Cluster-based flexible resource runnable domain aggregation error quantification method and system

Abstract

The invention provides a cluster-based flexible resource runnable domain aggregation error quantification method and system, and the method comprises the steps: generating energy storage resource runnable domain constraints according to a pre-constructed energy storage time sequence coupling output model, and carrying out the clustering of the energy storage resource runnable domain constraints according to each energy storage resource runnable domain constraint; after an energy storage cluster aggregation polyhedron is generated based on time coupling characteristics, active power demand sampling is carried out based on a Markov chain Monte Carlo method according to the energy storage cluster aggregation polyhedron and a preset sampling rule to obtain a cluster operation power time sequence set; and carrying out aggregation error clustering analysis on the cluster operation power time sequence set to obtain aggregation error clustering results, taking the aggregation error clustering cluster with the maximum average aggregation error in the aggregation error clustering results as a to-be-analyzed clustering cluster, and analyzing the clustering center of the to-be-analyzed clustering cluster to obtain an aggregation error quantification result. According to the method, the complexity of high-dimensional space aggregation analysis can be effectively reduced, and the accuracy of energy storage flexibility aggregation error evaluation is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of demand-side resource aggregation, and in particular to a cluster-based flexible resource operational domain aggregation error quantification method and system. Background Art

[0002] The increasing volatility of distributed energy resource feed-in makes the utilization of existing flexibility resources in the power system an important measure to address the stability of power grid operation, and aggregation research based on the flexibility feasible domain becomes an important basis for exploring the potential of flexibility for effective planning and use.

[0003] The existing method of actively quantifying the flexibility potential of distribution networks is mainly to aggregate individual flexibility resources into a flexible power system, and then perform a steady-state evaluation of the feasible domain based on the linear optimal power flow. However, in practical applications, although the coupling complexity can be reduced by aggregating individual flexibility resources, it is easy to cause information loss and decomposing the flexibility demand from a single optimization problem will cause the result to deviate from the planned set point; in addition, in the process of solving a single optimization problem, not only does it require a large number of grid model constraint calculations for the power system, resulting in excessively high computational costs and low efficiency, but the evaluation of the feasible domain convex hull often ignores the time coupling constraints caused by the use of flexibility, making it difficult to accurately evaluate the aggregation error of the feasible domain. Summary of the invention

[0004] The purpose of the present invention is to provide a cluster-based flexible resource operational domain aggregation error quantification method, which generates an energy storage cluster aggregation polyhedron by aggregating the operational domains of various energy storage resources based on time coupling characteristics, and uses the Markov chain Monte Carlo method to sample the active power demand to obtain a cluster operating power time series set, and then performs a structured analysis of the solution space inside the feasible domain based on the error clustering analysis of the cluster operating power time series, which effectively reduces the complexity of high-dimensional space analysis and improves the accuracy of energy storage flexibility aggregation error evaluation.

[0005] In order to achieve the above objectives, it is necessary to provide a cluster-based flexible resource operational domain aggregation error quantification method and system in response to the above technical problems.

[0006] In a first aspect, an embodiment of the present invention provides a cluster-based flexible resource operational domain aggregation error quantification method, the method comprising the following steps: Generate energy storage resource operational domain constraints based on a pre-built energy storage time-series coupled output model; the energy storage time-series coupled output model includes an energy storage active output model and an energy storage time-coupled flexibility model; According to the operational domain constraints of each energy storage resource, the corresponding energy storage cluster aggregation polyhedron is generated based on the time coupling characteristics; According to the energy storage cluster aggregation polyhedron and preset sampling rules, active power demand sampling is performed based on the Markov chain Monte Carlo method to obtain a cluster operation power time series set; the preset sampling rule is to obtain multiple cluster operation power time series of preset days according to a preset sampling step and a preset time period sampling period; Performing aggregation error clustering analysis on the cluster operation power time series set to obtain corresponding aggregation error clustering results; the aggregation error clustering results include multiple aggregation error clustering clusters; The aggregation error cluster with the largest average aggregation error in the aggregation error clustering results is taken as the cluster to be analyzed, and the cluster center of the cluster to be analyzed is analyzed to obtain the corresponding aggregation error quantification result; the aggregation error quantification result includes the location of the aggregation abnormality area, the abnormal power demand pattern and the aggregation error.

[0007] Furthermore, the steps of constructing the energy storage active output model include: According to the active power limit of energy storage resources, energy storage charging and discharging power constraints, and energy storage charging and discharging start and stop constraints, an energy storage charging and discharging time coupling model is constructed; According to the energy storage capacity, charging efficiency, energy storage charging power and energy storage discharging power of the energy storage resources, a time-coupled model of energy storage state of charge is constructed.

[0008] Furthermore, the energy storage charging and discharging time coupling model is expressed as: in, , and They represent the active power, energy storage charging power and energy storage discharging power of the energy storage resource at time t respectively; and They represent the maximum charging power and maximum discharging power of the energy storage resource respectively; and They respectively represent the charging state and discharging state of the energy storage resource at time t.

[0009] Furthermore, the energy storage state of charge time coupling model is expressed as: in, and They respectively represent the initial state charge and initial charging state of the energy storage resource; Indicates the energy storage charge state of the energy storage resource at time t; and They represent the energy storage capacity and charging efficiency of the energy storage resources respectively; and They represent the energy storage charging power and energy storage discharging power of the energy storage resource at time t respectively; Indicates the duration interval.

