Air conditioner load regulation and control evaluation method based on improved AHP-entropy weight combination weighting method

Abstract

The invention discloses an air conditioner load regulation and control evaluation method based on an improved AHP-entropy weight combination weighting method, and aims to solve the problems that traditional evaluation excessively depends on expert experience and lacks a data objectivity and subjective and objective information fusion mechanism. According to the technical scheme, the method comprises the following steps: constructing an index system for evaluating effectiveness, reliability and economy, and designing a quantization formula of each index; a subjective weight is obtained by adopting an analytic hierarchy process, an objective weight is obtained by adopting an entropy weight method, a game theory combination weighting method is introduced, and an optimal combination coefficient and a comprehensive weight are solved by constructing an optimization model for minimizing the difference between the subjective weight and the objective weight; based on this, comprehensive evaluation of the regulation effect is realized. The method gives consideration to experience and data values, is objective and accurate in evaluation result, can balance power grid demands and user experience, supports regulation and control system optimization and power grid dispatching, and is adaptive to multi-scene application.

Description

Technical Field

[0001] This invention relates to the field of new energy microgrid power generation control technology, specifically to an evaluation method for air conditioning load regulation based on an improved AHP-entropy weight combination weighting method. Background Technology

[0002] With the transformation and upgrading of the energy structure, the power system's ability to regulate demand-side resources is constantly increasing. Especially against the backdrop of large-scale integration of new energy sources and widening peak-valley differences in power load, demand response is gradually becoming a key technical means to improve the flexibility and stability of the power grid. As the most important adjustable resource on the residential side, air conditioning load has a fast response speed and great regulation potential, and has broad application prospects in demand response.

[0003] However, in the actual operation of air conditioning group control, the load regulation effect is often affected by a variety of factors, such as load adjustment rate, adjustable capacity ratio, user comfort, communication reliability, and response delay. Traditional regulation evaluation methods often rely on a single or few indicators, such as only examining the reduction amount or user comfort, lacking multi-dimensional comprehensive consideration and making it difficult to reflect the overall level of regulation.

[0004] Existing research attempts to establish a multi-indicator evaluation system, but the following problems often exist in the allocation of indicator weights: 1. Over-reliance on expert experience: The traditional Analytic Hierarchy Process (AHP) relies mainly on expert scoring, resulting in highly subjective results; 2. Lack of data-driven objectivity: Although the entropy weight method can reflect the differences in indicator data, it may lead to results that deviate from actual needs when there is a lack of expert guidance; 3. Lack of a reasonable integration mechanism: Existing methods often use AHP or entropy weight method alone, lacking a scientific combination method to balance subjective and objective information. Summary of the Invention

[0005] The purpose of this invention is to address the problems of traditional air conditioning load control evaluation methods, such as the use of single indicators, lack of multi-dimensional comprehensive consideration, over-reliance on expert experience or simple emphasis on data characteristics, and lack of a scientific integration mechanism of subjective and objective information, resulting in insufficient accuracy and practicality of evaluation results. Therefore, this invention proposes an air conditioning load control performance evaluation method based on an improved AHP-entropy weighting method. By constructing a two-level hierarchical indicator system, subjective weights are first calculated using the Analytic Hierarchy Process (AHP) combined with expert experience. Then, objective weights are determined based on actual operating data characteristics using the entropy weighting method. Finally, a game theory optimization model is introduced to integrate subjective and objective weights to obtain a comprehensive weight, achieving a comprehensive, scientific, and accurate evaluation of the air conditioning load control effect, providing a reliable reference for power grid operation optimization and demand response compensation settlement.

