Flexible load regulation capability evaluation method and system based on integrated clustering algorithm

Abstract

The present application relates to the technical field of distribution network flexible resource regulation capacity evaluation, and provides a flexible load regulation capacity evaluation method and system based on integrated clustering algorithm, which comprises: collecting historical daily load sample sets and the day before the regulation day daily load sample sets of power users; dividing the sample sets into K clusters through clustering and integrated clustering algorithm, and then obtaining the cluster to which the baseline load of the regulation day belongs; obtaining the regulation day baseline load of all power users participating in demand response and its upward and downward regulation amount limits through the cluster samples, and calculating the upward and downward regulation amount limits of the distribution network node load; constructing a multi-dimensional evaluation model of the flexible load system level regulation capacity based on the distribution network power flow constraint; and quantifying the output regulation capacity. The above method improves the accuracy of the distribution network flexible load regulation capacity evaluation and increases the diversity of the evaluation dimensions.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of distribution network flexible resource regulation capability evaluation, in particular to a flexible load regulation capability evaluation method and system based on integrated clustering algorithm. BACKGROUND

[0002] At present, the new energy installed capacity is 3.6 million kilowatts in 2024, the total capacity is 14.1 million kilowatts (8.9 million kilowatts of photovoltaic, 5.2 million kilowatts of wind power), and it is expected to exceed 17 million kilowatts in 2025, and reach about 300 million kilowatts in 2030.

[0003] However, the inherent volatility and uncontrollable characteristics of new energy pose a serious challenge to the safe and stable operation of the power system. The randomness and intermittency characteristics will lead to intensified system net load fluctuations and significantly increased peak-valley differences, thereby leading to a dramatic increase in system peak shaving pressure; at the same time, the high uncertainty of the temporal and spatial distribution of new energy power generation further exacerbates the curtailment of wind and light. Therefore, it is urgent to deeply tap the regulation potential of controllable resources of the power system to improve the safe and stable operation level of the power grid.

[0004] As an important part of controllable resources of the power system, flexible load has unique advantages in improving system regulation capability due to its flexible response characteristics. Compared with traditional generating units, flexible load not only can realize more refined power regulation, but also has the characteristics of low regulation cost, strong scalability and bidirectional regulation capability. Scientific evaluation of the regulation capability of flexible load is an important technical basis for fully releasing its regulation potential and improving the safe and stable operation of the power system.

[0005] In the evaluation of flexible load regulation capability, multiple factors such as load characteristics, response speed, external environmental constraints and user participation willingness need to be considered. Therefore, how to comprehensively, accurately and quickly evaluate the flexible load regulation capability becomes a problem to be solved, both the practicality and feasibility of the evaluation method and the comprehensiveness of the evaluation method need to be considered. The traditional flexible load regulation capability evaluation method mainly includes two types: one type depends on complex physical modeling, usually needs to depend on fine modeling of equipment level, collects multi-dimensional data such as equipment parameters, user behavior characteristics and environmental parameters of multiple types of flexible load, and realizes regulation capability evaluation through complex calculation. Although this method has accuracy advantage in theory, in the actual scene of large-scale heterogeneous flexible load access, the multi-dimensional data acquisition is difficult, the calculation complexity is high, which leads to the increase of engineering implementation cost and the limitation of operability. The second type of method uses clustering algorithm to analyze historical load data to evaluate the flexible regulation capability of users, which avoids complex physical modeling. For example, the Chinese invention patent with the publication number CN114219205B "Flexible load credible capacity calculation method for power grid planning", according to the historical power load data of the regional distribution network, the K-means clustering algorithm is used to extract the characteristics; in the dispatching period, the regional distribution network operation cost is minimized as the target, the regional distribution network optimization model considering the flexible load is established, the constraint conditions of the model are configured, the iteration is solved, and the typical day demand response value is obtained; according to the calculated typical day demand response value, combined with the flexible load response amount, the response amount and the determination of the user under the corresponding incentive level are calculated; but the above method uses a single clustering algorithm, which is difficult to fully adapt to the time sequence, high dimensionality and uncertainty of the load data, which leads to the clustering effect being easily disturbed, and affects the accuracy and stability of the subsequent evaluation results.

[0006] At the same time, like the above method, the existing research method has the problems of insufficient evaluation dimension and lack of systematic evaluation. The current research mostly focuses on the adjustable power range, response time and other basic parameters of flexible load, the evaluation dimension is too single, and lacks system-level indicators such as load curve smoothness, peak load reduction amount and new energy consumption contribution rate. This single-dimensional evaluation mode fails to establish a multi-level evaluation framework from users to nodes to distribution systems, making it difficult to comprehensively evaluate the overall regulation capability of flexible load from the macroscopic level of power grid safety and optimal scheduling. SUMMARY

[0007] The technical problem to be solved by the present application is how to improve the accuracy of the evaluation of the regulation capability of the flexible load of the distribution network and increase the diversity of the evaluation dimension.