[0010] Furthermore, the steps of constructing the energy storage time-coupled flexibility model include: Based on the power balance of a single flexibility resource in the power system, the energy storage time coupling flexibility model is obtained; the energy storage time coupling flexibility model is expressed as: in, , and They represent the active power, static active power and flexible active power at time t respectively.

[0011] Furthermore, the step of sampling active power demand based on the energy storage cluster aggregation polyhedron and preset sampling rules and obtaining a cluster operation power time series set based on the Markov chain Monte Carlo method includes: Obtaining the energy storage cluster operation time series data of each preset time period sampling period; the energy storage cluster operation time series data includes the energy storage cluster active power demand time series data and the operation parameter data of each energy storage resource in the cluster; According to the energy storage cluster operation time series data of each preset time period sampling period and the preset sampling step, uniform sampling is performed in the energy storage cluster aggregation polyhedron based on the Markov chain Monte Carlo method to obtain the corresponding cluster operation power time series.

[0012] Furthermore, the step of performing aggregation error cluster analysis on the cluster operation power time series set to obtain a corresponding aggregation error cluster result includes: Performing power plan output deviation analysis on each cluster operation power time series in the cluster operation power time series set respectively to obtain corresponding time series aggregation errors; According to each time series aggregation error, K-means clustering analysis is performed on the cluster operation power time series set to obtain a corresponding aggregation error clustering result.

[0013] Furthermore, the step of analyzing the cluster centers of the clusters to be analyzed to obtain corresponding aggregation error quantification results includes: Obtaining the operating power point with the largest power plan output deviation in the cluster operating power time series corresponding to the cluster center as the error abnormality sample point; The active power demand data of the energy storage cluster corresponding to the error abnormal sample point is obtained, and the feasible domain position and power mode analysis is performed on the active power demand data of the energy storage cluster to obtain the corresponding aggregate abnormal area position and abnormal power demand mode; the aggregate abnormal area position is the active power demand of the error abnormal sample point corresponding to the geometric position in the aggregate polyhedron of the energy storage cluster; the abnormal power demand mode includes a high power demand mode and a low power demand mode.

[0014] Further, the method comprises: According to the aggregate error quantification result, a corresponding flexibility resource scheduling adjustment strategy is generated; the flexibility resource scheduling adjustment strategy includes controlling storage capacity and optimizing power demand allocation.

[0015] In a second aspect, an embodiment of the present invention provides a cluster-based flexible resource executable domain aggregation error quantization system, the system comprising: A feasible domain acquisition module is used to generate energy storage resource operational domain constraints according to a pre-built energy storage time-series coupled output model; the energy storage time-series coupled output model includes an energy storage active output model and an energy storage time-coupled flexibility model; The cluster feasible domain generation module is used to generate the corresponding energy storage cluster aggregation polyhedron based on the time coupling characteristics according to the operational domain constraints of each energy storage resource; A power demand sampling module is used to perform active power demand sampling based on the Markov chain Monte Carlo method according to the energy storage cluster aggregation polyhedron and preset sampling rules to obtain a cluster operation power time series set; the preset sampling rule is to obtain multiple cluster operation power time series of preset days according to a preset sampling step and a preset time period sampling period; An aggregation error analysis module is used to perform aggregation error cluster analysis on the cluster operation power time series set to obtain a corresponding aggregation error cluster result; the aggregation error cluster result includes a plurality of aggregation error cluster clusters; The aggregation error quantification module is used to take the aggregation error clustering cluster with the largest average aggregation error in the aggregation error clustering results as the clustering cluster to be analyzed, and analyze the cluster center of the clustering cluster to be analyzed to obtain the corresponding aggregation error quantification result; the aggregation error quantification result includes the location of the aggregation abnormality area, the abnormal power demand pattern and the aggregation error.

[0016] The present invention provides a cluster-based flexible resource operational domain aggregation error quantification method and system, through which the method realizes the generation of energy storage resource operational domain constraints according to a pre-constructed energy storage time series coupling output model including an energy storage active output model and an energy storage time coupling flexibility model, and generates corresponding energy storage cluster aggregation polyhedrons based on the time coupling characteristics according to the operational domain constraints of each energy storage resource. Then, according to the energy storage cluster aggregation polyhedron and preset sampling rules, active power demand sampling is performed based on the Markov chain Monte Carlo method to obtain a cluster operation power time series set, and aggregation error clustering analysis is performed on the cluster operation power time series set to obtain an aggregation error clustering result including multiple aggregation error clustering clusters, and the aggregation error clustering cluster with the largest average aggregation error in the aggregation error clustering result is used as the cluster cluster to be analyzed, and the cluster center of the cluster cluster to be analyzed is analyzed to obtain a technical solution of the corresponding aggregation error quantification result including the location of the aggregation abnormal area, the abnormal power demand pattern and the aggregation error. Compared with the existing technology, this cluster-based flexibility resource can run domain aggregation error quantification method, which can effectively reduce the complexity of high-dimensional space analysis, while improving the accuracy of energy storage flexibility aggregation error assessment, providing a reliable basis for subsequent optimization and adjustment of energy storage resource scheduling strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of a method for quantifying cluster-based flexible resource operational domain aggregation errors in an embodiment of the present invention; Figure 2 It is another flow chart of a method for quantifying cluster-based flexible resource operational domain aggregation errors in an embodiment of the present invention; Figure 3 It is a structural schematic diagram of a cluster-based flexible resource executable domain aggregation error quantization system in an embodiment of the present invention; Figure 4 It is another structural schematic diagram of the cluster-based flexible resource executable domain aggregation error quantization system in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical scheme and beneficial effects of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. Obviously, the embodiments described below are part of the embodiments of the present invention and are only used to illustrate the present invention, but are not used to limit the scope of the present invention. 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.

