Demand response baseline load estimation method, system and device and storage medium

Abstract

The invention discloses a demand response baseline load estimation method, system and device and a storage medium, and aims to solve the error problem caused by directly using day-ahead declaration data as a baseline and guarantee the compensation calculation accuracy. The method comprises the following steps: acquiring day-ahead / day-inside load and electricity price data of users, dividing the day-ahead / day-inside users according to response time, and dividing the users into peak clipping / valley filling types according to types; fitting two types of price elastic coefficient matrixes based on the load and the electricity price change rate in the non-excitation period, and calculating the load variation in combination with the electricity price change rate in the excitation period; estimating an intra-day user baseline in combination with the original declaration data and the variable quantity, and clustering the intra-day users into a multi-class cluster by using k-means; matching day-ahead users with similar intra-day class clusters, calculating a distance weight coefficient, and performing weighted accumulation on class cluster baselines to obtain a day-ahead user result; and S5, adjusting parameters according to response types, and outputting an individual baseline for compensation calculation. The system comprises a data acquisition and classification module, equipment and a storage medium which are used for realizing the method.

Description

Technical Field

[0001] This application relates to the field of electricity market demand response technology, and in particular to a method, system, device and storage medium for demand response baseline load estimation. Background Technology

[0002] The stable operation of the electricity market is crucial for ensuring the reliability of energy supply. Demand response is an important mechanism for balancing supply and demand, and also a vital means of improving the stability and efficiency of the power grid. With the continuous opening of the electricity market and the changing dynamics of electricity supply and demand, incentive-based demand response for individual users has gradually become a focus of industry attention.

[0003] Calculating user compensation for incentive-based demand response requires an accurate estimate of the user's baseline load, i.e., the load level before participating in demand response. If the baseline load estimate is too high, the load aggregator will have to pay more compensation to participating users, thus affecting the aggregator's profits and even causing losses. If the baseline load estimate is too low, participating users may not receive sufficient economic compensation, which will inevitably reduce their participation enthusiasm and thus affect the promotion of demand response projects.

[0004] In electricity spot market transactions, day-ahead declared load data can typically be used as baseline load. However, it's important to note that users settle their electricity bills using the intraday real-time price, not the day-ahead price at the time of declaration. User consumption patterns are influenced by the real-time price, causing actual load changes. Therefore, directly using day-ahead declared load data as baseline load will inevitably lead to significant errors. Furthermore, there are differences in day-ahead declared volumes between users participating in day-ahead and intraday demand response. For users participating in day-ahead demand response, their day-ahead declared volumes already account for the impact of incentive-based demand response, meaning the declared load data for the demand response period has been adjusted to reflect reduced or increased loads. For users participating in intraday demand response, their day-ahead declared volumes are not affected by incentive-based demand response, meaning the load level during incentive-based demand response is not considered. Therefore, when estimating baseline load for incentive-based demand response, the differences between day-ahead and intraday demand response must be taken into account. Summary of the Invention

[0005] This application provides a method, system, device, and storage medium for estimating demand response baseline load to solve the above-mentioned problems.

[0006] On the one hand, this application provides a method for estimating the baseline load of demand response, the method comprising the following steps: Step S1: Obtain the day-ahead declared load data, the intra-day actual load data, the day-ahead electricity price data and the intra-day actual electricity price data of the user, divide the user into a day-ahead demand response user and an intra-day demand response user according to the time of participating in demand response, and divide the user into a peak-cutting type demand response user and a valley-filling type demand response user according to the response type; Step S2: Fit the peak-cutting type and valley-filling type price elasticity coefficient matrices based on the load change rate and the electricity price change rate in the non-incentive type demand response period, and calculate the user load change amount in combination with the electricity price change rate in the incentive type demand response period; Step S3: Estimate the baseline load of the intra-day demand response user in combination with the original declared load data and the load change amount, and obtain multiple clusters by clustering the load curve of the intra-day demand response user by using a k-means clustering algorithm; Step S4: Match the day-ahead demand response user with the most similar intra-day demand response cluster based on the load data in the non-demand response period, calculate the distance weight coefficient of the user in the cluster, and weight and accumulate the baseline loads of the users in the cluster to obtain the baseline load estimation result of the day-ahead demand response user; Step S5: Adjust the baseline load model parameters according to the demand response type participated by the user, and output the baseline load estimation result of the individual user for power spot market compensation calculation.