[0008] The present application solves the above technical problems by the following technical means:

[0009] The application provides a flexible load regulation capacity evaluation method based on an integrated clustering algorithm, and comprises the following steps:

[0010] S1, obtaining a historical daily load sample set of any to-be-evaluated power user in a power distribution network and a regulation day ;

[0011] S2, calculating a regulation day baseline load of the power user ;

[0012] S3, clustering the historical daily load sample set of the power user to obtain an optimal clustering number K ;

[0013] S4, based on the optimal clustering number K obtained in step S3, using an integrated clustering algorithm to divide the sample set into K clusters , wherein ;

[0014] S5, calculating the regulation day baseline load of each cluster ; , obtaining the cluster to which the baseline load of the regulation day belongs ;

[0015] S6, based on the daily load sample data of the cluster to which the baseline load of the regulation day belongs , calculating the upward and downward regulation amount limits of the regulation day baseline load of the power user ;

[0016] S7, repeating steps S1-S6 to obtain the baseline load and the upward and downward regulation amount limits of all power users participating in demand response in the power distribution network, and calculating the upward and downward regulation amount limits of nodes in the power distribution network

[0017] S8, constructing a multi-dimensional evaluation model of the flexible load system-level regulation capacity of the power distribution network based on power flow constraints of the power distribution network

[0018] S9, based on the multi-dimensional evaluation model constructed in step S8, quantifying the output of the system-level flexible load regulation capacity of the power distribution network.

[0019] Further, the historical daily load sample set of the power user in step S1 is as follows:

[0020]

[0021] ​​​​

[0022] in, Indicates the electricity user number Daily load data for the day; Indicates the electricity user number Heavenly t Load data sampled at various times; The number of days for load data; The number of sampling times for the daily load;

[0023] The date of the regulation Daily load sample set As shown in the following formula:

[0024]

[0025]

[0026] in, This indicates that electricity users are on the [number]th day before the control date. Daily load data for the day; This indicates that electricity users are on the [number]th day before the control date. Heavenly t Load data sampled at various times; The number of sampling times for the daily load;

[0027] Step S2 describes the daily baseline load for the regulation of electricity users. The specific calculation is as follows:

[0028] .

[0029] Further, step S3 includes the following steps:

[0030] S31. Determine the number of clusters based on sample characteristics and sample size. k Range, traversal k Historical daily load sample set conduct Clustering;

[0031] S32. Calculate the number of clusters. k DBI indicator The The calculation is as follows:

[0032]

[0033] in, Representing the Each category / cluster; Representing the The average intra-class distance of each cluster; Representing thej the average distance within a cluster of the class cluster; the distance between the center point of the first cluster and the second cluster, calculated as follows: the distance between the center point of the first cluster and the second cluster, calculated as follows: j the distance between the center point of the first cluster and the second cluster, calculated as follows:

[0034]

[0035]

[0036] wherein, the number of samples within the cluster; the sample center of the cluster, calculated as follows: the sample center of the cluster, calculated as follows:

[0037]

[0038] wherein, the daily load data sample in the cluster.

[0039] S33, based on the calculation result of step S32, taking the minimum value corresponding to the optimal cluster number k .

[0040] Further, the step S4 comprises the following steps:

[0041] S41, using Bootstrap resampling technique to perform times of independent replacement sampling on the sample set , each time extracting samples to form a new sample set, calculated as follows:

[0042]

[0043] wherein, the sample set obtained by the th resampling; the daily load data of the th resampling sample set on the th day;

[0044] S42, taking the optimal cluster number as a unified parameter, performing clustering on each Bootstrap sample set to generate a base clustering result, calculated as follows:

[0045]

[0046] ​​​

[0047] wherein, represents a base cluster set; represents the first base cluster of the second sampling; represents the first cluster of the base cluster ;

[0048] S43, constructing a consensus matrix with dimension , the element of the consensus matrix represents the frequency that the sample and the sample are classified into the same cluster in all base cluster sets , as follows:

[0049]

[0050]

[0051] S44, based on the consensus matrix A, integrating the base cluster by the method of hierarchical clustering to obtain the final clustering result, the specific implementation method is:

[0052] (1) regarding each sample data in the sample set as an independent cluster, based on the consensus matrix obtained in step S43, the similarity between any two clusters is calculated in turn, as follows:

[0053]

[0054] wherein, and are the number of samples in the cluster ; are the sample index set in the cluster ;

[0055] (2) merging the two clusters with the largest similarity, until the number of clusters is , and the final clustering result is obtained, as follows:

[0056]

[0057] wherein, is the first daily load data in the cluster ; is the number of samples in the cluster .

[0058] Furthermore, the membership degree described in step S5 The calculation is as follows:

[0059]

[0060] in, and Representing clusters and clusters The sample center;

[0061] Adjusting the daily baseline load Classified into the cluster with the highest corresponding membership degree As shown in the following formula:

[0062]

[0063] in, For clusters The number of samples.

[0064] Further, step S6 includes the following steps:

[0065] S61, by cluster Power matrix constructed from sample data As shown in the following formula:

[0066]

[0067] in, Representing a cluster The Middle A sample, as shown in the following formula:

[0068]

[0069] Each column in Indicates the same sampling time Load data;

[0070] S62. Setting the significance level Calculate the load of electricity users at each sampling time. quantiles and Quantiles, specifically:

[0071] For sampling time The load, i.e. The first in The data was analyzed using the quantile estimation method to obtain the electricity user load at the sampling time. of quantiles and quantiles ;

[0072] S63, calculating the up-regulation limit and the down-regulation limit of the regulated daily load respectively as follows:

[0073]

[0074]

[0075] Further, the step S7 comprises the following steps:

[0076] S71, repeating the steps S1 to S6 to obtain the baseline load, the up-regulation limit and the down-regulation limit of all the power users participating in demand response in the power distribution network, respectively denoted as ; wherein, is the number of power users participating in demand response; is the number of sampling time points of daily load;

