Method for evaluating demand response potential of interruptible load

Abstract

The invention discloses an interruptible load demand response potential evaluation method. The method comprises the steps of collecting user power consumption behavior data and performing preprocessing to obtain preprocessed user power consumption behavior data; clustering the preprocessed user power consumption behavior data to divide the users into different categories, wherein the users in each category have similar power consumption behaviors; constructing an interruptible load demand prediction model, and analyzing and quantifying the interruptible power load of the user according to the power consumption behavior data of different types of users according to the interruptible load demand prediction model; and carrying out weighted summation on the user interruptible load quantities of different user types obtained by prediction to obtain an interruptible load demand response potential evaluation result.

Description

Technical Field

[0001] The present invention relates to the technical field of interruptible load demand assessment, and in particular to a method for evaluating interruptible load demand response potential. Background Art

[0002] In modern power systems, with the rapid development of renewable energy and the annual growth of electricity load demand, the power system is facing unprecedented challenges; traditional power system dispatching methods mainly rely on the regulation capabilities of the power generation side, but with the increasing penetration of intermittent renewable energy such as wind power and solar energy in the power grid, the regulation capabilities of the power generation side are subject to certain restrictions; due to the instability and intermittency of renewable energy, the imbalance between power supply and demand has gradually emerged, and the demand for flexibility in the power system is increasing.

[0003] In this context, demand response has gradually become an important means to improve the flexibility of the power system. Demand response refers to the use of incentives or price signals to stimulate users to actively change their electricity consumption behavior during periods of tight grid load or peak electricity prices, thereby helping the grid to achieve supply and demand balance. Interruptible load is the core of demand response. Interruptible load refers to the load that power users can temporarily interrupt or reduce when the grid needs it. By reasonably evaluating the demand response potential of interruptible load, the power system can better manage the contradiction between electricity supply and demand and achieve more efficient scheduling. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a method for evaluating interruptible load demand response potential, the method comprising:

[0005] S1: Collecting user electricity usage behavior data and preprocessing it to obtain preprocessed user electricity usage behavior data;

[0006] S2: clustering the pre-processed user power consumption behavior data to divide the users into different categories. Users in each category have similar power consumption behaviors, wherein fuzzy clustering based on optimized dynamic bending distance is a measurement method for the similar power consumption behaviors;

[0007] S3: constructing an interruptible load demand prediction model, wherein the model quantifies the interruptible power load of users by analyzing the power consumption behavior data of different categories of users;

[0008] S4: Perform weighted summation on the predicted interruptible loads of different user categories to obtain an evaluation result of interruptible load demand response potential.

[0009] Preferably: the user's electricity consumption behavior data in step S1 includes timestamp, electricity load, temperature and electricity price.

[0010] Preferably, in the step S1, the preprocessing adopts Gaussian weighted K nearest neighbor algorithm to fill in missing data.

[0011] Preferably, the preprocessing uses a Gaussian weighted K nearest neighbor algorithm to fill in missing data, including:

[0012] S11: Initialize a filling window K, and determine neighboring data based on the filling window K with the position of the data to be filled as the center;

[0013] S12: Use the Gaussian weighting function to calculate the weight of each neighbor data corresponding to the missing data. The closer the distance, the greater the weight of the neighbor data. The calculation formula is:

[0014]

[0015] in:

[0016] x represents the user's electricity consumption behavior data to be filled;

[0017] x j represents the user's electricity consumption behavior data at time j;

[0018] w j Represents the weight of the user's electricity consumption behavior data at time j

[0019] d(x,x j ) represents the time interval between the user's electricity usage behavior data at time j and the user's electricity usage behavior data to be filled;

[0020] σ represents the bandwidth parameter;

[0021] exp represents exponential function;

[0022] S13: Perform weighted filling on the data to be filled according to the calculated weight of the neighboring data. The calculation formula is:

[0023]

[0024] in:

[0025] Indicates the calculated data to be filled;

[0026] S14: Repeat steps S12 to S13 for the missing data in the user's electricity usage behavior data until all the missing data are filled in to obtain the pre-processed user's electricity usage behavior data.

