infrastructure equipment data of the power infrastructure to be evaluated and historical meteorological data of the area where the power infrastructure to be evaluated is located; Determine whether the historical meteorological data and the electric power infrastructure equipment data can pass factor analysis, and if so, perform dimensionality reduction processing on the historical meteorological data and the electric power infrastructure equipment data to obtain dimensionality-reduced data, and input the dimensionality-reduced data into a risk prediction model for prediction to obtain a risk prediction result; According to the risk prediction results and the constructed fault probability model, the fault probability model is sampled to generate a set of fault scenarios for the power infrastructure to be evaluated, and the power infrastructure to be evaluated is evaluated based on the constructed resilience assessment indicators under extreme disasters and the set of fault scenarios to obtain a resilience assessment result.
2. The method for evaluating the resilience of new power infrastructure under extreme disasters according to claim 1, characterized in that: The determining whether the historical meteorological data and the electric power infrastructure equipment data can pass the factor analysis, and if so, performing dimensionality reduction processing on the historical meteorological data and the electric power infrastructure equipment data to obtain the reduced dimensionality data, includes: Calculating the correlation coefficient between each data in the historical meteorological data and the power infrastructure equipment data, and obtaining a correlation coefficient matrix according to each of the correlation coefficients; The covariance matrix is obtained by using the correlation coefficient matrix and the inverse matrix of the correlation coefficient. The KMO test result and the Bartlett sphericity test result are calculated according to the covariance matrix. If the KMO test result is greater than a first preset value and the Bartlett sphericity test result is greater than a second preset value, the factor analysis passes, and the historical meteorological data and the power infrastructure equipment data are subjected to dimensionality reduction processing to obtain the reduced-dimensional data.
3. The method for evaluating the resilience of new power infrastructure under extreme disasters according to claim 1 is characterized in that: The risk prediction model is obtained by training based on a radial basis function neural network, and includes: Obtain historical meteorological sample data and power infrastructure equipment sample data; Determine whether the historical meteorological sample data and the electric power infrastructure equipment sample data can pass factor analysis. If they can, perform dimensionality reduction processing on the historical meteorological data and the electric power infrastructure equipment data to obtain dimensionality-reduced data, and input the dimensionality-reduced data into the initial risk prediction model for prediction to obtain a prediction result, wherein the activation function of the initial risk prediction model is: In the formula, c i represents the neurons in the hidden layer, σ is the variance of the Gaussian function, x p is the pth sample data, ||x p -c i || is the Euclidean norm; According to the prediction result, the error is calculated using the loss function, the error and the initial value are used as individual fitness values, the fitness function is used to select among the individual fitness values, the optimal fitness value is obtained, the optimal fitness value is determined as the weight, the initial risk prediction model is adjusted according to the weight, and the adjusted initial risk prediction model is trained by returning to use other historical meteorological sample data until the obtained weight meets the expected value, and a trained risk prediction model is obtained, wherein the expression of the fitness function is: In the formula, y i represents the prediction result output by the i-th node, o i Represents the true value.
4. The method for evaluating the resilience of new power infrastructure under extreme disasters according to claim 1, characterized in that: The fault probability model includes a power subsystem fault probability model, a natural gas subsystem fault probability model and a thermal subsystem fault probability model, wherein the power subsystem fault probability model is: In the formula, represents the total failure probability of the power subsystem, k represents the weight coefficient, represents the failure probability of the jth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and j represents the jth power system component; The natural gas subsystem failure probability model is: In the formula, represents the total failure probability of the natural gas subsystem, q represents the weight coefficient, represents the failure probability of the g-th component in the electronic system, β is the Weibull slope, α is the scaling parameter, and g represents the j-th natural gas system component; The failure probability model of the thermal subsystem is: In the formula, represents the total failure probability of the natural gas subsystem, l represents the weight coefficient, represents the failure probability of the hth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and h represents the hth thermal system component.
5. The method for evaluating the resilience of new power infrastructure under extreme disasters according to claim 1, characterized in that: The resilience assessment index under extreme disasters constructed includes an optimal load reduction model and a power grid resilience index, wherein the optimal load reduction model is: In the formula, f j 、f g 、f h represents the reduction in electrical load, natural gas load, and thermal load at the jth, gth, and hth nodes, respectively. jmax 、f gmax 、f hmax Respectively represent the upper limit of the active output of the jth, gth, and hth nodes, f jmin 、f gmin 、f hmin Respectively represent the lower limit of the active output of the jth, gth, and hth nodes; The grid resilience indicators are: Where P f (t) is the probability of load loss, p(t) is the failure probability of the power infrastructure to be evaluated at time t, and f(t) is the ideal load curve under normal operation.