[0019] The cluster-based flexible resource operational domain aggregation error quantification method provided by the present invention can be understood as a proposed method based on the current application status of the complex analysis process of the flexibility potential assessment of the existing distribution network, high computational cost, low assessment efficiency, and difficulty in accurately assessing the feasible domain aggregation error. The method generates an energy storage cluster aggregation polyhedron by aggregating the operational domains of various energy storage resources based on the time coupling characteristics, and uses the Markov chain Monte Carlo method to sample the active power demand to obtain a cluster operating power time series set, and then performs a structured analysis of the feasible domain internal solution space based on the error clustering analysis of the cluster operating power time series. The following embodiments will provide a detailed explanation of the cluster-based flexible resource operational domain aggregation error quantification method of the present invention.

[0020] In one embodiment, Figure 1 As shown, a cluster-based flexible resource operational domain aggregation error quantification method is provided, comprising the following steps: S11. Generate energy storage resource operational domain constraints based on a pre-constructed energy storage time series coupled output model; the energy storage time series coupled output model includes an energy storage active output model and an energy storage time coupled flexibility model; wherein the energy storage active output model can be understood as a model that simultaneously considers the energy storage flexibility analysis conditions and the operational constraints followed by a single flexibility resource in the power system, and is constructed based on the time series coupling analysis, and can accurately describe the energy storage resource operational constraints, preferably including an energy storage charging and discharging time coupling model and an energy storage charge state time coupling model; specifically, the steps of constructing the energy storage active output model include: According to the active power limit of energy storage resources, energy storage charging and discharging power constraints, and energy storage charging and discharging start and stop constraints, an energy storage charging and discharging time coupling model is constructed; wherein, the energy storage charging and discharging time coupling model can be understood as a mathematical expression of the active power of energy storage resources at each time point t obtained by statically restricting the energy storage resources according to the upper and lower limits of their active power, and describing the energy storage charging state and the time coupling constraints of the charge, which is expressed as: (1) (2) (3) in, , and They represent the active power, energy storage charging power and energy storage discharging power of the energy storage resource at time t respectively; and They represent the maximum charging power and maximum discharging power of the energy storage resource respectively; and They represent the charging state and discharging state of the energy storage resource at time t respectively. 1 indicates that the energy storage resource is in the corresponding state at time t, and 0 indicates that the energy storage resource is not in the corresponding state at time t.

[0021] According to the energy storage capacity, charging efficiency, energy storage charging power and energy storage discharging power of the energy storage resources, a time coupling model of the energy storage state of charge is constructed; wherein the time coupling model of the energy storage state of charge can be understood as a mathematical expression constructed based on the energy storage capacity, charging efficiency, energy storage charging power and energy storage discharging power of the energy storage resources, which is used to describe the initial state of the charge of the energy storage resources and the connection between the energy storage state of charge at adjacent operating times, and is expressed as: (4) in, and They respectively represent the initial state charge and initial charging state of the energy storage resource; is the energy storage charge state of the energy storage resource at time t; and are the energy storage capacity and charging efficiency of the energy storage resources respectively; and They represent the energy storage charging power and energy storage discharging power of the energy storage resource at time t respectively; Indicates the duration interval.

[0022] The energy storage time coupling flexibility model in this embodiment can be understood as a mathematical model based on the fact that in addition to the active output constraints of the energy storage resources themselves, the energy storage resources also need to comply with the flexibility assessment constraints of a single flexibility resource or a flexible power system (energy storage resource cluster) in the power system; specifically, the steps of constructing the energy storage time coupling flexibility model include: Based on the power balance of a single flexibility resource in the power system, the energy storage time coupling flexibility model is obtained; the energy storage time coupling flexibility model is expressed as: (5) in, , and They represent the active power, static active power and flexible active power at time t respectively.

[0023] By integrating the energy storage time-series coupled output model constructed by the above method steps, the operational domain constraints of a single energy storage resource can be obtained, which is expressed as: (6) in, and is the parameter matrix; represents the i-th operable point of the energy storage resource at time t; It means that the energy storage resource at time t is the set of all feasible operating points.

[0024] S12. According to the operational domain constraints of each energy storage resource, a corresponding energy storage cluster aggregation polyhedron is generated based on the time coupling characteristics; wherein the process of obtaining the energy storage cluster aggregation polyhedron is as follows: First, the operational domain constraints of each energy storage resource in the actual energy storage cluster are coupled in time series to obtain the time series coupled operational domain; assuming that there are n time steps, the corresponding time series coupled operational domain can be expressed as: (7) (8) in, Represents the time series coupled runnable domain; Represents a time series coupled runnable domain The dimension of Then, the operational domain constraints of each energy storage resource are aggregated in a hierarchical manner to obtain the cluster resource aggregated operational area based on the energy storage cluster.

