A data-driven method and terminal for assessing demand response potential

Abstract

This invention discloses a data-driven method and terminal for assessing demand response potential. It constructs an objective function for demand-side response costs based on user demand response data and an electricity price model. By combining Wasserstein distance, a fuzzy set of response behaviors considering user behavioral uncertainty is obtained. Furthermore, the fuzzy set is corrected using moment information, avoiding assessment biases caused by a small sample size. This method not only considers the impact of user demand response on the potential assessment results but also incorporates the probability distribution information of user demand response behavior, thereby effectively narrowing the boundary of the uncertain set and better calculating the scale of user demand response. This helps to formulate more reasonable user demand response strategies and further optimize power system operation and planning.

Description

Technical Field

[0001] This invention relates to the field of power system operation technology, and in particular to a data-driven method and terminal for assessing demand response potential. Background Technology

[0002] In recent years, deep-seated contradictions such as low overall efficiency of the power system and insufficient complementarity among various power sources have become increasingly prominent, urgently requiring comprehensive optimization. The new power system is dominated by new energy sources, and with the massive influx of new energy, the randomness on both the source and load sides of the power system further increases. Furthermore, the strong volatility and low rotational inertia brought by new energy sources pose severe challenges to medium- and long-term operation simulation and planning. On the other hand, the integration of flexible DC transmission, large-scale offshore wind power, distributed photovoltaic grid connection, solar thermal power plants, electric vehicles, and charging stations into the power system has significantly increased the dimensions of issues related to power system operation and planning. Simultaneously, with the rapid development of smart grids, the conditions for power system supply and demand interaction are becoming increasingly mature, and the controllable scale on the demand side is gradually increasing. Demand response methods can effectively optimize power operation, thereby delaying or avoiding major power grid investment projects.

[0003] Currently, demand response has been considered to some extent in various aspects of power system operation and planning. By assessing the potential scale of user demand response, certain benefits can be obtained, optimizing system operation and delaying grid investment. Demand response guides users to adjust their energy consumption behavior through price or incentive signals, thereby improving the power system load curve and playing a role in peak shaving, valley filling, and promoting the consumption of renewable energy, thus having significant benefits in optimizing various aspects of system planning and operation. However, most current research treats demand response resources as fully controllable regulatory resources, and few studies consider the uncertainty of user response behavior when assessing the potential scale of demand response. In fact, demand response is a user-initiated behavior, and to some extent, uncertainty still exists from the users themselves. Furthermore, the number of users is enormous, and their behavioral characteristics are complex, making it difficult to characterize them through specific physical models.

[0004] Therefore, assessing the scale of user demand response potential is a key issue in power system planning and operation, playing a fundamental role in numerous studies and applications. However, existing assessment methods are too idealistic and struggle to create accurate assessment models. Summary of the Invention

[0005] The technical problem to be solved by this invention is to provide a data-driven method for assessing demand response potential, so as to better calculate the scale of user demand response and optimize the operation and planning of the power system.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0007] A data-driven method for assessing demand response potential includes the following steps:

[0008] Obtain user demand response data and electricity price model, and construct an objective function for demand-side response costs based on the demand response data and electricity price model;

[0009] Obtain user historical sample data, and construct a fuzzy set of user behavior uncertainty in the user historical sample data based on Wasserstein distance;

[0010] The fuzzy set is corrected based on the moment information;

[0011] The objective function is constrained based on the modified fuzzy set and the uncertainty of user behavior;

[0012] Based on the constrained objective function, evaluate the demand response potential boundary for uncertainty in user response behavior.

[0013] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows:

[0014] A data-driven demand response potential assessment terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the data-driven demand response potential assessment method described above.