[0027] Preferably, in step S2, clustering the pre-processed user electricity consumption behavior data to divide the users into different categories includes:

[0028] S21: randomly initialize the user category center according to the preset number of user categories C;

[0029] S22: Randomly initialize the membership matrix U = [u ij ];

[0030] in:

[0031] u ij Indicates the membership of the electricity consumption behavior data of the ith user to user category j. The membership satisfies the constraint condition and is calculated as follows:

[0032]

[0033] S23: Update the center of each user category center according to the membership degree of the electricity consumption behavior data. The update calculation formula is:

[0034]

[0035] in:

[0036] v j represents the center of the j-th user category;

[0037] s i represents the electricity consumption behavior data of the i-th user;

[0038] u ij Indicates the membership degree of the electricity consumption behavior data of the i-th user to user category j;

[0039] m represents the fuzzification coefficient;

[0040] N represents the number of user electricity consumption behavior data;

[0041] S24: Update the membership of the sample according to the optimized DTW distance between the center of the user category and the user's electricity consumption behavior data. The membership update formula is:

[0042]

[0043] in:

[0044] DTW(s i ,v j ) represents the electricity consumption behavior data s of the i-th user i and the center v of the jth user category j The optimized DTW distance between them;

[0045] S25: Repeat steps S23 and S24 until the membership matrix U converges, and divide the users into different categories according to the calculated memberships. The convergence means that the membership change after each iteration is lower than a preset threshold.

[0046] Preferably, the optimized DTW distance between the user power consumption behavior data in step S24 includes:

[0047] The DTW distance with boundary constraints is used to limit the search range of the DTW path and reduce the computational complexity. The specific process includes:

[0048] S24.1: Initialize the cumulative cost matrix D for storing the distance between the first user's electricity usage behavior data s1 and the second user's electricity usage behavior data s2. The initial state of the cost matrix is ​​set to D(i,j)=∞, indicating that the distance between the first user's electricity usage behavior data before time i and the second user's electricity usage behavior data before time j is infinite. Initialize the strip boundary constraint width parameter w.

[0049] S24.2: Calculate the data point distance d(i, j) between the data of the first user's electricity usage behavior data at time i and the data of the second user's electricity usage behavior data at time j, and the calculation formula is:

[0050] d(i,j)=(s1(i)-s2(j)) 2

[0051] in:

[0052] s1(i) represents the data of the first user's electricity consumption behavior at time i;

[0053] s2(j) represents the data of the second user's electricity consumption behavior at time j;

[0054] S24.3: Iteratively update and calculate the cumulative cost matrix, and use the D(L1, L2) value in the calculated cumulative cost matrix as the optimized DTW distance between the first user's power consumption behavior data and the second user's power consumption behavior data. The updated calculation formula is:

[0055] D(i,j)=d(i,j)+min(D(i-1,j),D(i,j-1),D(i-1,j-1))

[0056] in:

[0057] The iterative update is performed only within the strip boundary constraint area satisfying |ij|≤w, and other positions remain unchanged. L1 represents the length of the first user's power usage behavior data, and L2 represents the length of the second user's power usage behavior data.

[0058] Preferably, the interruptible load demand prediction model is constructed in step S3, including:

[0059] S31: using ADF test to detect the stability of electricity consumption behavior data of different categories of users. If the user electricity consumption behavior data is not stable, first-order difference processing is performed to obtain stable user electricity consumption behavior data;

[0060] S32: using the autocorrelation function and the partial autocorrelation function to calculate the autoregressive order and the moving average order of the interruptible load demand prediction model according to the stabilized user power consumption behavior data;

[0061] S33: Optimizing the model using the improved quasi-Newton algorithm to obtain the optimal model parameters;

[0062] S34: Instantiate the model according to the optimal model parameters obtained by the optimization solution, and use the model instance to predict the power load, and use the difference between the predicted user power load and the actual user power load as the interruptible load demand of the corresponding user.