[0025] S13. According to the energy storage cluster aggregation polyhedron and preset sampling rules, active power demand sampling is performed based on the Markov chain Monte Carlo method to obtain a cluster operation power time series set; the preset sampling rule is to obtain multiple cluster operation power time series of preset days according to the preset sampling step and the preset time period sampling cycle; wherein, the preset sampling step can be understood as the time interval between two sampling moments, and the preset time period sampling cycle can be understood as the length of the sampling period used to divide the 24 hours of a day. Assuming that the preset time period sampling cycle is 8 hours and the preset sampling step is 1 hour, there are 3 preset time period sampling cycles in a day, and each preset time period sampling cycle includes 8 sampling moments, and the sampling data corresponding to the 8 sampling moments form a time series, that is, 3 cluster operation power time series can be sampled within one day. It should be noted that the preset number of days, the preset sampling time and the preset time period sampling cycle can be determined according to actual application requirements, and are not specifically limited here.

[0026] Considering the direct impact between the feasibility of the current energy storage cluster operating point and the future planned operating point in practical applications, resulting in the feasible domain dimension of the energy storage cluster aggregation polyhedron being proportional to the time step, and the geometric shape of the aggregation polyhedron being very complex, in order to uniformly sample the energy storage cluster aggregation polyhedron, this embodiment preferably uses the Markov chain Monte Carlo method (MCMC) to perform analysis sampling within the aggregation polyhedron to obtain the energy storage cluster active power output point that meets the constraints of the above-mentioned energy storage time series coupling output model and meets the active power demand at the sampling time. Specifically, the steps of sampling the active power demand based on the Markov chain Monte Carlo method according to the energy storage cluster aggregation polyhedron and the preset sampling rules to obtain the cluster operation power time series set include: Obtain the energy storage cluster operation timing data of each preset time period sampling period; the energy storage cluster operation timing data includes the energy storage cluster active power demand timing data and the operation parameter data of each energy storage resource in the cluster; wherein, the energy storage cluster active power demand timing data can be understood as the active power demand at each sampling moment within the preset time period sampling period, and the operation parameter data includes energy storage capacity, charging and discharging efficiency and maximum charging power, etc.

[0027] According to the energy storage cluster operation time series data of each preset time period sampling period and the preset sampling step length, uniform sampling is performed within the energy storage cluster aggregation polyhedron based on the Markov chain Monte Carlo method to obtain the corresponding cluster operation power time series; wherein the cluster operation power time series can be understood as a series of energy storage cluster active output feasible points that can meet the constraints of the above energy storage time series coupled output model obtained by sampling with random length in a randomly selected direction. The specific acquisition process is as follows: 1) Select the initial solution :Based on constraints (1)-(8), a single energy storage resource feasible domain is constructed. The feasible domain of the energy storage cluster is constructed with the feasible domain set of each energy storage flexibility resource in the energy storage cluster. The feasible point of the active output of the energy storage resource in the feasible domain is randomly selected as the initial point. , randomly select a direction ; 2) Calculate the distance in the hyperplane direction ,in, represents the lower and upper limits of the change in the active output of the energy storage resources in the cluster operation power time series, that is, the output at the next moment is constrained by the output at this moment and the upper and lower limits of the output change, and needs to satisfy constraints (2) and (4), where The minimum output of each energy storage resource in formula (2) is determined by accumulating the minimum output of all energy storage resources to obtain the energy storage cluster ; Similarly, The maximum output of each energy storage resource in formula (2) and (4) is determined by accumulating the maximum output of all energy storage resources to obtain the energy storage cluster. ; 3) Select a random distance , the distance represents the gap between the feasible charging power point at this moment and the next moment, which is used to determine the active output of the energy storage cluster at the next moment; 4) Update the operational domain of the energy storage cluster The active output of the energy storage cluster at the next moment is based on the active output sampling value at this moment. The gap to the feasible point of charging power Determine, among which It is the operational domain of the energy storage cluster at the next moment.

[0028] 5) Repeat the above steps to obtain the sampling values ​​of the operable points at n moments, and obtain multiple uniform sample points in the energy storage cluster aggregation polyhedron solution space corresponding to a preset time period sampling period, and each sample point represents an operable point of the time energy storage cluster.

[0029] In actual application scenarios, the Markov chain Monte Carlo method is used to sample the feasible points of active power output of the energy storage cluster. Each sampling point represents the operating status of the energy storage cluster at different times, thereby generating a time series consisting of feasible points of effective active power output corresponding to n active power demands, and the n active power demands reflect the power demand and scheduling plan of the energy storage flexibility resources. This embodiment uses the Markov chain Monte Carlo method to perform Monte Carlo sampling on complex time series and a large number of energy storage flexibility resources in the energy storage cluster, which can effectively reduce the time complexity of sampling high-dimensional and time-coupled variables, reduce the computational cost of error evaluation and analysis, and provide reliable technical support for improving the efficiency of error evaluation and analysis.