[0015] The beneficial effects of this invention are as follows: It constructs an objective function for demand-side response costs based on user demand response data and an electricity price model, and obtains a fuzzy set of response behaviors considering user behavior uncertainty by combining Wasserstein distance. Furthermore, it corrects the fuzzy set using moment information, avoiding assessment bias caused by a small sample size. In other words, it not only considers the impact of user demand response on potential assessment results but also incorporates the probability distribution information of user demand response behavior, thereby effectively narrowing the boundary of the uncertain set and better realizing the measurement of user demand response scale. This helps to formulate more reasonable user demand response strategies and further optimize power system operation and planning. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the steps of a data-driven demand response potential assessment method according to an embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of the structure of a data-driven demand response potential assessment terminal according to an embodiment of the present invention;

[0018] Figure 3This is a sample data fitting curve diagram of a data-driven demand response potential assessment method according to an embodiment of the present invention;

[0019] Figure 4 This diagram shows a comparison between a data-driven demand response potential assessment method according to an embodiment of the present invention and existing assessment methods. Detailed Implementation

[0020] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0021] Please refer to Figure 1 A data-driven method for assessing demand response potential includes the following steps:

[0022] Obtain user demand response data and electricity price model, and construct an objective function for demand-side response costs based on the demand response data and electricity price model;

[0023] Obtain user historical sample data, and construct a fuzzy set of user behavior uncertainty in the user historical sample data based on Wasserstein distance;

[0024] The fuzzy set is corrected based on the moment information;

[0025] The objective function is constrained based on the modified fuzzy set and the uncertainty of user behavior;

[0026] Based on the constrained objective function, evaluate the demand response potential boundary for uncertainty in user response behavior.

[0027] As described above, the beneficial effects of this invention are as follows: It constructs an objective function for demand-side response costs based on user demand response data and an electricity price model, and obtains a fuzzy set of response behaviors considering user behavior uncertainty by combining Wasserstein distance. Furthermore, it corrects the fuzzy set using moment information, avoiding evaluation biases caused by a small sample size. This not only considers the impact of user demand response on potential assessment results but also incorporates the probability distribution information of user demand response behavior, thereby effectively narrowing the boundary of the uncertain set, better realizing the calculation of user demand response scale, and helping to formulate more reasonable user demand response strategies, further optimizing power system operation and planning.

[0028] Furthermore, the electricity price model includes an electricity price elasticity coefficient and an electricity price signal;

[0029] The user demand response data includes user load data. The objective function for constructing demand-side response costs includes:

[0030] A user energy consumption model is constructed based on the electricity price elasticity coefficient, electricity price signal, and user load data.

[0031] The objective function is obtained based on the user energy consumption model and the electricity price signal.

[0032] As described above, by constructing a user energy consumption model using the electricity price elasticity coefficient, electricity price signal, and user load data, the electricity price is correlated with user electricity consumption behavior, thereby improving the accuracy of the measurement of the user demand response scale.

[0033] Furthermore, the step of constructing a user energy consumption model based on the electricity price elasticity coefficient, the electricity price signal, and the user load data includes:

[0034] The electricity price elasticity coefficient and electricity price signal change according to different time periods;

[0035] Based on the electricity price elasticity coefficient and electricity price signal for different time periods, a time-of-use model is constructed, yielding:

[0036]

[0037]

[0038] Construct the user energy consumption model based on the time-segmentation model;

[0039] Wherein, ΔP ii This represents the change in user energy consumption behavior during time period i caused by the change in the electricity price signal during time period i; P i 0 For the user's original energy consumption behavior during time period i, ε ii Δρ is the self-elasticity coefficient for time period i; i This indicates the change in electricity price signal during time period i; Represents the original electricity price signal for time period i; ΔP ij ε represents the change in user energy consumption behavior in time period i caused by the change in electricity price signal in time period j; ij Δρ is the cross-elasticity coefficient between time period i and time period j. j This indicates the change in electricity price signal during time period i; This represents the original electricity price signal for time period i.

[0040] As described above, by constructing a time-segmented model based on the electricity price elasticity coefficient and electricity price signal for different time periods, a continuous long period can be segmented, thereby accurately reflecting the impact of electricity prices on users' energy consumption behavior at different time periods and effectively improving the accuracy of uncertainty measurement.

[0041] Furthermore, constructing the user energy consumption model based on the time-segmentation model includes:

[0042] Based on the influence between the current time period and other time periods, the user energy consumption model is obtained:

[0043]

[0044] Where, p i 1 This refers to the user's energy consumption behavior during time period i.

[0045] As described above, by taking into account the impact of changes in electricity price signals during this period and changes in user energy consumption behavior on other periods, and by adjusting the electricity price signals to obtain the user energy consumption model, the accuracy of the calculation of the user demand response scale is greatly improved.