[0063] Preferably, in step S33, the improved quasi-Newton algorithm is used to optimize the model to obtain the optimal model parameters, including:

[0064] S33.1: Construct the objective function based on the interruptible load demand prediction model, and the calculation formula is:

[0065]

[0066] in:

[0067] N ′ Indicates the total number of user electricity usage behavior data included in the user category;

[0068] θ represents the interruptible load demand prediction model parameter;

[0069] ∈ i (θ) represents the prediction deviation of the interruptible load demand prediction model with θ as parameter;

[0070] S33.2: Randomly initialize the model parameters θ0 and the Hessian matrix B0, and calculate the initial model prediction bias ∈(θ0) and gradient Initialize the maximum number of iterations;

[0071] S33.3: Calculate the search direction based on the Hessian matrix and gradient calculated in the current iteration round. The calculation formula is:

[0072] p k =-B k g k

[0073] in:

[0074] B kRepresents the Hessian matrix obtained by k rounds of iterative calculation;

[0075] g k Represents the gradient calculated after k rounds of iterations;

[0076] S33.4: Update the model parameters according to the calculated search direction. The calculation formula is:

[0077] θ k+1 =θ k +α k p k

[0078] in:

[0079] p k Represents the search direction obtained by k rounds of iterative calculation;

[0080] θ k represents the model parameters of round k;

[0081] θ k+1 represents the model parameters of the k+1 round;

[0082] α k represents the step size parameter calculated by the line search method in round k;

[0083] S33.5: Update the gradient and Hessian matrix according to the calculated model parameters;

[0084] S33.6: Iterate steps S33.3 to S33.5 until the maximum number of iterations is reached and the optimal model parameters are output.

[0085] Preferably, in step S33.5, updating the gradient and Hessian matrix according to the calculated model parameters includes:

[0086] A1: Update the gradient g according to the calculated model parameters k+1 , and calculate the parameter difference s k =θ k+1 -θ k and the gradient difference y k =g k+1 -g k ;

[0087] in:

[0088] s k represents the parameter difference of k rounds;

[0089] y k Represents the gradient difference of k rounds;

[0090] A2: Use the calculated parameter difference and gradient difference to estimate and update the Hessian matrix. The calculation formula is:

[0091]

[0092] in:

[0093] B k Represents the Hessian matrix obtained by k rounds of iterative calculation;

[0094] B k+1 Represents the Hessian matrix obtained by k+1 rounds of iterative calculation;

[0095] T represents the transpose operation.

[0096] Preferably, in step S4, the weighted sum of the predicted interruptible loads of different user categories is performed to obtain an evaluation result of interruptible load demand response potential, including:

[0097] S41: using AHP analytic hierarchy process to calculate the weight parameters of user interruptible load for different user categories;

[0098] S42: Obtain an interruptible load demand response potential evaluation result by performing weighted summation based on the calculated user interruptible load weight parameter and the user interruptible load.

[0099] Technical effects and advantages of the present invention:

[0100] 1. In the data preprocessing stage, the present invention uses the Gaussian weighted K nearest neighbor algorithm to fill in missing data. Compared with the traditional KNN, the use of Gaussian weighting can better consider the distance weight between adjacent data points, reduce filling errors, and improve the accuracy and robustness of preprocessed data.

[0101] 2. In the clustering stage, the present invention uses a fuzzy clustering algorithm that optimizes the dynamic bending distance to capture the user's electricity consumption behavior pattern, which improves the calculation efficiency and can more accurately characterize the electricity consumption behavior pattern.

[0102] 3. Aiming at the problem of predicting interruptible load, the present invention introduces the ARIMA model to perform periodic modeling based on historical electricity consumption behavior data, which has good adaptability and accuracy for predicting user interruptible load.

[0103] 4. The present invention utilizes an improved quasi-Newton algorithm to perform model optimization, thereby improving model optimization efficiency and achieving rapid demand response. BRIEF DESCRIPTION OF THE DRAWINGS

[0104] Figure 1 It is a flowchart of this application; DETAILED DESCRIPTION

[0105] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0106] See also Figure 1 In this embodiment, a method for evaluating the potential of interruptible load demand response is provided. The execution subject of the method includes but is not limited to a server, a terminal, etc. that can be configured to execute at least one of the electronic devices provided in the embodiment of the present application. In other words, the method for evaluating the potential of interruptible load demand response can be executed by software or hardware installed in a terminal device or a server device. The software can be a blockchain platform, and the server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0107] A method for evaluating interruptible load demand response potential comprises the following steps:

[0108] S1: Collect user electricity usage behavior data and preprocess it to obtain preprocessed user electricity usage behavior data, wherein the user electricity usage behavior data includes timestamp, electricity load, temperature and electricity price. The preprocessing adopts Gaussian weighted K nearest neighbor algorithm to fill in missing data.