[0030] S14. Performing aggregation error clustering analysis on the cluster operating power time series set to obtain corresponding aggregation error clustering results; wherein, the aggregation error clustering analysis can be understood as taking into account that the active power demand reflects the power demand and scheduling plan of the energy storage flexibility resources, and the various energy storage resources in the energy storage cluster are aggregated together, and the corresponding time series must be effectively coupled so that the various energy storage resources can cooperate to complete the corresponding energy storage cluster active power demand situation, and the power demand aggregation error of each cluster operating power time series is clustered, so as to reasonably allocate the active power demand based on the aggregation error analysis. Specifically, the step of performing aggregation error clustering analysis on the cluster operating power time series set to obtain the corresponding aggregation error clustering results includes: The power planned output deviation of each cluster operation power time series in the cluster operation power time series set is analyzed respectively to obtain the corresponding time series aggregation error; wherein the time series aggregation error can be understood as a deviation quantization value obtained by summarizing the deviation between the active output and the planned active power at each sampling moment in the cluster operation power time series, which can be expressed as: (9) in, represents the flexible active power demand delivered to the energy storage cluster at time t; represents the active power output of the u-th energy storage resource allocation at time t; represents the set of operating point sequences of all energy storage resources at time t; Represents the resulting deviation of the active power output operating point and the flexible active power planning set point of the cluster operating power time series.

[0031] According to the aggregation errors of each time series, K-means clustering analysis is performed on the cluster operation power time series set to obtain the corresponding aggregation error clustering results; wherein, the aggregation error clustering results can be understood as taking into account that the aggregation error may increase proportionally with the parameter differences of different energy storage resources, but the high-dimensional aggregation polyhedron is difficult to directly analyze the source of the error. The analysis results of the dependency relationship between the aggregation error and the geometric position of the active power demand (the geometric spatial position of the active power demand point in the energy storage cluster aggregation polyhedron) obtained by aggregating the active power demand of all cluster operation power time series (different geometric positions correspond to different power demand modes, and the aggregation errors under different power demand modes are different) include multiple aggregation error clustering clusters, and the cluster center of each aggregation error clustering cluster can represent a typical mode of active power demand and the corresponding aggregation error, that is, the distribution of the geometric spatial positions of the power demand points corresponding to each cluster center in the energy storage cluster aggregation polyhedron can be analyzed to obtain the degree of aggregation deviation under different location areas and different power demand combinations. The specific process of obtaining the aggregation error clustering results is as follows: 1) Set an appropriate constant k as the number of clusters for the active power demand aggregation error of the cluster operation power time series set, and select a random centroid for each cluster; 2) Calculate the distance from each cluster operation power time series in the cluster operation power time series set to the centroid, and assign the cluster operation power time series to the centroid closest to it; 3) By calculating the difference between the active output value of the energy storage flexibility resources of all cluster operating power time series in each cluster and the current cluster center, the average value of these differences is taken to update the cluster center, so that the cluster center after each round of iteration more accurately represents the active output of the class, thereby reducing errors and optimizing clustering results; 4) Repeat the above steps until the centroid remains in place and the desired aggregation error clustering result is obtained.

[0032] Through the above cluster analysis of the aggregation error of coupled flexible resources in the energy storage cluster, the dependence of the aggregation error on the geometric position of the corresponding active power demand time series in the energy storage cluster aggregation polyhedron can be determined. In order to accurately capture the dependence of the aggregation error on the geometric position of the active power demand of the aggregated flexible resources, the silhouette coefficient method is used to quantify the clustering quality in the above cluster analysis. If a cluster cluster is given Cluster operation power time series in , the silhouette coefficient of the time series is expressed as follows: (10) in, Distance clustering Other recent clusters, and Represent the cluster operation power time series With clustering and clusters The distance to the cluster center; Represents the cluster operating power time series The silhouette coefficient is the strength of quantifying the clustering of a given data. It can be used to measure the degree of alignment between the objects in the cluster and the rest of the cluster. The silhouette coefficient of the entire cluster is as follows: (11) in, Represents clusters The silhouette coefficient; Represents clusters The number of time series; Represents clusters Time series of cluster operation power The silhouette coefficient.

[0033] This embodiment uses the K-means clustering method combined with the silhouette coefficient method to perform cluster analysis on the flexibility resource aggregation error of the energy storage cluster, which can effectively capture the relationship between the aggregation error and the active power demand, and provide a reliable reference for further adjusting the active power demand, effectively improving the aggregation effect of the energy storage cluster and reducing the aggregation error.

[0034] S15. The clustering error cluster with the largest average aggregation error in the aggregation error clustering results is taken as the clustering cluster to be analyzed, and the cluster center of the clustering cluster to be analyzed is analyzed to obtain the corresponding aggregation error quantification result; wherein, the clustering cluster to be analyzed can be understood as a situation in which the power demand pattern with a large aggregation error needs to be adjusted and optimized, and the average value of the time series aggregation error of all cluster operating power time series in the cluster selected from the aggregation error clustering results is the largest; considering that the cluster center of each cluster can represent a type of active power demand, this embodiment preferably conducts an in-depth analysis of the cluster center time series of the clustering cluster to be analyzed, so as to dig out the power demand pattern adjusted by the actual demand for optimizing the subsequent energy storage flexibility resource adjustment strategy, and the corresponding aggregation error quantification result obtained by the analysis includes the location of the aggregation abnormal area, the abnormal power demand pattern and the aggregation error.