[0007] In an implementation manner of the present application, in the step S1, the sampling data further includes a day-ahead declared load curve before the user participates in the power spot market, an actual load curve of the demand response intra-day, and corresponding day-ahead electricity price data, intra-day actual electricity price data and incentive type demand response information.

[0008] In an implementation manner of the present application, in the step S2, fitting the price elasticity coefficient matrix includes: in the non-incentive type demand response period, calculating the difference rate of the declared load and the actual load at each time to obtain the load change rate, calculating the difference rate of the day-ahead electricity price and the intra-day electricity price at each time to obtain the electricity price change rate, and respectively establishing the peak-cutting type and valley-filling type price elasticity coefficient matrices by linear regression to represent the sensitivity of the load change to the electricity price change; wherein the peak-cutting type price elasticity coefficient matrix corresponds to the load reduction when the electricity price rises, and the valley-filling type price elasticity coefficient matrix corresponds to the load increase when the electricity price drops.

[0009] In an implementation manner of the present application, in the step S2, calculating the user load change amount includes: extracting the electricity price change rate in the incentive type demand response peak-cutting period and valley-filling period, respectively substituting into the fitted peak-cutting type and valley-filling type price elasticity coefficient matrices, calculating the theoretical load change rate of the user in the period, and then obtaining the load change amount.

[0010] In an implementation of the present application, in step S3, the estimation of the baseline load of the intra-day demand response user comprises: deducting or superimposing the load variation amount caused by the incentive demand response period price from the original declared load curve of the user to generate a load curve after the price adjustment as the baseline load estimation value of the intra-day user.

[0011] In an implementation of the present application, in step S3, the k-means clustering algorithm adopts the Euclidean distance to measure the similarity of the load curve, and the elbow method is used to determine the optimal clustering number, and the intra-day demand response users with similar load patterns are classified into the same cluster.

[0012] In an implementation of the present application, in step S4, the matching of the most similar cluster comprises: extracting the load data of the non-demand response period of the day-ahead user by using the synchronous pattern matching principle, calculating the Euclidean distance between the day-ahead user and all users in each intra-day cluster, and matching the day-ahead user to the intra-day cluster with the minimum distance.

[0013] In an implementation of the present application, in step S4, the calculation formula of the distance weight coefficient is: wherein d is the distance between the day-ahead user and the intra-day user, w is the weight, and n is the number of users in the cluster; the weight coefficient needs to be normalized to ensure that the baseline load curves of all intra-day users in the matched cluster are summed according to the corresponding weight coefficient when weighted and added.

[0014] In an implementation of the present application, in step 5, the parameter adjustment comprises: the baseline load correction formula of the peak-cut demand response is: The correction formula of the valley-filling demand response is: And a and β are calibrated through historical data.

[0015] Secondly, the present application also provides a demand response baseline load estimation system, which comprises: A data acquisition and classification module is used to acquire the load data, price data and demand response information of the user, and complete the time dimension and response type classification of the user; An elasticity coefficient and load variation calculation module is used to fit the price elasticity coefficient matrix of the peak-cut and valley-filling, and calculate the load variation amount of the user in the incentive demand response period; An intra-day baseline estimation and clustering module is used to estimate the baseline load of the intra-day demand response user, and cluster the load curve of the intra-day user by using the k-means algorithm; A day-ahead baseline matching estimation module is used to match the day-ahead user with the similar intra-day cluster, calculate the distance weight coefficient, and obtain the baseline load of the day-ahead user by weighting; ​Parameter adjustment and output module: used for adjusting model parameters according to response type, outputting baseline load estimation results.

[0016] Further, the present application also provides a demand response baseline load estimation device, the device comprises: at least one processor; and, a memory in communication connection with the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to complete the foregoing demand response baseline load estimation method.

[0017] Further, the present application also provides a demand response baseline load estimation nonvolatile computer storage medium, which stores computer executable instructions, and the computer executable instructions are executed by a processor to implement the foregoing demand response baseline load estimation method.