[0006] The technical solution provided by this invention is as follows: A method for evaluating the performance of air conditioning load control based on an improved AHP-entropy weight combination method, comprising: Based on the characteristics of air conditioning load group regulation, an air conditioning load regulation index system is constructed; Based on the expert scoring method, the relative importance of each indicator in the air conditioning load control index system is assigned to each level of the index system and a judgment matrix is ​​constructed. The basic subjective weights are calculated based on the judgment matrix using the analytic hierarchy process. Collect multi-dimensional control data corresponding to indicators during the actual operation of the air conditioning load group and perform standardization processing. Based on the standardized multi-dimensional control data, calculate the information entropy value of each indicator using the entropy weight method, and determine the basic objective weights based on the entropy value. A comprehensive weight optimization model is constructed based on the game theory combinatorial weighting method. Subjective weight vectors are obtained through each basic subjective weight, and objective weight vectors are obtained through each basic objective weight. The comprehensive weight optimization model is solved with the goal of minimizing the comprehensive deviation between the comprehensive weight and the subjective weight vector plus the comprehensive deviation between the comprehensive weight and the objective weight vector. The optimal combination coefficients of subjective and objective weights are obtained, and the comprehensive weights of each indicator are obtained based on the optimal combination coefficients, the subjective weight vector, and the objective weight vector. The operation data of the air conditioning load control process to be evaluated is obtained and standardized to obtain the standardized values ​​of each indicator. The standardized values ​​of each indicator are then weighted and summed with their corresponding comprehensive weights to calculate the comprehensive evaluation result of the air conditioning load control effect.

[0007] Preferably, the air conditioning load control index system includes a primary index system, which includes effectiveness indexes, reliability indexes, and economic indexes.

[0008] Each indicator in the primary indicator system corresponds to at least one secondary indicator; The secondary indicators corresponding to the effectiveness indicators include load regulation rate, adjustable capacity ratio, response accuracy, and load fluctuation rate; the secondary indicators corresponding to the reliability indicators include communication reliability, regulation interruption rate, response delay time, and data loss rate; and the secondary indicators corresponding to the economic indicators include user comfort deviation and load reduction amount.

[0009] The specific settings for the secondary indicators are as follows: Load regulation rate R power This represents the change in air conditioning load per unit time, reflecting the response speed of the group control system, as shown below: in, P target For the target load, Pinitial For the initial load, T adjust The total time required to complete the composite adjustment; Adjustable capacity percentage C afj This represents the ratio of the actual adjustable capacity to the rated capacity under current conditions, reflecting the release potential of group control, as shown below: in, P i,max For a single user, the maximum power within the permissible temperature range. P i,min This is the minimum power required for a single user within the permissible temperature range. P rated This refers to the rated capacity of the air conditioner. Response progress delta error The following table shows the degree of agreement between the actual load adjustment results and the target value: in, P actual The actual regulating power of the air conditioning cluster. P target The target power required by the power grid; Load fluctuation sigma p The power fluctuation amplitude during the regulation process reflects the stability of the execution, as shown below: in, P t Let be the total power of the air conditioning cluster at time t. T To detect the length of the time window, The average power during the control period; Communication reliability C success This indicates the ratio of power grid command issuance to user confirmation of receipt, as detailed below: in, N received Confirm the number of instructions received by the user. N sent This represents the total number of instructions issued by the power grid. Adjusting the interrupt rate I break The percentage of time interrupted during the regulation process is as follows: in, T interruptTo calculate the cumulative interruption duration, T total Adjust the total duration according to the plan; Response latency T delay The average delay from the issuance of a power grid command to the initiation of a user load response is shown below: in, N The number of air conditioners participating in regulation, T i,start Let the corresponding time start for the i-th user. t cmd This refers to the time when the power grid issues the instruction; Data missing rate L loss This indicates the proportion of missing data points during the data collection process, reflecting data integrity, as shown below: in, M loss For the number of missing households, M Total number of users; User comfort Score Indicates the degree to which room temperature deviates from the user's comfort range. in, CTO This refers to the cumulative degree to which the temperature deviates from the comfortable range within the control period; CTO max Δ represents the degree to which the temperature deviates from the comfortable range during the control period. T To regulate the temperature difference during the period, delta T max The maximum temperature difference during the regulation period, T t Let t be the room temperature at the t-th time point; T 0 The initial room temperature; Δ t To regulate the duration; Load reduction LRA-E This indicates the total electricity load reduction of air conditioning groups during the control period. in, P baseline Let t be the baseline load power; P real (t) represents the actual operating power at time t, Δ t To regulate the duration.

[0010] When evaluating air conditioning load control, it is first necessary to construct a scientific and reasonable indicator system. Based on the characteristics of air conditioning group control in actual operation, this invention establishes a two-level hierarchical model with primary and secondary indicators, considering the effectiveness, reliability, and economy of control. This model reflects the dynamic characteristics of air conditioning load response while also taking into account the needs of both the power grid and the user side.