[0077] S72, calculating the baseline load of the power user at the node i in the power distribution network as follows:

[0078]

[0079] wherein, is the set of power users at the node ; is the number of nodes of the power distribution network; is the number of sampling time points of daily load;

[0080] S73, calculating the up-regulation limit and the down-regulation limit of the power user at the node i in the power distribution network as follows:

[0081]

[0082]

[0083] Further, the step S8 comprises the following steps:

[0084] S81, setting a target function from three evaluation dimensions of smoothing the system load curve, increasing the new energy consumption and reducing the peak shaving pressure respectively, as follows:

[0085]

[0086] ​​​​​​​​​

[0087]

[0088] wherein, is the load curve smoothness; is the new energy consumption; is the system load peak-valley difference; is the number of nodes of the distribution network; is the number of sampling time instants of the daily load; and respectively represent the maximum value and the minimum value of the daily load power of the distribution network node . represent the load value of the distribution network node at the time instant t . represent the average value of the daily load of the distribution network node . represent the active power output of the new energy of the distribution network node .

[0089] S82, set the distribution network safety constraint of the distribution network level flexible load regulation capability evaluation model, the constraint condition is as follows:

[0090]

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097] wherein, represent the head node set of the branch with as the end node; represent the end node set of the branch with as the head node; and respectively represent the active power and the reactive power flowing through the branch at the time instant . and respectively represent the active power and the reactive power flowing through the branch at the time instant . denotes the voltage amplitude of the node at the time instant ; denotes the resistance of the branch ; denotes the reactance of the branch ; denotes the impedance of the branch ; denote the upper and lower limits of the voltage, respectively; denotes the power flow of the branch at the time instant ; denotes the safety current of the branch ; is the load regulation capability of the node at the time instant ; is the power factor angle of the load of the node at the time instant ; is the active power output of the new energy of the node at the time instant ; is the new energy generation prediction value of the node at the time instant .

[0098] Further, the step S9 comprises the following steps:

[0099] S91, based on the objective function and the constraint condition constructed in step S8, solving three optimization models to obtain the optimal system operation state under each optimization target, and specifically outputting the following optimal indexes:

[0100] (1) the minimum value of the load curve smoothness ;

[0101] (2) the maximum value of the new energy consumption ;

[0102] (3) the minimum value of the system load peak-valley difference ;

[0103] S92, based on the optimal indexes obtained in step S91, respectively calculating the regulation rates of the flexible load on the load curve smoothness, the new energy consumption and the system load peak-valley difference, as follows:

[0104]

[0105] wherein, are respectively the regulation rates of the flexible load on the load curve smoothness, the new energy consumption and the system load peak-valley difference; These are the smoothness of the load curve, the amount of new energy consumption, and the peak-valley difference of the system load before the implementation of flexible load control.

[0106] This invention also provides a flexible load adjustment capability assessment system based on an ensemble clustering algorithm. The system operates using the above-mentioned method and includes the following modules:

[0107] The data acquisition module is used to obtain the historical daily load sample set of any power user to be evaluated in the distribution network. and regulation days Daily load sample set ;

[0108] The control baseline load calculation module is used to calculate the control baseline load of electricity users. ;

[0109] The clustering module is used to analyze historical daily load samples from electricity users. conduct Clustering yields the optimal number of clusters K;

[0110] The ensemble clustering module is used to cluster the sample set based on the optimal number of clusters K obtained from the clustering module. Divided into K clusters ,in ;

[0111] The classification module is used to calculate and regulate the daily baseline load. For each cluster membership degree Obtain the baseline load on the control day. To which the cluster belongs ;

[0112] User-level adjustment calculation module for baseline load based on the adjustment day. To which category Based on the daily load sample data, calculate the upward and downward adjustment limits of the daily baseline load for power user regulation;

[0113] The node-level regulation calculation module is used to repeatedly execute the above modules in sequence to obtain the baseline load of all power users participating in demand response in the distribution network and their upward and downward regulation limits, and to obtain the upward and downward regulation limits of system nodes.

[0114] The system-level multidimensional evaluation model construction module is used to construct a multidimensional evaluation model of the system-level regulation capability of flexible loads in distribution networks based on distribution network power flow constraints.

[0115] The output module is used to quantify the flexible load regulation capability of the distribution network system based on the multi-dimensional evaluation model constructed by the multi-dimensional evaluation model construction module.

[0116] The present application has the advantages of:

[0117] (1) The present application adopts an integrated clustering algorithm to mine the characteristics of the historical load data of power users, generates multiple base clusters through Bootstrap resampling, and performs hierarchical clustering integration based on a consensus matrix, effectively overcoming the problem that a single clustering algorithm is sensitive to noise and outliers, significantly improving the stability and accuracy of load clustering, and thereby laying a foundation for accurate quantification of user-level regulation capacity.