[0109] The preprocessing in step S1 uses a Gaussian weighted K nearest neighbor algorithm to fill in missing data, including:

[0110] S11: Initialize a filling window K, and determine neighboring data based on the filling window K with the position of the data to be filled as the center;

[0111] S12: Use the Gaussian weighting function to calculate the weight of each neighbor data corresponding to the missing data. The closer the distance, the greater the weight of the neighbor data. The calculation formula is:

[0112]

[0113] in:

[0114] x represents the user's electricity consumption behavior data to be filled;

[0115] x j represents the user's electricity consumption behavior data at time j;

[0116] w jRepresents the weight of the user's electricity consumption behavior data at time j

[0117] d(x,x j ) represents the time interval between the user's electricity usage behavior data at time j and the user's electricity usage behavior data to be filled;

[0118] σ represents the bandwidth parameter;

[0119] exp represents exponential function;

[0120] S13: Perform weighted filling on the data to be filled according to the calculated weight of the neighboring data. The calculation formula is:

[0121]

[0122] in:

[0123] Indicates the calculated data to be filled;

[0124] S14: Repeat steps S12 to S13 for the missing data in the user's electricity usage behavior data until all the missing data are filled in to obtain the pre-processed user's electricity usage behavior data.

[0125] S2: Clustering the pre-processed user power usage behavior data to divide the users into different categories. Users in each category have similar power usage behaviors, wherein fuzzy clustering based on optimized dynamic bending distance is a measurement method for the similar power usage behaviors.

[0126] In step S2, clustering is performed on the pre-processed user electricity consumption behavior data to divide the users into different categories, including:

[0127] S21: randomly initialize the user category center according to the preset number of user categories C;

[0128] S22: Randomly initialize the membership matrix U = [u ij ];

[0129] in:

[0130] u ij Indicates the membership of the electricity consumption behavior data of the ith user to user category j. The membership satisfies the constraint condition and is calculated as follows:

[0131]

[0132] S23: Update the center of each user category center according to the membership degree of the electricity consumption behavior data. The update calculation formula is:

[0133]

[0134] in:

[0135] v j represents the center of the j-th user category;

[0136] s i represents the electricity consumption behavior data of the i-th user;

[0137] u ij Indicates the membership degree of the electricity consumption behavior data of the i-th user to user category j;

[0138] m represents the fuzzification coefficient;

[0139] N represents the number of user electricity consumption behavior data;

[0140] S24: Update the membership of the sample according to the optimized DTW distance between the center of the user category and the user's electricity consumption behavior data. The membership update formula is:

[0141]

[0142] in:

[0143] DTW(s i ,v j ) represents the electricity consumption behavior data s of the i-th user i and the center v of the jth user category j The optimized DTW distance between them;

[0144] S25: Repeat steps S23 and S24 until the membership matrix U converges, and divide the users into different categories according to the calculated memberships. The convergence means that the membership change after each iteration is lower than a preset threshold.

[0145] The optimized DTW distance between the user power consumption behavior data in step S24 includes:

[0146] The DTW distance with boundary constraints is used to limit the search range of the DTW path and reduce the computational complexity. The specific process includes:

[0147] S24.1: Initialize the cumulative cost matrix D for storing the distance between the first user's electricity usage behavior data s1 and the second user's electricity usage behavior data s2. The initial state of the cost matrix is ​​set to D(i,j)=∞, indicating that the distance between the first user's electricity usage behavior data before time i and the second user's electricity usage behavior data before time j is infinite. Initialize the strip boundary constraint width parameter w.

[0148] S24.2: Calculate the data point distance d(i, j) between the data of the first user's electricity usage behavior data at time i and the data of the second user's electricity usage behavior data at time j, and the calculation formula is:

[0149] d(i,j)=(s1(i)-s2(j)) 2

[0150] in:

[0151] s1(i) represents the data of the first user's electricity consumption behavior at time i;

[0152] s2(j) represents the data of the second user's electricity consumption behavior at time j;

[0153] S24.3: Iteratively update and calculate the cumulative cost matrix, and use the D(L1, L2) value in the calculated cumulative cost matrix as the optimized DTW distance between the first user's power consumption behavior data and the second user's power consumption behavior data. The updated calculation formula is:

[0154] D(i,j)=d(i,j)+min(D(i-1,j),D(i,j-1),D(i-1,j-1))

[0155] in:

[0156] The iterative update is performed only within the strip boundary constraint area satisfying |ij|≤w, and other positions remain unchanged. L1 represents the length of the first user's power usage behavior data, and L2 represents the length of the second user's power usage behavior data.