[0046] Furthermore, obtaining the objective function based on the user energy consumption model and the electricity price signal includes:

[0047]

[0048] DR0 represents the demand-side response cost; P represents the original electricity price signal for time period i; i 0 This represents the user's original energy consumption behavior during time period i; P represents the adjusted electricity price signal at time i; i 1 This refers to the user's energy consumption behavior during time period i.

[0049] As described above, by using the original electricity price signal, the user's original energy consumption behavior, the changed electricity price signal, and the changes in the user's energy consumption behavior, the corresponding demand-side response cost can be obtained. This allows for accurate prediction of user demand response costs and the prediction of user behavior based on user demand response costs.

[0050] Furthermore, the fuzzy set for constructing user behavior uncertainty based on Wasserstein distance includes:

[0051] A user historical sample dataset is generated based on the aforementioned user historical sample data, denoted as:

[0052]

[0053] in, This represents the k-th historical sample data, and N represents the total number of historical samples for the user.

[0054] Calculate the Wasserstein distance:

[0055]

[0056]

[0057] Where ε(N) is the radius of the Wasserstein sphere; β is the confidence level of the constructed Wasserstein sphere; The average value of the user's historical sample data; C and α are both coefficients, and ||·||1 represents the 1-norm of the vector;

[0058] Construct a fuzzy set of user behavior uncertainty based on the Wasserstein distance.

[0059] As described above, the Wasserstein sphere, based on Wasserstein distance, describes the distance between probability distributions and is used to characterize the true and probability distributions of users' future response behaviors, thereby improving model accuracy.

[0060] Furthermore, the Wasserstein distance constructs a fuzzy set of user behavior uncertainty, including:

[0061]

[0062]

[0063]

[0064] In the formula: λ is the corresponding parameter; ε is the radius of the Wasserstein sphere; ·|| ∞ The infinite norm of a vector; (·) + This indicates taking the maximum value between the function within the parentheses and 0; σ represents the k-th normalized historical sample data; * The optimal solution to the optimization problem shown by σ is... The variance of the user's historical sample data; The mean of the user's historical sample data; This represents the possible future actual response behavior of the user, with the superscript ~ indicating that uncertainty is taken into account; U is a fuzzy set of user behavior uncertainty.

[0065] As described above, by solving for the optimal solution of σ, the fuzzy set of user behavior uncertainty can be determined based on the optimal solution of σ, which can effectively reflect the user's consideration of uncertainty.

[0066] Furthermore, correcting the fuzzy set based on the moment information includes:

[0067] U m =U∩U′

[0068]

[0069] Among them, U mThis represents the corrected fuzzy set, where the superscript m indicates correction; U′ represents the correction term that considers moment information; and τ is a parameter greater than or equal to 1.

[0070] As can be seen from the above description, considering that insufficient historical sample data of users may have an adverse impact on the evaluation results in actual operation, and that the possible future response behavior of users should have a similar distribution pattern to the historical sample data of users, the correction of the fuzzy set by moment information can effectively avoid the problem of evaluation performance error when the number of samples is small.

[0071] Furthermore, constraining the objective function based on the modified fuzzy set and the uncertainty of user behavior includes:

[0072]

[0073] Among them, κ, y n (n = 1, 2, 3, ..., N), ζ n (n=1,2,3…,N), Γ are auxiliary variables; ||·|| ∞ a, β are the infinite norms of the contents within the parentheses. i For the corresponding coefficients; The change in response potential is caused by the uncertainty of the user's response behavior to changes in price signals at time i.

[0074] As can be seen from the above description, the above formula makes the modified fuzzy set have similar moment information characteristics to the user's historical data, and by setting auxiliary variables, it is more conducive to forming a solution form for commercial solvers, which facilitates the solution model.

[0075] Another embodiment of the present invention provides a data-driven demand response potential assessment terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the data-driven demand response potential assessment method described above.