6. A new power infrastructure resilience assessment system under extreme disasters, characterized by: Including acquisition module, prediction module and evaluation module, Wherein, the acquisition module is used to acquire the power infrastructure equipment data of the power infrastructure to be evaluated and the historical meteorological data of the area where the power infrastructure to be evaluated is located; The prediction module is used to determine whether the historical meteorological data and the power infrastructure equipment data can pass the factor analysis. If they can, the historical meteorological data and the power infrastructure equipment data are subjected to dimensionality reduction processing to obtain the reduced-dimensional data, and the reduced-dimensional data are input into the risk prediction model for prediction to obtain the risk prediction result; The evaluation module is used to sample the fault probability model according to the risk prediction result and the constructed fault probability model, generate a fault scenario set of the power infrastructure to be evaluated, evaluate the power infrastructure to be evaluated based on the constructed resilience assessment index under extreme disasters and the fault scenario set, and obtain a resilience assessment result.
7. A new type of power infrastructure resilience assessment system under extreme disasters as claimed in claim 6, characterized in that: The determining whether the historical meteorological data and the electric power infrastructure equipment data can pass the factor analysis, and if so, performing dimensionality reduction processing on the historical meteorological data and the electric power infrastructure equipment data to obtain the reduced dimensionality data, includes: Calculating the correlation coefficient between each data in the historical meteorological data and the power infrastructure equipment data, and obtaining a correlation coefficient matrix according to each of the correlation coefficients; The covariance matrix is obtained by using the correlation coefficient matrix and the inverse matrix of the correlation coefficient. The KMO test result and the Bartlett sphericity test result are calculated according to the covariance matrix. If the KMO test result is greater than a first preset value and the Bartlett sphericity test result is greater than a second preset value, the factor analysis passes, and the historical meteorological data and the power infrastructure equipment data are subjected to dimensionality reduction processing to obtain the reduced-dimensional data.
8. A new type of power infrastructure resilience assessment system under extreme disasters as claimed in claim 6, characterized in that: The risk prediction model is obtained by training based on a radial basis function neural network, and includes: Obtain historical meteorological sample data and power infrastructure equipment sample data; Determine whether the historical meteorological sample data and the electric power infrastructure equipment sample data can pass factor analysis. If they can, perform dimensionality reduction processing on the historical meteorological data and the electric power infrastructure equipment data to obtain dimensionality-reduced data, and input the dimensionality-reduced data into the initial risk prediction model for prediction to obtain a prediction result, wherein the activation function of the initial risk prediction model is: In the formula, c i represents the neurons in the hidden layer, σ is the variance of the Gaussian function, x p is the pth sample data, ||x p -c i || is the Euclidean norm; According to the prediction result, the error is calculated using the loss function, the error and the initial value are used as individual fitness values, the fitness function is used to select among the individual fitness values, the optimal fitness value is obtained, the optimal fitness value is determined as the weight, the initial risk prediction model is adjusted according to the weight, and the adjusted initial risk prediction model is trained by returning to use other historical meteorological sample data until the obtained weight meets the expected value, and a trained risk prediction model is obtained, wherein the expression of the fitness function is: In the formula, y i represents the prediction result output by the i-th node, o i Represents the true value.
9. A new power infrastructure resilience assessment system under extreme disasters as claimed in claim 6, characterized in that: The fault probability model includes a power subsystem fault probability model, a natural gas subsystem fault probability model and a thermal subsystem fault probability model, wherein the power subsystem fault probability model is: In the formula, represents the total failure probability of the power subsystem, k represents the weight coefficient, represents the failure probability of the jth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and j represents the jth power system component; The natural gas subsystem failure probability model is: In the formula, represents the total failure probability of the natural gas subsystem, q represents the weight coefficient, represents the failure probability of the g-th component in the electronic system, β is the Weibull slope, α is the scaling parameter, and g represents the j-th natural gas system component; The failure probability model of the thermal subsystem is: In the formula, represents the total failure probability of the thermal subsystem, l represents the weight coefficient, represents the failure probability of the hth component in the electronic system, β is the Weibull slope, α is the scaling parameter, and h represents the hth thermal system component.
10. A new type of power infrastructure resilience assessment system under extreme disasters as claimed in claim 6, characterized in that: The resilience assessment index under extreme disasters constructed includes an optimal load reduction model and a power grid resilience index, wherein the optimal load reduction model is: In the formula, f j 、f g 、f h represents the reduction in electrical load, natural gas load, and thermal load at the jth, gth, and hth nodes, respectively. jmax 、f gmax 、f hmax Respectively represent the upper limit of the active output of the jth, gth, and hth nodes, f jmin 、f gmin 、f hmin Respectively represent the lower limit of the active output of the jth, gth, and hth nodes; The grid resilience indicators are: Where P f (t) is the probability of load loss, p(t) is the failure probability of the power infrastructure to be evaluated at time t, and f(t) is the ideal load curve under normal operation.