[0157] S3: Construct an interruptible load demand prediction model, which quantifies the user's interruptible power load by analyzing the power consumption behavior data of different categories of users.

[0158] The interruptible load demand prediction model is constructed in step S3, including:

[0159] S31: using ADF test to detect the stability of electricity consumption behavior data of different categories of users. If the user electricity consumption behavior data is not stable, first-order difference processing is performed to obtain stable user electricity consumption behavior data;

[0160] S32: using the autocorrelation function and the partial autocorrelation function to calculate the autoregressive order and the moving average order of the interruptible load demand prediction model according to the stabilized user power consumption behavior data;

[0161] S33: Use the improved quasi-Newton algorithm to optimize the model to obtain the optimal model parameters and improve the efficiency of model parameter solution;

[0162] S34: Instantiate the model according to the optimal model parameters obtained by the optimization solution, and use the model instance to predict the power load, and use the difference between the predicted user power load and the actual user power load as the interruptible load demand of the corresponding user.

[0163] In the step S33, the improved quasi-Newton algorithm is used to optimize the model to obtain the optimal model parameters, including:

[0164] S33.1: Construct the objective function based on the interruptible load demand prediction model, and the calculation formula is:

[0165]

[0166] in:

[0167] N ′ Indicates the total number of user electricity usage behavior data included in the user category;

[0168] θ represents the interruptible load demand prediction model parameter;

[0169] ∈ i (θ) represents the prediction deviation of the interruptible load demand prediction model with θ as parameter;

[0170] S33.2: Randomly initialize the model parameters θ0 and the Hessian matrix B0, and calculate the initial model prediction bias ∈(θ0) and gradient Initialize the maximum number of iterations;

[0171] S33.3: Calculate the search direction based on the Hessian matrix and gradient calculated in the current iteration round. The calculation formula is:

[0172] p k =-B k g k

[0173] in:

[0174] B k Represents the Hessian matrix obtained by k rounds of iterative calculation;

[0175] g k Represents the gradient calculated after k rounds of iterations;

[0176] S33.4: Update the model parameters according to the calculated search direction. The calculation formula is:

[0177] θ k+1 =θ k +α k p k

[0178] in:

[0179] p k Represents the search direction obtained by k rounds of iterative calculation;

[0180] θ k represents the model parameters of round k;

[0181] θ k+1 represents the model parameters of the k+1 round;

[0182] α k represents the step size parameter calculated by the line search method in round k;

[0183] S33.5: Update the gradient and Hessian matrix according to the calculated model parameters;

[0184] S33.6: Iterate steps S33.3 to S33.5 until the maximum number of iterations is reached and the optimal model parameters are output.

[0185] In step S33.5, the gradient and Hessian matrix are updated according to the calculated model parameters, including:

[0186] A1: Update the gradient g according to the calculated model parameters k+1 , and calculate the parameter difference s k =θ k+1 -θ k and the gradient difference y k =g k+1 -g k ;

[0187] in:

[0188] s k represents the parameter difference of k rounds;

[0189] y k Represents the gradient difference of k rounds;

[0190] A2: Use the calculated parameter difference and gradient difference to estimate and update the Hessian matrix. The calculation formula is:

[0191]

[0192] in:

[0193] B k Represents the Hessian matrix obtained by k rounds of iterative calculation;

[0194] B k+1 Represents the Hessian matrix obtained by k+1 rounds of iterative calculation;

[0195] T represents the transpose operation.

[0196] S4: The predicted interruptible loads of different user categories are weighted and summed to obtain the interruptible load demand response potential evaluation results, including:

[0197] S41: using AHP analytic hierarchy process to calculate the weight parameters of user interruptible load for different user categories;

[0198] S42: Obtain an interruptible load demand response potential evaluation result by performing weighted summation based on the calculated user interruptible load weight parameter and the user interruptible load.