[0118] (2) The present application proposes a three-layer evaluation system from the user level, the node level to the system level. In the system level evaluation, multiple objectives such as smoothing the system load curve, improving new energy consumption capacity and reducing peak regulation pressure are comprehensively considered, and an optimization model considering power distribution network power flow and safety constraints is established to realize comprehensive and systematic evaluation of flexible load regulation capacity, and provide multi-dimensional decision support for power grid dispatching and new energy consumption. BRIEF DESCRIPTION OF DRAWINGS

[0119] Figure 1 FIG. 1 is a flowchart of the flexible load regulation capacity evaluation method based on the integrated clustering algorithm of the present application embodiment;

[0120] Figure 2 FIG. 2 is a schematic diagram of the modified IEEE33 node power distribution system in the simulation experiment of the present application embodiment;

[0121] Figure 3 FIG. 3 is a schematic diagram of the upward and downward regulation amount limits of the power load of a certain power user in the simulation experiment of the present application embodiment;

[0122] Figure 4 FIG. 4 is a schematic diagram of the upward and downward regulation amount limits of the power load of node 9 in the simulation experiment of the present application embodiment;

[0123] Figure 5 FIG. 5 is a schematic diagram of the system level regulation capacity index of flexible load regulation in the simulation experiment of the present application embodiment. DETAILED DESCRIPTION

[0124] In order to make the purpose, technical scheme and advantages of the present application embodiment clearer, the technical scheme in the present application embodiment will be described clearly and completely below in combination with the present application embodiment. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0125] Embodiment 1

[0126] The embodiment provides a flexible load regulation capacity evaluation method based on an integrated clustering algorithm. Through multi-level, systematic design and structured coupling, a set of layer-by-layer progressive collaborative optimization evaluation architecture is constructed. Specifically, the user-level evaluation firstly relies on the integrated clustering algorithm to accurately cluster the flexible load historical data, thereby avoiding the problem that a single algorithm is difficult to fully adapt to the time sequence, high dimensionality and uncertainty of the load data, so that the clustering effect is easily disturbed, and then the accuracy of the subsequent evaluation results is affected. On this basis, the node-level evaluation integrates the user-level quantitative results, and forms accurate input parameters required by the system-level optimization; finally, in the system-level evaluation, by establishing an optimization model considering the power flow and safety constraints of the power distribution network, the regulation rate of the flexible load regulation on the load curve smoothness, new energy consumption capacity and system load peak-valley difference is calculated from the three targets of smoothing the load curve, improving the new energy consumption capacity and reducing the system peak-valley difference, so that the multi-dimensional quantitative evaluation and scheduling decision support of the system regulation capacity are realized.

[0127] Specifically, the evaluation method process is as shown in Figure 1 , and includes the following steps:

[0128] S1, obtaining a historical daily load sample set of any to-be-evaluated power user in a power distribution network and a day-ahead day daily load sample set ; in the embodiment, the historical daily load sample of the power user participating in the flexible load regulation adopts the historical load data of the past year, and the sampling interval of the daily load data is 15 minutes, that is, there are 96 sampling data per day. As follows:

[0129]

[0130]

[0131] , wherein, represents the daily load data of the power user on the day; represents the load data sampled at the hour on the t day; 365 is the number of days of daily load data; 96 is the number of sampling time points of daily load;

[0132] In the embodiment, the "regulation day" refers to a specific date in the operation of the power system, during which the flexible load is planned to be regulated to respond to the system demand, that is, the target day for which the load regulation needs to be performed; the date selection is within a week before the regulation day, that is, the daily load sample set of the past 7 days before the regulation day , as follows:

[0133]

[0134]

[0135] in, This indicates that electricity users are on the [number]th day before the control date. Daily load data for the day; This indicates that electricity users are on the [number]th day before the control date. Heavenly t Load data sampled at specific times; 96 represents the number of sampling times for the daily load.

[0136] S2, Calculate the daily baseline load for electricity user regulation. ; as shown in the following formula:

[0137]

[0138] S3, Historical daily load sample set of electricity users conduct Clustering yields the optimal number of clusters K; the specific implementation includes the following steps:

[0139] S31. Determine the number of clusters based on sample characteristics and sample size. k Range, traversal Historical daily load sample set conduct Clustering; based on the sample size in this embodiment, k The initial value range is 2-10;

[0140] S32. Calculate the number of clusters. k DBI indicator The The calculation is as follows:

[0141]

[0142] in, Representing the Each category / cluster; Representing the The average intra-class distance of each cluster; Representing the j The average intra-class distance of each cluster; Representing the The first cluster and the first j The distance between the center points of each cluster is calculated using the following formula:

[0143]

[0144]

[0145] in, Representative clusters The number of samples within; Representative clusters The sample center is taken as its mean in this embodiment, as shown in the following formula:

[0146]

[0147] in, Representative clusters Daily load data samples.

[0148] S33. Based on the calculation results of step S32, with the minimum Corresponding k Value as the optimal cluster number .

[0149] S4. Based on the optimal clustering number K obtained in step S3, use an ensemble clustering algorithm to cluster the sample set. Divided into K clusters ,in The specific implementation method includes the following steps:

[0150] S41. In order to reduce K To address the instability of random initialization in the Means clustering algorithm, Bootstrap resampling is employed for the sample set. Perform 100 independent samplings with replacement, drawing 365 samples each time to form a new sample set. The probability of each sample being selected is 1 / 2. As shown in the following formula:

[0151]

[0152] in, Representing the The sample set obtained by resampling; Representing the The second resampling sample set Daily load data for the day;

[0153] S42, with the optimal number of clusters As a uniform parameter, for each Bootstrap sample set conduct Clustering, generating the base clustering results, as shown in the following formula:

[0154]

[0155]

[0156] in, Represents the base cluster set; Representing the Sub-sampling base clusters; representative base cluster