[0076] The data-driven demand response potential assessment method and terminal of the present invention can be applied to assess user demand response potential, so as to better realize the calculation of user demand response scale, optimize power system operation and planning, and delay grid investment. The following is a detailed description of the implementation method:

[0077] Example 1

[0078] Please refer to Figure 1 A data-driven method for assessing demand response potential includes the following steps:

[0079] S1. Obtain user demand response data and an electricity price model; construct an objective function for demand-side response costs based on the demand response data and the electricity price model; describe the user's response to changes in electricity price (incentive) signals using the electricity price elasticity coefficient ε, wherein the electricity price model includes the electricity price elasticity coefficient and the electricity price signal, specifically:

[0080] S11. Construct a user energy consumption model based on the electricity price elasticity coefficient, electricity price signal, and user load data:

[0081]

[0082] In the formula, ΔP represents the change in user energy consumption behavior; P 0 The user's original load; Δρ represents the change in the electricity price (incentive) signal; ρ 0 This is the original electricity price (incentive) signal;

[0083] In an optional implementation, considering that changes in electricity price signals at different times within a continuous period (e.g., 24 hours in a practical sense) will affect users' energy consumption behavior during that time period, the electricity price elasticity coefficient and electricity price signal are configured to change according to different time periods. Then, a time-of-use model is constructed based on the electricity price elasticity coefficient and electricity price signal for different time periods, resulting in:

[0084]

[0085]

[0086] Based on the influence between the current time period and other time periods, the user energy consumption model is obtained. That is, after adjusting the price signal by taking into account the influence of the current time period and other time periods, the user's energy consumption behavior P is obtained. i 1 It can be represented as:

[0087]

[0088] Wherein, ΔP ii This represents the change in user energy consumption behavior during time period i caused by the change in the electricity price signal during time period i; P i 0 For the user's original energy consumption behavior during time period i, ε ii Δρ is the self-elasticity coefficient for time period i; i This indicates the change in electricity price signal during time period i; Represents the original electricity price signal for time period i; ΔP ij ε represents the change in user energy consumption behavior in time period i caused by the change in electricity price signal in time period j; ij Δρ is the cross-elasticity coefficient between time period i and time period j. j This indicates the change in electricity price signal during time period i; P represents the original electricity price signal for time period i; i 1 This refers to user energy consumption behavior during time period i.

[0089] S12. The objective function obtained based on the user energy consumption model and electricity price signal includes:

[0090]

[0091] Because of the adjustment to the electricity price signal, the electricity sales revenue of the power grid company will change, which will be included in the cost of implementing price-based demand-side response and recorded as DR0; that is, DR0 is the demand-side response cost. P represents the original electricity price signal for time period i; i 0 This represents the user's original energy consumption behavior during time period i; P represents the adjusted electricity price signal at time i; i 1 This refers to user energy consumption behavior during time period i.

[0092] S2. Obtain user historical sample data, and construct a fuzzy set of user behavior uncertainty in the user historical sample data based on Wasserstein distance; wherein, in another optional implementation, constructing the fuzzy set of user behavior uncertainty in the user historical sample data by means of multi-scene information characterization, KL divergence characterization or moment information characterization can achieve similar results; the main advantage of using Wasserstein is that it has high computational efficiency.

[0093] S21. Generate a user historical sample dataset based on the aforementioned user historical sample data, denoted as:

[0094]

[0095] in, This represents the k-th historical sample data, and N represents the total number of historical samples for the user.

[0096] S22. The Wasserstein sphere, based on Wasserstein distance, is used to describe the distance between probability distributions, characterizing the true distribution and probability distribution of a user's future response behavior. In response potential assessment, it can be used to characterize the distance between the distribution of a user's future actual response behavior and the distribution of historical samples. A feasible Wasserstein sphere radius ε can be obtained by the following method:

[0097]

[0098]

[0099] Where ε(N) is the radius of the Wasserstein sphere; β is the confidence level of the constructed Wasserstein sphere; The average value of the user's historical sample data; C and α are both coefficients, and ||·||1 represents the 1-norm of the vector;

[0100] S23. Construct a fuzzy set representing the uncertainty of user behavior based on the Wasserstein distance; that is, based on the radius ε of the Wasserstein sphere, further construct a fuzzy set U describing the user's future response behavior. The fuzzy set can be obtained by solving the following problem:

[0101]

[0102] In the formula: λ is the corresponding parameter; ε is the radius of the Wasserstein sphere; ||·|| ∞ The infinite norm of a vector; (·) + This indicates taking the maximum value between the function within the parentheses and 0; σ represents the k-th normalized historical sample data; * The optimal solution to the optimization problem shown by σ is... The variance of the user's historical sample data; The mean of the user's historical sample data; This represents the possible future actual response behavior of the user; the superscript ~ indicates that uncertainty is taken into account; U is a fuzzy set of user behavior uncertainty.