[0199] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. A method for evaluating interruptible load demand response potential, characterized in that: The method comprises: S1: Collecting user electricity usage behavior data and preprocessing it to obtain preprocessed user electricity usage behavior data; S2: clustering the pre-processed user power consumption behavior data to divide the users into different categories. Users in each category have similar power consumption behaviors, wherein fuzzy clustering based on optimized dynamic bending distance is a measurement method for the similar power consumption behaviors; S3: constructing an interruptible load demand prediction model, wherein the model quantifies the interruptible power load of users by analyzing the power consumption behavior data of different categories of users; S4: Perform weighted summation on the predicted interruptible loads of different user categories to obtain an evaluation result of interruptible load demand response potential.

2. The method for evaluating interruptible load demand response potential according to claim 1, characterized in that: The user's electricity consumption behavior data in step S1 includes timestamp, electricity load, temperature and electricity price.

3. The method for evaluating interruptible load demand response potential according to claim 1, characterized in that: In the S1 step, the preprocessing adopts the Gaussian weighted K nearest neighbor algorithm to fill in the missing data.

4. The method for evaluating interruptible load demand response potential according to claim 3, characterized in that: The preprocessing uses a Gaussian weighted K nearest neighbor algorithm to fill in missing data, including: S11: Initialize a filling window K, and determine neighboring data based on the filling window K with the position of the data to be filled as the center; S12: Use the Gaussian weighting function to calculate the weight of each neighbor data corresponding to the missing data. The closer the distance, the greater the weight of the neighbor data. The calculation formula is: in: x represents the user's electricity consumption behavior data to be filled; x j represents the user's electricity consumption behavior data at time j; w j Represents the weight of the user's electricity consumption behavior data at time j d(x,x j ) represents the time interval between the user's electricity usage behavior data at time j and the user's electricity usage behavior data to be filled; σ represents the bandwidth parameter; exp represents exponential function; S13: Perform weighted filling on the data to be filled according to the calculated weight of the neighboring data. The calculation formula is: in: Indicates the calculated data to be filled; S14: Repeat steps S12 to S13 for the missing data in the user's electricity usage behavior data until all the missing data are filled in to obtain the pre-processed user's electricity usage behavior data.

5. The method for evaluating interruptible load demand response potential according to claim 1, characterized in that: In step S2, clustering is performed on the pre-processed user electricity consumption behavior data to divide the users into different categories, including: S21: randomly initialize the user category center according to the preset number of user categories C; S22: Randomly initialize the membership matrix U = [u ij ]; in: u ij Indicates the membership of the electricity consumption behavior data of the ith user to user category j. The membership satisfies the constraint condition and is calculated as follows: S23: Update the center of each user category center according to the membership degree of the electricity consumption behavior data. The update calculation formula is: in: v j represents the center of the j-th user category; s i represents the electricity consumption behavior data of the i-th user; u ij Indicates the membership degree of the electricity consumption behavior data of the i-th user to user category j; m represents the fuzzification coefficient; N represents the number of user electricity consumption behavior data; S24: Update the membership of the sample according to the optimized DTW distance between the center of the user category and the user's electricity consumption behavior data. The membership update formula is: in: DTW(s i ,v j ) represents the electricity consumption behavior data s of the i-th user i and the center v of the jth user category j The optimized DTW distance between them; S25: Repeat steps S23 and S24 until the membership matrix U converges, and divide the users into different categories according to the calculated memberships. The convergence means that the membership change after each iteration is lower than a preset threshold.

6. A method for evaluating interruptible load demand response potential according to claim 5, characterized in that: The optimized DTW distance between the user power consumption behavior data in step S24 includes: The DTW distance with boundary constraints is used to limit the search range of the DTW path and reduce the computational complexity. The specific process includes: S24.1: Initialize the cumulative cost matrix D for storing the distance between the first user's electricity usage behavior data s1 and the second user's electricity usage behavior data s2. The initial state of the cost matrix is ​​set to D(i,j)=∞, indicating that the distance between the first user's electricity usage behavior data before time i and the second user's electricity usage behavior data before time j is infinite. Initialize the strip boundary constraint width parameter w. S24.2: Calculate the data point distance d(i, j) between the data of the first user's electricity usage behavior data at time i and the data of the second user's electricity usage behavior data at time j, and the calculation formula is: d(i,j)=(s1(i)-s2(j)) 2 in: s1(i) represents the data of the first user's electricity consumption behavior at time i; s2(j) represents the data of the second user's electricity consumption behavior at time j; S24.3: Iteratively update and calculate the cumulative cost matrix, and use the D(L1, L2) value in the calculated cumulative cost matrix as the optimized DTW distance between the first user's power consumption behavior data and the second user's power consumption behavior data. The updated calculation formula is: D(i,j)=d(i,j)+min(D(i-1,j),D(i,j-1),D(i-1,j-1)) in: The iterative update is performed only within the strip boundary constraint area satisfying |ij|≤w, and other positions remain unchanged. L1 represents the length of the first user's power usage behavior data, and L2 represents the length of the second user's power usage behavior data.