[0035] Specifically, the step of analyzing the cluster center of the cluster to be analyzed to obtain the corresponding aggregation error quantification result includes: The operating power point with the largest planned power output deviation in the cluster operating power time series corresponding to the cluster center is obtained as the error abnormality sample point; wherein the error abnormality sample point can be understood as the sample point with the largest deviation between the active power demand plan point and the active power demand sampling point in the cluster operating power time series corresponding to the cluster center, which has a greater impact on the overall aggregation error of the cluster to be analyzed.

[0036] The active power demand data of the energy storage cluster corresponding to the abnormal error sample point is obtained, and the feasible domain position and power mode analysis are performed on the active power demand data of the energy storage cluster to obtain the corresponding aggregate abnormal area position and abnormal power demand mode; wherein the feasible domain position can be understood as the geometric position of the active power demand data of the energy storage cluster in the energy storage cluster aggregation polyhedron; the power mode can be understood as a high power allocation demand or a low power allocation demand; that is, the aggregate abnormal area position in the obtained aggregate error quantization result is the active power demand of the abnormal error sample point corresponding to the geometric position in the energy storage cluster aggregation polyhedron, and the abnormal power demand mode is one of a high power demand mode and a low power demand mode.

[0037] Through the above steps, we can obtain the geometric position relationship of the active power demand of the energy storage cluster, the relationship between the active power demand pattern and the aggregation error, especially the dependence between different power demand patterns at different geometric positions and the size of the aggregation error. Through analysis, it can be seen that the geometric positions with large aggregation errors indicate that these active power demand patterns are prone to deviations under the current scheduling strategy and urgently need to be adjusted, thereby providing guidance for the subsequent optimization of the scheduling strategy.

[0038] The embodiment of the present invention generates energy storage resource operational domain constraints according to a pre-constructed energy storage time series coupling output model including an energy storage active output model and an energy storage time coupling flexibility model, generates a corresponding energy storage cluster aggregation polyhedron based on the time coupling characteristics according to each energy storage resource operational domain constraint, performs active power demand sampling based on the Markov chain Monte Carlo method according to the energy storage cluster aggregation polyhedron and preset sampling rules to obtain a cluster operation power time series set, performs aggregation error clustering analysis on the cluster operation power time series set to obtain an aggregation error clustering result including multiple aggregation error clustering clusters, and clusters the aggregation error demand sampling result to obtain a corresponding aggregation error clustering result including multiple aggregation error clustering clusters. The clustering error clustering cluster with the largest average aggregation error in the error clustering results is used as the clustering cluster to be analyzed, and the clustering center of the clustering cluster to be analyzed is analyzed to obtain the corresponding aggregation error quantification result including the aggregation abnormal area location, abnormal power demand pattern and aggregation error. The technical solution can effectively solve the application defects of the existing distribution network flexibility potential evaluation and analysis process, high calculation cost, low evaluation efficiency and difficulty in accurately evaluating the feasible domain aggregation error. It can not only reduce the complexity of high-dimensional space analysis, but also improve the accuracy of energy storage flexibility aggregation error evaluation, and provide a reliable basis for the subsequent optimization and adjustment of energy storage resource scheduling strategies.

[0039] In addition, considering that a large aggregation error in actual applications may lead to a low utilization efficiency of flexible resources, in order to minimize the aggregation error of energy storage flexible resources and improve the utilization effect of flexible resources, this embodiment preferably generates a corresponding energy storage resource scheduling adjustment strategy based on the aggregation error quantification result obtained by error cluster analysis. Specifically, Figure 2 As shown, the method includes: S16. Generate a corresponding flexibility resource scheduling adjustment strategy based on the quantification result of the aggregation error; wherein, the flexibility resource scheduling adjustment strategy can be understood as optimizing the scheduling strategy for the cluster center with a large aggregation error, so that future time periods can be more finely regulated to reduce the aggregation error; correspondingly, the flexibility resource scheduling adjustment strategy includes controlling storage capacity and optimizing power demand allocation, wherein controlling storage capacity can be understood as adjusting the charging and discharging conditions of energy storage resources, etc., and optimizing power demand allocation can be understood as determining whether to reduce the active power demand allocation value or increase the active power demand allocation value in the future based on the specific identified abnormal power demand mode (low power demand or high power demand); for example, when the identified abnormal power demand mode is high power demand, it is considered that the high power demand setting is unreasonable and the power demand needs to be appropriately reduced; when the identified abnormal power demand mode is low power demand, it is considered that the low power demand setting is unreasonable and the power demand needs to be appropriately increased.

[0040] This embodiment generates a flexibility resource scheduling adjustment strategy based on the aggregation error clustering analysis results for optimizing the flexibility resource scheduling, which can effectively reduce the flexibility resource aggregation error, improve the accuracy of energy storage flexibility aggregation error assessment, and effectively improve the utilization efficiency of flexibility resources.

[0041] It should be noted that although the steps in the above flowchart are shown in sequence according to the arrows, 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.