[0011] Effectiveness indicators primarily reflect the actual response and control effect of air conditioning groups after the power grid issues control commands. The core issue is whether the load adjusts according to the target curve required by the power grid. Evaluation dimensions include adjustment speed, adjustment capacity, response accuracy, and load fluctuation. Effectiveness indicators embody the "execution power" of the control system and are the primary factor in measuring the effectiveness of group control. In demand response compensation mechanisms, effectiveness indicators often directly determine the compensation amount and the reliability of power grid dispatch.

[0012] Reliability indicators primarily reflect the stability and sustainability of the control process. Air conditioning group control involves communication and data transmission among a large number of users; any signal interruption or data loss will severely impact the control effect. Evaluation dimensions include communication reliability, control interruption rate, response delay time, and data loss rate. Reliability indicators emphasize "process assurance," ensuring that control actions do not fail due to communication link or equipment malfunctions. For the power grid, reliability indicators are a necessary condition for the long-term implementation of control measures in actual operation.

[0013] Economic indicators primarily examine the cost-benefit characteristics of regulation from both the user and power grid perspectives. For users, regulation should not lead to an excessive decrease in indoor comfort, otherwise user participation will be affected. For the power grid, it is necessary to examine the actual load-reduction capacity of air conditioning groups to ensure the effectiveness of peak-valley regulation. Evaluation dimensions include user comfort and load reduction. Economic indicators reflect the sustainability and promotion potential of regulation behavior and are a key prerequisite for achieving large-scale demand response.

[0014] To more precisely reflect the connotation of the primary indicators, this invention sets secondary indicators under each primary indicator, totaling 10, as follows: The secondary indicators corresponding to the effectiveness indicators include load adjustment rate, adjustable capacity ratio, response accuracy, and load fluctuation rate; the secondary indicators corresponding to the reliability indicators include communication reliability, adjustment interruption rate, response delay time, and data loss rate; the secondary indicators corresponding to the economic indicators include user comfort deviation and load reduction amount.

[0015] After determining the evaluation index system, it is necessary to assign weights to the importance of each index. Since the indexes involve multiple factors from both the power grid and user sides, relying solely on objective data may not fully reflect the decision-making preferences of managers and the experience of industry experts. Therefore, the Analytic Hierarchy Process (AHP) is introduced to calculate the index weights.

[0016] Preferably, the method of assigning relative importance to each indicator based on expert scoring and constructing a judgment matrix, calculating the subjective weight of each indicator based on the judgment matrix using the analytic hierarchy process (AHP), and performing a consistency check on the judgment matrix to obtain the subjective weights of each indicator that meet the consistency requirements, includes: Based on the Saaty scaling method, multiple experts were invited to assign relative importance to each indicator at the same level and construct a judgment matrix for the indicators at the same level. Scale the 1-9 scale (Saaty scale method): 1 indicates that the two indicators are equally important; 3 indicates that the former is slightly more important than the latter; 5 indicates that the former is significantly more important than the latter; 7 indicates that the former is strongly more important than the latter; 9 indicates that the former is extremely more important than the latter; 2, 4, 6, and 8 are intermediate values; if the importance of indicator i relative to j is... ,but .

[0017] The geometric mean of each row of the judgment matrix corresponding to each level indicator is calculated based on the geometric mean method. After the geometric mean vector is normalized, the subjective weight of the first-level indicator and the local subjective weight of the second-level indicator are obtained. Based on the largest eigenvalue γ max Calculate the consistency index CI and the consistency ratio CR. When CR < 0.1, the judgment matrix passes the consistency test, which means that the subjective weights of each index are credible. The subjective weights of the calculated primary indicators are multiplied by the local subjective weights of the corresponding secondary indicators to obtain the subjective weights of each secondary indicator. The local subjective weights of the secondary indicators refer to the relative importance of the secondary indicator within the category of primary indicators, and need to be multiplied by the weights of the primary indicators to be converted into the global importance of the entire evaluation system.

[0018] In evaluating the effectiveness of actual load regulation, the importance of indicators depends not only on expert experience (subjective factors) but also on the variability of actual operating data. To ensure the objectivity of weight allocation, this invention introduces the Entropy Weight Method, which assigns weights based on the dispersion of each indicator's data distribution.