[0018] The demand response baseline load estimation method, system, device and storage medium provided by the present application have the following beneficial effects: (1) Filling the research gap of baseline load estimation based on day-ahead declared capacity. In the electricity spot market transaction, users need to declare load capacity in advance according to the day-ahead electricity price, but the existing research has not fully tapped the reference value of this data. The present application innovatively relates the day-ahead declared load to the real-time electricity price, quantifies the influence of price elasticity on user behavior, solves the baseline deviation problem caused by directly using day-ahead declared data, and significantly improves the estimation accuracy.

[0019] (2) Fusion of price elasticity theory and load pattern clustering to improve model adaptability. In view of the characteristics of the intra-day demand response users affected by the real-time electricity price fluctuation, the present application constructs a dynamic price elasticity coefficient matrix, combines with the load pattern clustering technology, and accurately identifies the user electricity consumption behavior. This method not only effectively distinguishes the load characteristics of price-sensitive and non-sensitive users, but also optimizes the model parameters through data driving, and enhances the generalization ability of baseline estimation for different user types and scenes.

[0020] (3) In view of the differences in the declaration characteristics of day-ahead and intra-day demand response users, the present application designs a phased baseline estimation logic: for intra-day users, the price elasticity is used to restore the real baseline; for day-ahead users, the synchronous pattern matching principle is used to strip the demand response influence. This hierarchical processing strategy takes into account the market transaction rules and user behavior characteristics, reduces the compensation cost risk of aggregators, at the same time guarantees the user participation enthusiasm, promotes the large-scale landing of demand response projects. BRIEF DESCRIPTION OF DRAWINGS

[0021] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings: Figure 1 A flow chart of a demand response baseline load estimation method provided by an embodiment of the application; Figure 2 A system composition diagram of a demand response baseline load estimation system provided by an embodiment of the application; Figure 3 A schematic diagram of a demand response baseline load estimation device provided by an embodiment of the application. DETAILED DESCRIPTION

[0022] To make the objects, technical solutions and advantages of the application clearer, the technical solutions of the application will be described below in detail with reference to the embodiments of the application and the accompanying drawings. Obviously, the described embodiments are only some of the embodiments of the application, but not all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the application.

[0023] The embodiments of the application provide a demand response baseline load estimation method, system, device and storage medium, and the technical solutions of the embodiments of the application will be described in detail below with reference to the accompanying drawings.

[0024] Figure 1 A flow chart of a demand response baseline load estimation method provided by an embodiment of the application. As shown in the figure, the method mainly includes the following steps: Figure 1 Step S1: Obtain day-ahead declared load data, intra-day actual load data, day-ahead electricity price data and intra-day actual electricity price data of a user, divide the user into a day-ahead demand response user and an intra-day demand response user according to a time of participating in demand response, and divide the user into a peak-cut type demand response user and a valley-fill type demand response user according to a response type; Step S2: Fit a peak-cut type and a valley-fill type price elasticity coefficient matrix respectively based on a load change rate and an electricity price change rate in a non-incentive type demand response period, and calculate a user load change amount in combination with an electricity price change rate in an incentive type demand response period; Step S3: Estimate a baseline load of the intra-day demand response user in combination with original declared load data and the load change amount, and perform clustering on a load curve of the intra-day demand response user by using a k-means clustering algorithm to obtain multiple clusters; ​Step S4: Based on the load data of the non-demand response period, match the day-ahead demand response user with the most similar day-ahead demand response cluster, calculate the distance weight coefficient of the users in the cluster and weight the baseline load of the users in the cluster to obtain the baseline load estimation result of the day-ahead demand response user; Step S5: Adjust the baseline load model parameters according to the demand response type participated by the user, and output the baseline load estimation result of the individual user for power spot market compensation calculation.

[0025] Further, the day-ahead declared load curve and the demand response day-ahead actual load curve before the user participates in the power spot market are collected. The corresponding day-ahead price data and day-ahead actual price data are synchronously obtained. The incentive demand response information is obtained, the users are divided into day-ahead demand response and day-ahead demand response according to the time of participating in demand response, and the type of participating in demand response is divided into peak shaving demand response and valley filling demand response.