[0042] In one embodiment, Figure 3 As shown, a cluster-based flexible resource executable domain aggregation error quantization system is provided, and the system includes: A feasible domain acquisition module 1 is used to generate energy storage resource operational domain constraints according to a pre-built energy storage time-series coupled output model; the energy storage time-series coupled output model includes an energy storage active output model and an energy storage time-coupled flexibility model; Cluster feasible domain generation module 2, used to generate corresponding energy storage cluster aggregation polyhedron based on time coupling characteristics according to the operational domain constraints of each energy storage resource; The power demand sampling module 3 is used to perform active power demand sampling based on the Markov chain Monte Carlo method according to the energy storage cluster aggregation polyhedron and preset sampling rules to obtain a cluster operation power time series set; the preset sampling rule is to obtain multiple cluster operation power time series of preset days according to a preset sampling step and a preset time period sampling period; Aggregation error analysis module 4, used for performing aggregation error cluster analysis on the cluster operation power time series set to obtain corresponding aggregation error cluster results; The aggregation error quantification module 5 is used to take the aggregation error clustering cluster with the largest average aggregation error in the aggregation error clustering results as the clustering cluster to be analyzed, and analyze the cluster center of the clustering cluster to be analyzed to obtain the corresponding aggregation error quantification result; the aggregation error quantification result includes the location of the aggregation abnormality area, the abnormal power demand pattern and the aggregation error.

[0043] In one embodiment, Figure 4 As shown, a cluster-based flexible resource executable domain aggregation error quantization system is provided, and the system also includes: The resource scheduling adjustment module 6 is used to generate a corresponding flexible resource scheduling adjustment strategy according to the aggregation error quantization result; the flexible resource scheduling adjustment strategy includes controlling storage capacity and optimizing power demand allocation.

[0044] For the specific limitations of the cluster-based flexible resource operational domain aggregation error quantization system, please refer to the limitations of the cluster-based flexible resource operational domain aggregation error quantization method above, and the corresponding technical effects can also be obtained equivalently, which will not be repeated here. Each module in the above-mentioned cluster-based flexible resource operational domain aggregation error quantization system can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0045] In summary, the embodiment of the present invention provides a cluster-based flexible resource operational domain aggregation error quantification method and system, wherein the cluster-based flexible resource operational domain aggregation error quantification method realizes the generation of energy storage resource operational domain constraints according to a pre-constructed energy storage time series coupling output model including an energy storage active output model and an energy storage time coupling flexibility model, and after generating a corresponding energy storage cluster aggregation polyhedron based on the time coupling characteristics according to each energy storage resource operational domain constraint, active power demand sampling is performed based on the Markov chain Monte Carlo method according to the energy storage cluster aggregation polyhedron and preset sampling rules to obtain a cluster operating power time series set, and the cluster operating power time series is analyzed. The clustering analysis of the clustering errors is performed on the column set to obtain the clustering error results corresponding to multiple clustering error clustering clusters, and the clustering error clustering cluster with the largest average clustering error in the clustering error clustering results is used as the clustering cluster to be analyzed. The cluster center is analyzed to obtain the corresponding clustering error quantification results including the location of the abnormal aggregation area, the abnormal power demand pattern and the aggregation error, as well as the technical solution of generating the flexible resource scheduling adjustment strategy according to the aggregation error quantification results. It can not only reduce the complexity of high-dimensional space analysis, improve the efficiency of aggregation error evaluation and the accuracy of evaluation results, but also provide a reliable basis for the formulation of energy storage resource scheduling strategies, thereby effectively improving the utilization efficiency of flexible resources.

[0046] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not 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.

[0047] The above-mentioned embodiments only express several preferred implementation modes of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principle of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be based on the protection scope of the claims.

Claims

1. A cluster-based flexible resource operational domain aggregation error quantification method, characterized in that: The method comprises the following steps: Generate energy storage resource operational domain constraints based on a pre-built energy storage time-series coupled output model; the energy storage time-series coupled output model includes an energy storage active output model and an energy storage time-coupled flexibility model; According to the operational domain constraints of each energy storage resource, the corresponding energy storage cluster aggregation polyhedron is generated based on the time coupling characteristics; According to the energy storage cluster aggregation polyhedron and preset sampling rules, active power demand sampling is performed based on the Markov chain Monte Carlo method to obtain a cluster operation power time series set; the preset sampling rule is to obtain multiple cluster operation power time series of preset days according to a preset sampling step and a preset time period sampling period; Performing aggregation error clustering analysis on the cluster operation power time series set to obtain corresponding aggregation error clustering results; the aggregation error clustering results include multiple aggregation error clustering clusters; The aggregation error cluster with the largest average aggregation error in the aggregation error clustering results is taken as the cluster to be analyzed, and the cluster center of the cluster to be analyzed is analyzed to obtain the corresponding aggregation error quantification result; the aggregation error quantification result includes the location of the aggregation abnormality area, the abnormal power demand pattern and the aggregation error.

2. The cluster-based flexible resource operational domain aggregation error quantification method according to claim 1, characterized in that: The steps of constructing the energy storage active output model include: According to the active power limit of energy storage resources, energy storage charging and discharging power constraints, and energy storage charging and discharging start and stop constraints, an energy storage charging and discharging time coupling model is constructed; According to the energy storage capacity, charging efficiency, energy storage charging power and energy storage discharging power of the energy storage resources, a time-coupled model of energy storage state of charge is constructed.