[0019] Preferably, the step of collecting multi-dimensional control data corresponding to indicators during the actual operation of the air conditioning load group and performing standardized processing, calculating the information entropy value of each indicator based on the processed multi-dimensional control data using the entropy weight method, and determining the objective weight of each indicator based on the entropy value, includes: The collected data on load regulation rate, adjustable capacity percentage, response accuracy, communication reliability, and load reduction are standardized, and the standardized value of the i-th data sample corresponding to the j-th indicator is obtained. y ij The expression is equal to the original data of the i-th sample of the indicator minus the minimum value among all samples of the indicator's original data, divided by the maximum value among all samples of the indicator's original data minus the minimum value among all samples of the indicator's original data. in, Let max( be the original value of the i-th data sample on the j-th indicator) x j ) represents the maximum original value of the j-th index among all samples, min( x j ) represents the minimum original value of the j-th index among all samples. The standardized value of the i-th sample data on the j-th indicator; For load fluctuation rate, regulation interruption rate, response delay time, data missing rate, and comfort deviation, standardization is performed. The standardized value yij of the i-th data sample corresponding to the j-th indicator is equal to the original data of the i-th sample of the indicator minus the maximum value among all original data of the indicator, and then divided by the maximum value among all original data of the indicator minus the minimum value among all original data of the indicator. The expression is as follows: in, Let max( be the original value of the i-th data sample on the j-th indicator) x j ) represents the maximum original value of the j-th index among all samples, min( x j ) represents the minimum original value of the j-th index among all samples. The standardized value of the i-th sample data on the j-th indicator; Based on standardized numerical values, the basic objective weights are calculated as follows: Calculate the weight of the i-th data sample on the j-th indicator. Where m represents the number of samples. This represents the weight of the i-th data sample on the j-th indicator; For index j, its entropy value e j Defined as: in: The entropy value is guaranteed to be in the interval [0, 1]; when all When they are completely equal, the entropy value This indicates that the indicator contributes no information; when The greater the difference, the smaller the entropy value, indicating that the indicator contributes more to the overall difference; Based on the deviation g of the entropy value i, the deviation of the secondary indicators. i : The j-th basic objective weight is defined as: in: is the j-th basic objective weight; n is the number of secondary indicators.

[0020] Preferably, a comprehensive weight optimization model is constructed based on the game theory combinatorial weighting method. Subjective weight vectors are obtained through each basic subjective weight, and objective weight vectors are obtained by combining each basic objective weight. The objective is to minimize the sum of the comprehensive deviation between the comprehensive weight and the subjective weight vector and the comprehensive deviation between the comprehensive weight and the objective weight vector. The comprehensive weight optimization model is then solved to obtain the optimal combination coefficients of the subjective and objective weights. Based on these optimal combination coefficients, the subjective weight vector, and the objective weight vector, the comprehensive weights of each indicator are obtained, including: The comprehensive weight optimization model is constructed as follows: in, α 1 Subjective weighting coefficient; α 2 For objective weighting coefficients, For comprehensive weighting, The subjective weight vector is a set of basic subjective weights of all secondary indicators arranged in a fixed order. The objective weight vector is the set of basic objective weights of all secondary indicators arranged in a fixed order. The objective function is defined as minimizing the sum of the combined deviation between the combined weights and the subjective weights vector, and the combined deviation between the combined weights and the objective weights vector, as shown below: Will WSubstituting the expression into the objective function and obtaining the coefficients through mathematical transformation, we can solve for the coefficients. α 1 Sum of coefficients α 2 The system of linear equations is as follows: in, for The transpose of , for The transpose of ; Solve the system of linear equations to obtain the coefficients. α 1 Sum of coefficients α 2 For coefficients α 1 Sum of coefficients α 2 Normalization is performed to obtain the optimal combination coefficients. Based on the optimal combination coefficients, subjective weight vector, and objective weight vector, the comprehensive weights of each indicator are obtained.