[0026] Further, in this embodiment, 120 residential users in a certain region are randomly selected as research objects, including 20 users participating in day-ahead demand response and 100 users participating in day-ahead demand response, and the day-ahead declared load data and demand response day-ahead actual load data and corresponding price data are obtained, the data sampling interval is 60 minutes, and the type of participating in demand response is divided into peak shaving demand response and valley filling demand response.

[0027] Further, in the non-incentive demand response period, the difference rate of the declared load and the actual load of each user at each time is calculated to obtain the load change rate, and the difference rate of the day-ahead price and the day-ahead price at each time is calculated to obtain the price change rate, and the result is represented as:

[0028]

[0029] wherein, is the actual power load of the user i at the time t , and corresponds to the actual price at the time t ; is the day-ahead declared power load of the user t at the time , and t corresponds to the day-ahead price at the time .

[0030] Based on the load and price change rate of the user at each time in the non-incentive demand response period, the price elasticity coefficient matrix of the peak shaving type and the valley filling type is respectively established by linear regression, which represents the sensitivity of the price change to the load. The time when the load is reduced due to the price increase in the non-incentive demand response period is selected as , the load and the price change rate are used to fit the peak clipping type price elasticity coefficient matrix. Similarly, since the time when the load increases due to the price drop is , the load and the price change rate are used to fit the valley filling type price elasticity coefficient matrix:

[0031]

[0032] , wherein, is the peak clipping type price elasticity coefficient of the user i , and is the valley filling type price elasticity coefficient of the user i .

[0033] Further, the load change amount of the user caused by the time period price of the incentive type demand response is calculated. Specifically, the response time period of the peak clipping type incentive demand response is selected as , and the response time period of the valley filling type incentive demand response is . The peak clipping time period price change rate is , and the valley filling time period price change rate is . The peak clipping time period and the valley filling time period price change rates are respectively substituted into the fitted peak clipping type and valley filling type price elasticity coefficient matrices, and the corresponding load change rates are calculated:

[0034]

[0035] The load change amount of the user caused by the time period price of the incentive type demand response is calculated according to the load change rate:

[0036]

[0037] Further, combined with the original declared load data, the baseline load of the intra-day demand response user is estimated. Specifically, based on the load change amount of the user caused by the time period price of the incentive type demand response calculated in step S3, the load change amount caused by the time period price of the incentive type demand response is deducted or superimposed from the original declared load curve of the user, to generate a load curve after the price adjustment, as the baseline load estimation value of the intra-day user:

[0038]

[0039] Further, the k-means clustering algorithm is used to cluster the load curves of the intra-day demand response users.

[0040] Specifically, using the load curve after electricity price adjustment as input, the k-means algorithm is used to cluster users based on load pattern similarity, and the number of clusters is set using a method similar to the elbow rule. Users with similar load patterns are grouped into the same cluster, and those with dissimilar patterns are grouped into different clusters, resulting in multiple clusters. Each cluster contains a group of users with similar electricity consumption behaviors within the day. In this embodiment, 100 users participating in intraday demand response are clustered into 4 categories, and the clustering results are as follows: Figure 2 As shown.

[0041] Furthermore, based on load data from non-demand response periods, day-ahead demand response users are matched with the most similar intraday demand response clusters. Specifically, day-ahead user load data from non-demand response periods is extracted, and the Euclidean distance or correlation coefficient between these users and all users within each intraday cluster is calculated to measure load pattern similarity. The Euclidean distance between any two H-dimensional vectors can be expressed as:

[0042] Each day-ahead demand response user is matched to the nearest intraday demand response cluster, which is the intraday demand response cluster with the most similar load patterns.

[0043] Further, the distance weighting coefficients for users within the matching cluster are calculated. Specifically, for each user within a matching cluster within a given day, the inverse distance between that user and the user from the previous day is calculated. This distance is then normalized to generate the weighting coefficients for each user within a day relative to the user from the previous day. A higher weight indicates a greater contribution to the baseline estimation.

[0044]

[0045] in, For users responding to recent demands y Responding to user demand within the day x i distance, For the similarity between the two, For the corresponding weighting coefficients, n The number of users in the matched intraday demand response cluster.