[0103] S3. Correct the fuzzy set according to the moment information;

[0104] U m =U∩U′

[0105]

[0106] Among them, U m The modified fuzzy set is represented by the superscript m, which indicates the modification; U′ represents the modification term that considers moment information; τ is a parameter greater than or equal to 1; the moment information mainly refers to the second moment information, which mathematically means the expectation of the square of the difference between the random variable and the expectation, and practically means that the response behavior of future users should have a certain similarity with the historical data pattern, so as to eliminate abnormal data that differs greatly from historical samples.

[0107] S4. Constrain the objective function based on the modified fuzzy set and the uncertainty of user behavior. Specifically: Based on the modified fuzzy set, the distribution of the user's possible future response potential can be obtained. Furthermore, considering the uncertainty of user behavior, the constructed demand response constraints and objective function should be rewritten as follows:

[0108]

[0109]

[0110]

[0111]

[0112]

[0113] Among them, κ, y n (n = 1, 2, 3, ..., N), ζ n (n=1,2,3…,N), Γ are auxiliary variables; ||·|| ∞ The infinite norm of the content in parentheses; a and β i These are preset coefficients; The change in response potential due to the uncertainty of the user's response behavior to changes in price signals at time i;

[0114] S5. Based on the constrained objective function, evaluate the demand response potential boundary of user response behavior uncertainty; obtain the demand response potential boundary of user response behavior uncertainty.

[0115] Example 2

[0116] The application of the data-driven demand response potential assessment method in Example 1 in a specific scenario is as follows: User demand response data with a sample size of N=10000 is generated based on Monte Carlo simulation; due to the uncertainty of user response behavior, the actual user response load and response electricity will deviate from the expectation to a certain extent. Therefore, it is assumed that the user response behavior follows a normal distribution, and the generated sample data and fitted curve are as follows: Figure 3 As shown;

[0117] The sample data is processed according to the data-driven demand response potential assessment method as described in Example 1;

[0118] S2. Obtain historical sample data of users and construct a fuzzy set of user behavior uncertainty based on Wasserstein distance;

[0119] S3. Then, the fuzzy set is corrected based on the moment information;

[0120] S4. Constrain the objective function based on the modified fuzzy set and the uncertainty of user behavior;

[0121] S5. Finally, the potential scale of demand response considering the uncertainty of user response behavior is described as follows: Figure 4As shown; simultaneously, to verify that the method of this embodiment can effectively utilize the probability distribution information of historical data, thereby achieving a better balance between the economy and conservatism in assessing the scale of response potential, the evaluation results of this embodiment are compared with other schemes; compared with the response potential assessment results that only consider the expected probability distribution of user behavior, the evaluation results of this embodiment also take into account the uncertainty of user behavior, broadening the boundary of the response potential scale and exhibiting better robustness; compared with the response potential assessment results that consider the extreme cases of various uncertain user behaviors, the method of this embodiment further considers the probability of various response behaviors occurring, effectively shrinking the boundary of the uncertainty set, thereby exhibiting better economy; therefore, the evaluation results of this embodiment can directly achieve a good balance between economy and conservatism, and have good evaluation results.

[0122] Example 3

[0123] A data-driven demand response potential assessment terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps in the data-driven demand response potential assessment method as described in Embodiment 1.

[0124] In summary, the data-driven demand response potential assessment method and terminal provided by this invention constructs an objective function for demand-side response costs based on user demand response data and an electricity price model. It constructs a user energy consumption model using electricity price elasticity coefficients and signals from different time periods, as well as user load data, thus linking electricity prices with user electricity consumption behavior. This improves the accuracy of calculating the scale of user demand response. Furthermore, it uses Wasserstein distance to obtain a fuzzy set of response behaviors that considers the uncertainty of user behavior, and further corrects the fuzzy set using moment information. This avoids assessment bias caused by a small sample size. In other words, it not only considers the impact of user demand response on the potential assessment results but also incorporates the probability distribution information of user demand response behavior, thereby effectively narrowing the boundary of the uncertain set and better calculating the scale of user demand response. This helps to formulate more reasonable user demand response strategies and further optimize power system operation and planning.