[0157] S43, constructing a consensus matrix with dimension , the element of the consensus matrix represents the frequency that sample and sample are classified into the same cluster in all base cluster sets , wherein sample and sample are the daily load data of day and day obtained in step S1, as follows:

[0158]

[0159]

[0160] S44, based on the consensus matrix A, the base clusters are integrated by the hierarchical clustering method to obtain the final clustering result, and the specific implementation method is as follows:

[0161] (1) each sample data in the sample set is regarded as an independent cluster, and based on the consensus matrix obtained in step S43, the similarity between any two clusters is calculated in turn, as follows:

[0162]

[0163] wherein, and are the number of samples in cluster ; are the sample index set in cluster ;

[0164] (2) the two clusters with the largest similarity are merged until the number of clusters is , and the final clustering result is obtained, as follows:

[0165]

[0166] wherein, is the th daily load data in cluster ; is the number of samples in cluster .

[0167] ​​Thus, the instability of the simple clustering algorithm is avoided, and the sample set is completely divided into clusters.

[0168] S5, calculating the baseline load of the regulation day For each cluster membership , the baseline load of the regulation day belongs to the cluster ; the calculation of the membership , as follows:

[0169]

[0170] , and respectively represent the sample center of cluster and cluster ;

[0171] The baseline load of the regulation day is classified into the cluster with the largest membership, as follows:

[0172]

[0173] , where is the number of samples in cluster .

[0174] S6, based on the baseline load of the regulation day belongs to the cluster of the daily load sample data, the upward and downward adjustment amount limit value of the baseline load of the power user regulation day is calculated; the specific implementation includes the following steps:

[0175] S61, construct the power matrix from the sample data in the cluster , as follows:

[0176]

[0177] , where represents the sample in the cluster , as follows:

[0178]

[0179] Each column in represents the load data at the same sampling time;

[0180] ​​​S62, setting the significance level In this embodiment, The quantile of the power user load at each sampling moment is calculated The quantile and The quantile, specifically:

[0181] For the load at the sampling moment , that is, the column data in , the quantile estimation method is used to obtain the quantile and the 0.95 quantile of the power user load at the sampling moment ;

[0182] S63, respectively calculating the upward adjustment limit value and the downward adjustment limit value of the regulated daily load, as follows:

[0183]

[0184]

[0185] S7, repeating steps S1-S6 to obtain the baseline load and its upward and downward adjustment limit values of all power users participating in demand response in the power distribution network, and calculating the upward and downward adjustment limit values of the nodes in the power distribution network; the specific implementation includes the following steps:

[0186] S71, repeating steps S1 to S6 to obtain the baseline load, upward adjustment limit value and downward adjustment limit value of all power users participating in demand response in the power distribution network, respectively denoted as , and , ; wherein, is the number of power users participating in demand response;

[0187] S72, calculating the baseline load i of the power users on the nodes in the power distribution network, as follows:

[0188]

[0189] wherein, is the set of power users on the node ; is the number of nodes in the power distribution network;

[0190] S73, calculating the upward adjustment limit value i of the power users on the nodes And down-regulate the amount limit As follows:

[0191]

[0192]

[0193] S8, construct a multi-dimensional evaluation model of distribution network flexible load system-level regulation capacity based on distribution network power flow constraints; the specific implementation includes the following steps:

[0194] S81, set the objective function from three evaluation dimensions of smoothing the system load curve, improving new energy consumption and reducing peak shaving pressure, as follows:

[0195]

[0196]

[0197]

[0198] Wherein, The load curve smoothness; The new energy consumption; The system load peak-valley difference; The number of nodes of the distribution network; The number of sampling time of daily load; And Respectively represent the maximum and minimum of the daily load power of the distribution network node ; Represent the load value of the distribution network node At t Time; Represent the average daily load of the distribution network node ; Represent the new energy active power of the distribution network node ;

[0199] S82, set the distribution network safety constraint of the distribution network-level flexible load regulation capacity evaluation model, to ensure the consistency and comparability of the evaluation results of each model, the same constraint conditions are adopted for the three optimization models, including power flow balance constraint, power grid safety constraint and flexible load regulation amount limit constraint, etc. The specific constraint conditions are as follows:

[0200]

[0201]

[0202]

[0203]

[0204]

[0205]

[0206]

[0207] wherein, denotes the head node set of the branch with as the end node; denotes the end node set of the branch with as the head node; and respectively denote the active and reactive power flowing through the branch at the moment of ; and respectively denote the active and reactive power flowing through the branch at the moment of ; denotes the voltage amplitude of the node at the moment of ; denotes the resistance of the branch ; denotes the reactance of the branch ; denotes the impedance of the branch ; respectively denote the upper and lower limits of the voltage; denotes the current-carrying capacity of the branch at the moment of ; denotes the safe current of the branch ; is the load regulation capability of the node at the moment of ; is the power factor angle of the load of the node at the moment of ; is the active power output of the new energy of the node at the moment of ; is the new energy power generation prediction value of the node at the moment of .

[0208] S9, based on the multi-dimensional evaluation model constructed in step S8, quantifying the output distribution network system-level flexible load regulation capability. The specific implementation includes the following steps:

[0209] S91, based on the objective function and constraints constructed in step S8, three optimization models are solved using the Gurobi optimizer to obtain the optimal system operating state under each optimization objective, and the following optimal indicators are output:

[0210] (1) Minimum value of load curve smoothness ;

[0211] (2) Maximum value of new energy consumption ;

[0212] (3) Minimum value of system load peak-valley difference ;

[0213] S92, based on the optimal indicators obtained in step S91, the adjustment rates of flexible load on load curve smoothness, new energy consumption, and system load peak-valley difference are calculated, as follows:

[0214]

[0215] wherein, are the adjustment rates of flexible load on load curve smoothness, new energy consumption, and system load peak-valley difference, respectively; are the load curve smoothness, new energy consumption, and system load peak-valley difference before implementing flexible load regulation, respectively.