[0046] Furthermore, the baseline load of users within the same cluster is weighted and accumulated to obtain the baseline load estimate of day-ahead demand response users.

[0047] Specifically, the baseline load curves of all intraday users within the matching cluster are weighted and summed according to their corresponding weighting coefficients to generate the baseline load estimate curves for intraday users:

[0048] in, This represents the baseline load for the intraday demand response period for user-incentivized demand. This represents the baseline load for the day-ahead demand response period for user-incentivized demand response.

[0049] Furthermore, the baseline load model parameters are adjusted based on the user's demand response type (peak shaving / valley filling). Specifically, for peak shaving users, an additional compensation coefficient is added during peak electricity price periods. The baseline load correction formula for peak shaving demand response is as follows:

[0050] Accordingly, the modified formula for valley-filling demand response is:

[0051] The values ​​of α and β range from 0.05 to 0.15 and are calibrated using historical data.

[0052] Finally, the baseline load estimate results for individual users are output for use in electricity spot market compensation calculations.

[0053] Specifically, a baseline load time series is generated for each user as the benchmark for compensation calculation. The baseline load is compared with the actual load to calculate the demand response reduction or increase, determining the economic compensation amount. The compensation calculation expression is as follows:

[0054]

[0055] The above describes a demand response baseline load estimation method provided by embodiments of this application. Based on the same inventive concept, embodiments of this application also provide a demand response baseline load estimation system. Figure 2 A diagram illustrating the composition of a demand response baseline load estimation system provided in this application embodiment is shown below. Figure 2 As shown, the system mainly includes: a data acquisition and classification module 201, used to acquire user load data, electricity price data, and demand response information, and to classify users by time dimension and response type; an elasticity coefficient and load change calculation module 202, used to fit peak-shaving and valley-filling price elasticity coefficient matrices, and to calculate the user load change during incentive-type demand response periods; an intraday baseline estimation and clustering module 203, used to estimate the baseline load of intraday demand response users, and to cluster intraday user load curves using the k-means algorithm; a day-ahead baseline matching estimation module 204, used to match day-ahead users with similar intraday clusters, calculate distance weight coefficients, and weight them to obtain the day-ahead user baseline load; and a parameter adjustment and output module 205, used to adjust model parameters according to the response type and output the baseline load estimation results.

[0056] The above describes a demand response baseline load estimation system provided by an embodiment of this application. Based on the same inventive concept, this application also provides a demand response baseline load estimation device. Figure 3 A schematic diagram of a demand response baseline load estimation device provided in an embodiment of this application is shown below. Figure 3 As shown, the device mainly includes: at least one processor 301; and a memory 302 communicatively connected to the at least one processor; wherein the memory 302 stores instructions that can be executed by the at least one processor 301, and the instructions are executed by the at least one processor 301 to enable the at least one processor 301 to complete the aforementioned demand response baseline load estimation method.

[0057] In addition, embodiments of this application also provide a non-volatile computer storage medium for demand response baseline load estimation, which stores computer-executable instructions that are executed by a processor to implement the aforementioned demand response baseline load estimation method.

[0058] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0059] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0060] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0061] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0062] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0063] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0064] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for estimating baseline load in demand response, characterized in that, The method includes the following steps: Step S1: Obtain the user's day-ahead declared load data, day-ahead actual load data, day-ahead electricity price data, and day-ahead actual electricity price data. Based on the time when the user participates in demand response, they are divided into day-ahead demand response users and day-ahead demand response users. Based on the response type, they are divided into peak shaving demand response and valley filling demand response. Step S2: Based on the load change rate and electricity price change rate during the non-incentive demand response period, fit the peak-shaving and valley-filling price elasticity coefficient matrices respectively, and calculate the user load change by combining the electricity price change rate during the incentive demand response period. Step S3: Combine the original declared load data with the load change to estimate the baseline load of demand response users within the day, and use the k-means clustering algorithm to cluster the load curves of demand response users within the day to obtain multiple clusters; Step S4: Based on the load data during non-demand response periods, match day-ahead demand response users with the most similar intraday demand response clusters, calculate the distance weight coefficients of users within the clusters, and weight and accumulate the baseline loads of users within the clusters to obtain the baseline load estimation results for day-ahead demand response users. Step S5: Adjust the baseline load model parameters according to the demand response type of user participation, and output the baseline load estimation results of individual users for use in electricity spot market compensation calculation.