[0125] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A data-driven method for assessing demand response potential, characterized in that, Including the following steps: Obtain user demand response data and electricity price model, and construct an objective function for demand-side response costs based on the demand response data and electricity price model; Obtain user historical sample data, and construct a fuzzy set of user behavior uncertainty in the user historical sample data based on Wasserstein distance; The fuzzy set is corrected based on the moment information; The objective function is constrained based on the modified fuzzy set and the uncertainty of user behavior; Based on the constrained objective function, evaluate the demand response potential boundary of uncertainty in user response behavior; The fuzzy set for constructing user behavior uncertainty based on Wasserstein distance includes: A user historical sample dataset is generated based on the aforementioned user historical sample data, denoted as: ； in, This represents the k-th historical sample data, and N represents the total number of historical samples for the user. Calculate the Wasserstein distance: ； ； in, The radius of the Wasserstein sphere; The confidence level of the constructed Wasserstein sphere; The average value of historical sample data for users; C and All are coefficients. The 1-norm of a vector; Construct a fuzzy set of user behavior uncertainty based on the Wasserstein distance; The Wasserstein distance constructs a fuzzy set of user behavior uncertainty, including: ； ； ； In the formula: For the corresponding parameters; The radius of the Wasserstein sphere; The infinite norm of a vector; This indicates taking the maximum value between the function within the parentheses and 0; This represents the k-th normalized historical sample data; for The optimal solution to the optimization problem shown; The variance of the user's historical sample data; This indicates the possible future actual response behavior of the user; the superscript ~ indicates that uncertainty is taken into account. A fuzzy set representing the uncertainty of user behavior; Correcting the fuzzy set based on the moment information includes: ； in, This represents the corrected fuzzy set, where the superscript m indicates the correction; This indicates a correction term that takes into account moment information; Parameters that are greater than or equal to 1; The constraint on the objective function based on the modified fuzzy set and the uncertainty of user behavior includes: ； in, , ( ), ( ), As an auxiliary variable; The infinite norm of the content within the parentheses; a and These are preset coefficients; The change in response potential is caused by the uncertainty of the user's response behavior to changes in price signals at time i.

2. The data-driven demand response potential assessment method according to claim 1, characterized in that, The electricity price model includes an electricity price elasticity coefficient and an electricity price signal; The user demand response data includes user load data. The objective function for constructing demand-side response costs includes: A user energy consumption model is constructed based on the electricity price elasticity coefficient, electricity price signal, and user load data. The objective function is obtained based on the user energy consumption model and the electricity price signal.

3. The data-driven demand response potential assessment method according to claim 2, characterized in that, The step of constructing a user energy consumption model based on the electricity price elasticity coefficient, the electricity price signal, and the user load data includes: The electricity price elasticity coefficient and electricity price signal change according to different time periods; Based on the electricity price elasticity coefficient and electricity price signal for different time periods, a time-of-use model is constructed, yielding: ； ； Construct the user energy consumption model based on the time-segmentation model; in, This indicates the change in user energy consumption behavior during time period i caused by the change in the electricity price signal during time period i. For the user's original energy consumption behavior during time period i, Let i be the self-elasticity coefficient for time period i; This indicates the change in electricity price signal during time period i; This represents the original electricity price signal for time period i; This indicates the change in user energy consumption behavior during time period i caused by the change in electricity price signal during time period j; The cross-elasticity coefficient between time period i and time period j; This indicates the change in electricity price signal during time period j; This represents the original electricity price signal for time period j.

4. The data-driven demand response potential assessment method according to claim 3, characterized in that, The step of constructing the user energy consumption model based on the time-segmentation model includes: Based on the influence between the current time period and other time periods, the user energy consumption model is obtained: ； in, This refers to the user's energy consumption behavior during time period i.

5. The data-driven demand response potential assessment method according to claim 2, characterized in that, The step of obtaining the objective function based on the user energy consumption model and the electricity price signal includes: ； in, For demand-side response costs; This represents the original electricity price signal for time period i; This represents the user's original energy consumption behavior during time period i. The adjusted electricity price signal at time i; This refers to the user's energy consumption behavior during time period i.

6. A data-driven demand response potential assessment terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements each step of the data-driven demand response potential assessment method as described in any one of claims 1-5.

Images