[0216] The present embodiment also provides a simulation experiment applying the above method. The simulation experiment is based on MATLAB R2024b, Gurobi optimization solver, and uses a modified IEEE33 node distribution system as an example, as shown in Figure 2 The distributed power supply access situation is: node 5 accesses a distributed wind power supply with a rated capacity of 1.7 MVA, and node 13 accesses a distributed wind power supply with a rated capacity of 2 MVA. The flexible load access situation is: node 9 accesses 11 flexible power user loads, node 11 accesses 9 flexible power user loads, node 16 accesses 12 flexible power user loads, node 20 accesses 10 flexible power user loads, node 24 accesses 11 flexible power user loads, and node 29 accesses 9 flexible power user loads; the user loads of the remaining nodes are rigid loads.

[0217] A certain power user load at node 9 is used to illustrate the user-level flexible load regulation capability evaluation based on the integrated clustering algorithm described in the present embodiment. The power user's annual (365 days) electricity load data in 2023 is collected, with a time resolution of 15 minutes (i.e., 96 sampling points per day), and part of the data is shown in Table 1:

[0218] Table 1 Part of the electricity load data of a certain power user in 2023

[0219]

[0220] By implementing steps S1-S6, the baseline load and regulation capability of the user are obtained, as shown in Figure 3 .

[0221] The regulation capability of other flexible power loads on node 9 is obtained in the same way, and the power load regulation capability of node 9 is obtained (i.e., according to step S7), as shown in Figure 4 .

[0222] Similarly, the power load regulation capability of other nodes can be obtained. Finally, the system optimal indicators (the minimum value of load curve smoothness, the maximum value of new energy consumption, and the minimum value of system load peak-valley difference) are calculated according to steps S8-S9, and the regulation capability indicators of flexible load regulation (the regulation rate of flexible load on load curve smoothness, new energy consumption, and system load peak-valley difference) are obtained, as shown in Figure 5 .

[0223] In particular, in order to further verify the stability of the integrated clustering algorithm, the above power user load data is taken as an example, and the integrated clustering algorithm and the single clustering algorithm are respectively used to perform 10 times of clustering calculation on the user load data, and the DBI indicators of each clustering result are counted, and the results are shown in Table 2.

[0224] Table 2 Comparison of DBI indicators of clustering results

[0225]

[0226] As shown in Table 2, the DBI indicators of the single K-Means clustering algorithm fluctuate greatly in 10 times of clustering, for example, the DBI indicators of the 4th and 9th clustering results are 0.403 and 0.481 respectively (the greater the DBI value, the worse the clustering effect), reflecting that the clustering results are significantly affected by the random initial center, and the stability is poor; while the DBI indicators of the integrated clustering algorithm are always stable between 0.352-0.354, with very small changes, indicating that it has good robustness, and effectively overcomes the inconsistent clustering results caused by the initialization sensitivity of the traditional clustering algorithm.

[0227] Embodiment 2

[0228] It needs to be further explained that based on the same inventive concept, the embodiment provides a flexible load regulation capability evaluation system based on an integrated clustering algorithm, which adopts the method described in Embodiment 1 when running, including the following modules:

[0229] A data acquisition module is configured to obtain a historical daily load sample set and regulation day-ahead day load sample set ;

[0230] a regulation day baseline load calculation module, configured to calculate a regulation day baseline load of the power user ;

[0231] a clustering module, configured to cluster the historical day load sample set of the power user to obtain an optimal clustering number K ;

[0232] an integrated clustering module, configured to divide the sample set into K clusters based on the optimal clustering number K obtained by the clustering module and using an integrated clustering algorithm , wherein ; ;

[0233] a classification module, configured to calculate the regulation day baseline load based on the membership of each cluster ; , to obtain the cluster to which the baseline load of the regulation day belongs ; ;

[0234] a user-level adjustment calculation module, configured to calculate the upward and downward adjustment limits of the regulation day baseline load of the power user based on the day load sample data of the cluster to which the baseline load of the regulation day belongs ;

[0235] a node-level adjustment calculation module, configured to repeatedly execute the above modules in sequence to obtain the baseline loads and the upward and downward adjustment limits of all power users participating in demand response in the power distribution network, and to obtain the upward and downward adjustment limits of the system nodes

[0236] a system-level multi-dimensional evaluation model construction module, configured to construct a multi-dimensional evaluation model of the system-level adjustment capacity of the flexible load of the power distribution network based on the power flow constraints of the power distribution network

[0237] an output module, configured to quantitatively output the system-level flexible load adjustment capacity of the power distribution network based on the multi-dimensional evaluation model constructed by the multi-dimensional evaluation model construction module.