2. The demand response baseline load estimation method according to claim 1, characterized in that, In step S1, the sampling data also includes the day-ahead declared load curve and the intraday actual load curve of the demand response before the user participates in the electricity spot market, as well as the corresponding day-ahead electricity price data, intraday actual electricity price data and incentive-based demand response information.

3. The demand response baseline load estimation method according to claim 1, characterized in that, In step S2, fitting the price elasticity coefficient matrix includes: during the non-incentive demand response period, calculating the difference rate between the declared load and the actual load at each time point to obtain the load change rate, calculating the difference rate between the day-ahead electricity price and the intraday electricity price at each time point to obtain the electricity price change rate, and establishing peak-shaving and valley-filling price elasticity coefficient matrices respectively through linear regression to characterize the sensitivity of electricity price changes to load; wherein, the peak-shaving price elasticity coefficient matrix corresponds to a decrease in load when the electricity price increases, and the valley-filling price elasticity coefficient matrix corresponds to an increase in load when the electricity price decreases.

4. The demand response baseline load estimation method according to claim 1, characterized in that, In step S2, calculating the user load change includes: extracting the electricity price change rate during peak shaving and valley filling periods of the incentive-driven demand response, substituting them into the fitted peak shaving and valley filling price elasticity coefficient matrices respectively, calculating the theoretical load change rate of users during that period, and thus obtaining the load change.

5. The demand response baseline load estimation method according to claim 1, characterized in that, In step S3, estimating the baseline load of demand response users during the day includes: subtracting or superimposing the load change caused by the price during the incentive demand response period from the user's original declared load curve to generate a load curve after electricity price adjustment, which serves as the estimated baseline load value for the user during the day.

6. The demand response baseline load estimation method according to claim 1, characterized in that, In step S3, the k-means clustering algorithm uses Euclidean distance to measure the similarity of load curves and uses the elbow method to determine the optimal number of clusters, grouping intraday demand response users with similar load patterns into the same cluster.

7. The demand response baseline load estimation method according to claim 1, characterized in that, In step S4, matching the most similar cluster includes: using the principle of synchronous pattern matching, extracting the load data of users during non-demand response periods of the previous day, calculating the Euclidean distance between them and all users in each intraday cluster, and matching the previous day users to the intraday cluster with the smallest distance.

8. The demand response baseline load estimation method according to claim 1, characterized in that, In step S4, the formula for calculating the distance weighting coefficient is: ， Where d is the distance between users before the day and users within the day, w is the weight, and n is the number of users in the cluster. The weight coefficients need to be normalized. When weighted and summed, the baseline load curves of all users within the matching cluster are summed according to the corresponding weight coefficients.

9. The demand response baseline load estimation method according to claim 1, characterized in that, In step S5, parameter adjustment includes: the baseline load correction formula for peak-shaving demand response is: The corrected formula for valley-filling demand response is: α and β are calibrated using historical data.

10. A demand response baseline load estimation system, characterized in that, The system includes: a data acquisition and classification module: used to acquire users' load data, electricity price data and demand response information, and to classify users by time dimension and response type; Elasticity coefficient and load change calculation module: used to fit the peak-shaving and valley-filling price elasticity coefficient matrix, and calculate the change in user load during the incentive demand response period; Intraday Baseline Estimation and Clustering Module: Used to estimate the baseline load of intraday demand response users and cluster the intraday user load curves using the k-means algorithm; The day-ahead baseline matching estimation module is used to match day-ahead users with similar intraday clusters, calculate distance weighting coefficients, and weight them to obtain the day-ahead user baseline load. Parameter Adjustment and Output Module: Used to adjust model parameters according to response type and output baseline load estimation results.

11. A demand response baseline load estimation device, characterized in that, The device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform a demand response baseline load estimation method according to any one of claims 1-9.

12. A non-volatile computer storage medium for demand response baseline load estimation, storing computer-executable instructions, characterized in that, The computer-executable instructions are executed by a processor to implement a demand response baseline load estimation method as described in any one of claims 1-9.

Images