3. The cluster-based flexible resource operational domain aggregation error quantification method according to claim 2, characterized in that: The energy storage charge and discharge time coupling model is expressed as: in, , and They represent the active power, energy storage charging power and energy storage discharging power of the energy storage resource at time t respectively; and They represent the maximum charging power and maximum discharging power of the energy storage resource respectively; and They respectively represent the charging state and discharging state of the energy storage resource at time t.

4. The cluster-based flexible resource operational domain aggregation error quantification method according to claim 2, characterized in that: The energy storage state of charge time coupling model is expressed as: in, and They respectively represent the initial state charge and initial charging state of the energy storage resource; Indicates the energy storage charge state of the energy storage resource at time t; and They represent the energy storage capacity and charging efficiency of the energy storage resources respectively; and They represent the energy storage charging power and energy storage discharging power of the energy storage resource at time t respectively; Indicates the duration interval.

5. The cluster-based flexible resource operational domain aggregation error quantification method according to claim 1, characterized in that: The steps of constructing the energy storage time-coupled flexibility model include: Based on the power balance of a single flexibility resource in the power system, the energy storage time coupling flexibility model is obtained; the energy storage time coupling flexibility model is expressed as: in, , and They represent the active power, static active power and flexible active power at time t respectively.

6. The cluster-based flexible resource operational domain aggregation error quantification method according to claim 1, characterized in that: The step of sampling active power demand based on the energy storage cluster aggregation polyhedron and preset sampling rules and obtaining a cluster operation power time series set based on the Markov chain Monte Carlo method includes: Obtaining the energy storage cluster operation time series data of each preset time period sampling period; the energy storage cluster operation time series data includes the energy storage cluster active power demand time series data and the operation parameter data of each energy storage resource in the cluster; According to the energy storage cluster operation time series data of each preset time period sampling period and the preset sampling step, uniform sampling is performed in the energy storage cluster aggregation polyhedron based on the Markov chain Monte Carlo method to obtain the corresponding cluster operation power time series.

7. The cluster-based flexible resource operational domain aggregation error quantification method according to claim 1, characterized in that: The step of performing aggregation error cluster analysis on the cluster operation power time series set to obtain a corresponding aggregation error cluster result comprises: Performing power plan output deviation analysis on each cluster operation power time series in the cluster operation power time series set respectively to obtain corresponding time series aggregation errors; According to each time series aggregation error, K-means clustering analysis is performed on the cluster operation power time series set to obtain a corresponding aggregation error clustering result.

8. The cluster-based flexible resource operational domain aggregation error quantification method according to claim 1, characterized in that: The step of analyzing the cluster centers of the clusters to be analyzed to obtain corresponding aggregation error quantification results comprises: Obtaining the operating power point with the largest power plan output deviation in the cluster operating power time series corresponding to the cluster center as the error abnormality sample point; The active power demand data of the energy storage cluster corresponding to the error abnormal sample point is obtained, and the feasible domain position and power mode analysis is performed on the active power demand data of the energy storage cluster to obtain the corresponding aggregate abnormal area position and abnormal power demand mode; the aggregate abnormal area position is the active power demand of the error abnormal sample point corresponding to the geometric position in the aggregate polyhedron of the energy storage cluster; the abnormal power demand mode includes a high power demand mode and a low power demand mode.

9. The cluster-based flexible resource operational domain aggregation error quantification method according to claim 1, characterized in that: The method comprises: According to the aggregate error quantification result, a corresponding flexibility resource scheduling adjustment strategy is generated; the flexibility resource scheduling adjustment strategy includes controlling storage capacity and optimizing power demand allocation.

10. A cluster-based flexible resource operational domain aggregation error quantization system, characterized in that: The system comprises: A feasible domain acquisition module is used to generate energy storage resource operational domain constraints according to a pre-built energy storage time-series coupled output model; the energy storage time-series coupled output model includes an energy storage active output model and an energy storage time-coupled flexibility model; The cluster feasible domain generation module is used to generate the corresponding energy storage cluster aggregation polyhedron based on the time coupling characteristics according to the operational domain constraints of each energy storage resource; A power demand sampling module is used to perform active power demand sampling based on the Markov chain Monte Carlo method according to the energy storage cluster aggregation polyhedron and preset sampling rules to obtain a cluster operation power time series set; the preset sampling rule is to obtain multiple cluster operation power time series of preset days according to a preset sampling step and a preset time period sampling period; An aggregation error analysis module is used to perform aggregation error cluster analysis on the cluster operation power time series set to obtain a corresponding aggregation error cluster result; the aggregation error cluster result includes a plurality of aggregation error cluster clusters; The aggregation error quantification module is used to take the aggregation error clustering cluster with the largest average aggregation error in the aggregation error clustering results as the clustering cluster to be analyzed, and analyze the cluster center of the clustering cluster to be analyzed to obtain the corresponding aggregation error quantification result; the aggregation error quantification result includes the location of the aggregation abnormality area, the abnormal power demand pattern and the aggregation error.

Images