[0238] The above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.​

Claims

1. A method for evaluating flexible load adjustment capability based on ensemble clustering algorithm, characterized in that, Includes the following steps: S1. Obtain the historical daily load sample set of any power user to be evaluated in the distribution network. and regulation days Daily load sample set ; S2, Calculate the daily baseline load for electricity user regulation. ; S3, Historical daily load sample set of electricity users conduct Clustering yields the optimal number of clusters K; S4. Based on the optimal clustering number K obtained in step S3, use an ensemble clustering algorithm to cluster the sample set. Divided into K clusters ,in ; S5. Calculate and regulate daily baseline load. For each cluster membership degree Obtain the baseline load on the control day. To which the cluster belongs ; S6, Baseline load based on the control date To which category Based on the daily load sample data, calculate the upward and downward adjustment limits of the daily baseline load for power user regulation; S7. Repeat steps S1-S6 to obtain the baseline load of all power users participating in demand response in the distribution network and their upward and downward adjustment limits, and calculate the upward and downward adjustment limits of the distribution network nodes. S8. Construct a multi-dimensional evaluation model for the system-level regulation capability of flexible loads in distribution networks based on power flow constraints. Includes the following steps: S81. Set objective functions from three evaluation dimensions: smoothing the system load curve, increasing the amount of renewable energy absorbed, and reducing peak-shaving pressure, as follows: in, For the smoothness of the load curve; This refers to the amount of new energy consumed; This refers to the peak-to-valley difference in system load. This represents the number of nodes in the distribution network. The number of sampling times for the daily load; and Representing distribution network nodes Maximum and minimum daily load power; Indicates distribution network node exist Load value at any given time; Indicates distribution network node The average daily load; Indicates distribution network node New energy sources have contributed significantly; S82. Set the distribution network security constraints for the distribution network-level flexible load regulation capability assessment model. The constraint conditions are as follows: in, Indicates The set of the starting nodes of the branches of the terminal nodes; Indicates The set of end nodes of the branches of the first node; and They represent Flowing through the side road Active and reactive power; and They represent Flowing through the side road The active and reactive power; express Time Node The voltage amplitude; Indicates a branch The resistance; Indicates a branch The reactance; Indicates a branch impedance; These represent the upper and lower limits of the voltage, respectively. express Time Branch The carrying capacity; Indicates a branch Safe current; for Time Node Load regulation capability; For nodes The load at The power factor angle at time; for Time Node New energy sources have contributed significantly; for Time Node The predicted value of new energy power generation, for Time Node The reactive power of the baseline load of electricity users; S9. Based on the multi-dimensional evaluation model constructed in step S8, quantitatively output the flexible load regulation capability of the distribution network system.

2. The method for evaluating flexible load adjustment capability based on ensemble clustering algorithm according to claim 1, characterized in that, The historical daily load sample set of electricity users mentioned in step S1 As shown in the following formula: in, Indicates the electricity user number Daily load data for the day; Indicates the electricity user number Heavenly t Load data sampled at various times; The number of days for daily load data; The number of sampling times for the daily load; The date of the regulation Daily load sample set As shown in the following formula: in, This indicates that electricity users are on the [number]th day before the control date. Daily load data for the day; This indicates that electricity users are on the [number]th day before the control date. Heavenly t Load data sampled at various times; The number of sampling times for the daily load; Step S2 describes the daily baseline load for the regulation of electricity users. The specific calculation is as follows: 。 3. The method for evaluating flexible load adjustment capability based on ensemble clustering algorithm according to claim 2, characterized in that, Step S3 includes the following steps: S31. Determine the number of clusters based on sample characteristics and sample size. k Range, traversal k Historical daily load sample set conduct Clustering; S32. Calculate the number of clusters. k DBI indicator The The calculation is as follows: in, Representing the Each category / cluster; Representing the The average intra-class distance of each cluster; Representing the j The average intra-class distance of each cluster; Representing the The first cluster and the first j The distance between the center points of each cluster is calculated using the following formula: in, Representative clusters The number of samples within; Representative clusters The sample center is given by the following formula: in, Representative clusters Daily load data samples; S33. Based on the calculation results of step S32, with the minimum Corresponding k Value as the optimal cluster number .

4. The method for evaluating flexible load adjustment capability based on ensemble clustering algorithm according to claim 2, characterized in that, Step S4 includes the following steps: S41. Use Bootstrap resampling technology to sample the data set. conduct Sub-independent sampling with replacement, each time sampling Each sample forms a new sample set, as shown in the following formula: in, Representing the The sample set obtained by resampling; Representing the The second resampling sample set Daily load data for the day; S42, with the optimal number of clusters As a uniform parameter, for each Bootstrap sample set conduct Clustering, generating the base clustering results, as shown in the following formula: in, Represents the base cluster set; Representing the Base clustering based on secondary resampling; Representative base clustering The first in Each category / cluster; S43. Constructing a consensus matrix , dimension The elements of this symmetric matrix Indicates sample and samples In all base cluster sets The frequency of being assigned to the same cluster is as follows: S44. Based on the consensus matrix A, hierarchical clustering is used to integrate the base clusters to obtain the final clustering result. The specific execution method is as follows: (1) The sample set Each sample data point is considered an independent cluster. Based on the consensus matrix obtained in step S43, the consensus matrix is ​​calculated sequentially for any two clusters. Similarity between As shown in the following formula in, and Clusters The number of samples in the sample; Clusters The set of sample subscripts in; (2) Merge the two clusters with the highest similarity until the number of clusters is [number missing]. The final clustering result is then obtained, as shown in the following formula: in, For clusters The first in Daily load data; For clusters The number of samples.