7. The method for evaluating interruptible load demand response potential according to claim 1, characterized in that: The interruptible load demand prediction model is constructed in step S3, including: S31: using ADF test to detect the stability of electricity consumption behavior data of different categories of users. If the user electricity consumption behavior data is not stable, first-order difference processing is performed to obtain stable user electricity consumption behavior data; S32: using the autocorrelation function and the partial autocorrelation function to calculate the autoregressive order and the moving average order of the interruptible load demand prediction model according to the stabilized user power consumption behavior data; S33: Optimizing the model using the improved quasi-Newton algorithm to obtain the optimal model parameters; S34: Instantiate the model according to the optimal model parameters obtained by the optimization solution, and use the model instance to predict the power load, and use the difference between the predicted user power load and the actual user power load as the interruptible load demand of the corresponding user.

8. The method for evaluating interruptible load demand response potential according to claim 7, characterized in that: In the step S33, the improved quasi-Newton algorithm is used to optimize the model to obtain the optimal model parameters, including: S33.1: Construct the objective function based on the interruptible load demand prediction model, and the calculation formula is: in: N ′ Indicates the total number of user electricity usage behavior data included in the user category; θ represents the interruptible load demand prediction model parameter; ∈ i (θ) represents the prediction deviation of the interruptible load demand prediction model with θ as parameter; S33.2: Randomly initialize the model parameters θ0 and the Hessian matrix B0, and calculate the initial model prediction bias ∈(θ0) and gradient Initialize the maximum number of iterations; S33.3: Calculate the search direction based on the Hessian matrix and gradient calculated in the current iteration round. The calculation formula is: p k =-B k g k in: B k Represents the Hessian matrix obtained by k rounds of iterative calculation; g k Represents the gradient calculated after k rounds of iterations; S33.4: Update the model parameters according to the calculated search direction. The calculation formula is: i k+1 =θ k +a k p k in: p k Represents the search direction obtained by k rounds of iterative calculation; θ k represents the model parameters of round k; θ k+1 represents the model parameters of the k+1 round; α k represents the step size parameter calculated by the line search method in round k; S33.5: Update the gradient and Hessian matrix according to the calculated model parameters; S33.6: Iterate steps S33.3 to S33.5 until the maximum number of iterations is reached and the optimal model parameters are output.

9. The method for evaluating interruptible load demand response potential according to claim 8, characterized in that: In step S33.5, the gradient and Hessian matrix are updated according to the calculated model parameters, including: A1: Update the gradient g according to the calculated model parameters k+1 , and calculate the parameter difference s k =θ k+1 -θ k and the gradient difference y k =g k+1 -g k ; in: s k represents the parameter difference of k rounds; y k Represents the gradient difference of k rounds; A2: Use the calculated parameter difference and gradient difference to estimate and update the Hessian matrix. The calculation formula is: in: B k Represents the Hessian matrix obtained by k rounds of iterative calculation; B k+1 represents the Hessian matrix obtained by k+1 rounds of iterative calculation; T represents the transpose operation.

10. The method for evaluating interruptible load demand response potential according to claim 1, characterized in that: In step S4, the weighted sum of the predicted interruptible loads of different user categories is performed to obtain an interruptible load demand response potential evaluation result, including: S41: using AHP analytic hierarchy process to calculate the weight parameters of user interruptible load for different user categories; S42: Obtain an interruptible load demand response potential evaluation result by performing weighted summation based on the calculated user interruptible load weight parameter and the user interruptible load.

Images