5. The method for evaluating flexible load adjustment capability based on ensemble clustering algorithm according to claim 4, characterized in that, The membership degree described in step S5 The calculation is as follows: in, and Representing clusters and clusters The sample center; Adjusting the daily baseline load Classified into the cluster with the highest corresponding membership degree As shown in the following formula: in, For clusters The number of samples.

6. The method for evaluating flexible load adjustment capability based on ensemble clustering algorithm according to claim 5, characterized in that, Step S6 includes the following steps: S61, by cluster Power matrix constructed from sample data As shown in the following formula: in, Representing a cluster The Middle A sample, as shown in the following formula: Each column in Indicates the same sampling time Load data; S62. Setting the significance level Calculate the load of electricity users at each sampling time. quantiles and Quantiles, specifically: For sampling time The load, i.e. The first in The data was analyzed using the quantile estimation method to obtain the electricity user load at the sampling time. of quantiles and quantiles ; S63. Calculate the upper and lower adjustment limits for the daily load control respectively. and As shown in the following formula: 。 7. The method for evaluating flexible load adjustment capability based on ensemble clustering algorithm according to claim 6, characterized in that, Step S7 includes the following steps: S71. Repeat steps S1 to S6 to obtain the baseline load, upward adjustment limit, and downward adjustment limit of all power users participating in demand response in the distribution network, denoted as follows: , and , ;in, The number of electricity users participating in demand response; The number of sampling times for the daily load; S72, Calculate distribution network nodes i Baseline load of electricity users As shown in the following formula: in, It is a node The set of electricity users on the platform; It represents the number of nodes in the distribution network. The number of sampling times for the daily load; S73, Calculate distribution network nodes i Upward adjustment limit for electricity users and downward adjustment limit As shown in the following formula: 。 8. The method for evaluating flexible load adjustment capability based on ensemble clustering algorithm according to claim 1, characterized in that, Step S9 includes the following steps: S91. Based on the objective function and constraints constructed in step S8, solve the three optimization models to obtain the optimal system operating state under each optimization objective. Specifically, output the following optimal indices: (1) Minimum value of load curve smoothness ; (2) Maximum value of new energy consumption ; (3) Minimum value of system load peak-valley difference ; S92. Based on the optimal indicators obtained in step S91, calculate the adjustment rates of flexible load on load curve smoothness, renewable energy absorption, and system load peak-valley difference, as shown in the following formulas: in, , and These are the adjustment rates of flexible load on the smoothness of the load curve, the amount of new energy absorbed, and the peak-valley difference of the system load, respectively. , and These are the smoothness of the load curve, the amount of new energy consumption, and the peak-valley difference of the system load before the implementation of flexible load control.

9. A flexible load regulation capacity assessment system based on ensemble clustering algorithm, characterized in that, Includes the following modules: The data acquisition module is used to obtain the historical daily load sample set of any power user to be evaluated in the distribution network. and regulation days Daily load sample set ; The control baseline load calculation module is used to calculate the control baseline load of electricity users. ; The clustering module is used to analyze historical daily load samples from electricity users. conduct Clustering yields the optimal number of clusters K; The ensemble clustering module is used to cluster the sample set based on the optimal number of clusters K obtained from the clustering module. Divided into K clusters ,in ; The classification module is used to calculate and regulate the daily baseline load. For each cluster membership degree Obtain the baseline load on the control day. To which the cluster belongs ; User-level adjustment calculation module for baseline load based on the adjustment day. To which category Based on the daily load sample data, calculate the upward and downward adjustment limits of the daily baseline load for power user regulation; The node-level regulation calculation module is used to repeatedly execute the above modules in sequence to obtain the baseline load of all power users participating in demand response in the distribution network and their upward and downward regulation limits, and to obtain the upward and downward regulation limits of system nodes. The system-level multidimensional evaluation model construction module is used to construct a multidimensional evaluation model of the system-level regulation capability of flexible loads in a distribution network based on distribution network power flow constraints. The module performs the following steps: S81. Set objective functions from three evaluation dimensions: smoothing the system load curve, increasing the amount of renewable energy absorbed, and reducing peak-shaving pressure, as follows: in, For the smoothness of the load curve; This refers to the amount of new energy consumed; This refers to the peak-to-valley difference in system load. This represents the number of nodes in the distribution network. The number of sampling times for the daily load; and Representing distribution network nodes Maximum and minimum daily load power; Indicates distribution network node exist Load value at any given time; Indicates distribution network node The average daily load; Indicates distribution network node New energy sources have contributed significantly; S82. Set the distribution network security constraints for the distribution network-level flexible load regulation capability assessment model. The constraint conditions are as follows: in, Indicates The set of the starting nodes of the branches of the terminal nodes; Indicates The set of end nodes of the branches of the first node; and They represent Flowing through the side road Active and reactive power; and They represent Flowing through the side road The active and reactive power; express Time Node The voltage amplitude; Indicates a branch The resistance; Indicates a branch The reactance; Indicates a branch impedance; These represent the upper and lower limits of the voltage, respectively. express Time Branch The carrying capacity; Indicates a branch Safe current; for Time Node Load regulation capability; For nodes The load at The power factor angle at time; for Time Node New energy sources have contributed significantly; for Time Node The predicted value of new energy power generation, for Time Node The reactive power output of new energy sources; The output module is used to quantify the flexible load regulation capability of the distribution network system based on the multi-dimensional evaluation model constructed by the multi-dimensional evaluation model construction module.

Images