[0021] This invention also provides an air conditioning load control evaluation system based on an improved AHP-entropy weight combination method, comprising: The air conditioning load control index system construction module is used to construct an air conditioning load control index system based on the control characteristics of air conditioning load groups. The subjective weight setting module is used to assign relative importance to each indicator in the air conditioning load control index system based on the expert scoring method and construct a judgment matrix. The basic subjective weights are calculated based on the judgment matrix using the analytic hierarchy process. The objective weight setting module is used to collect multi-dimensional control data corresponding to indicators during the actual operation of the air conditioning load group and perform standardized processing. Based on the standardized multi-dimensional control data, the information entropy value of each indicator is calculated by the entropy weight method, and the basic objective weights are determined based on the entropy value. The final comprehensive weight generation module is used to construct a comprehensive weight optimization model based on the game theory combination weighting method. It obtains a subjective weight vector through each basic subjective weight and an objective weight vector through each basic objective weight. With the goal of minimizing the comprehensive deviation between the comprehensive weight and the subjective weight vector plus the comprehensive deviation between the comprehensive weight and the objective weight vector, it solves the comprehensive weight optimization model to obtain the optimal combination coefficient of subjective and objective weights. Based on the optimal combination coefficient, the subjective weight vector, and the objective weight vector, it obtains the comprehensive weight of each indicator. The comprehensive evaluation result generation module is used to acquire the operation data of the air conditioning load control process to be evaluated and perform standardization processing to obtain the standardized values ​​of each indicator. The standardized values ​​of each indicator are then weighted and summed with their corresponding comprehensive weights to calculate the comprehensive evaluation result of the air conditioning load control effect.

[0022] The present invention also provides a computer storage medium storing a computer program, which, when executed by a processor, implements the steps of the air conditioning load control evaluation method based on the improved AHP-entropy weight combination method as described above.

[0023] The present invention also provides an electronic device, including a memory and a processor: the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, wherein when the computer-executable instructions are executed by the processor, the steps of the air conditioning load control evaluation method based on the improved AHP-entropy weight combination weighting method described above are implemented.

[0024] The beneficial effects of this invention are as follows: This invention employs a process of constructing a judgment matrix through expert scoring → calculating subjective weights using AHP (Advanced Person Hierarchy Process) → consistency verification. This process preserves the experts' experiential understanding of the characteristics of the air conditioning group control industry while eliminating unreasonable biases in expert scoring through consistency verification. Furthermore, the subsequent integration of subjective weights with data-driven objective weights further mitigates the one-sidedness of a single expert's experience, ensuring that the weighting results conform to the industry logic of actual control scenarios while avoiding interference from subjective assumptions.

[0025] This invention first collects multi-dimensional control data from the actual operation of air conditioning groups, and then calculates objective weights using the entropy weight method after standardization processing, which accurately reflects the actual differences in indicator data. At the same time, by combining weights with subjective weights, the differences in objective data serve the actual needs of the industry, avoiding the risk that the data from the single entropy weight method may be reasonable but not consistent with the control scenario, and making the weight results both objective and practical for the industry.

[0026] This invention introduces a game theory-based combination weighting method to construct an optimization model that minimizes the difference between subjective and objective weights. By analytically solving the model, the optimal combination coefficients of subjective and objective weights are obtained. This approach fully utilizes the industry experience-guided value of subjective weights while also leveraging the actual difference value of objective weight data, achieving a scientific integration of the two. The resulting comprehensive weights are more comprehensive and reliable, providing accurate indicators of importance for subsequent evaluations.

[0027] Based on the aforementioned breakthroughs in traditional methods, this invention ultimately achieves a scientific evaluation process from indicator system to comprehensive evaluation: the indicator system covers the dimensions of effectiveness, reliability, and economy, and combined with quantitative formulas, transforms abstract control concepts into measurable data; the comprehensive evaluation results not only accurately quantify the actual effect of air conditioning group control, but also guide the optimization of the control system and the grid demand response scheduling, while balancing grid demand and user experience, ultimately enhancing the practical application value of air conditioning load group control. Attached Figure Description

[0028] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 A flowchart of an air conditioning load control evaluation method based on an improved AHP-entropy weight combination method is provided in Example 1; Figure 2 This is a block diagram of an air conditioning load control evaluation system based on an improved AHP-entropy weight combination method, provided in Example 2. Detailed Implementation

[0029] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0030] Example 1 like Figure 1 As shown in the figure, this embodiment presents a method for evaluating the performance of air conditioning load control based on an improved AHP-entropy weight combination method, including the following steps: S1. Construct an air conditioning load control index system based on the characteristics of air conditioning load group control; The air conditioning load control index system includes a primary index system composed of effectiveness, reliability, and economic indicators. Each primary index has secondary indicators, as shown below: The secondary indicators set for the effectiveness index (A1) include load regulation rate, adjustable capacity percentage, response accuracy, and load fluctuation rate. The secondary indicators set for reliability index (A2) include communication reliability, adjustment interruption rate, response latency, and data loss rate; The secondary indicators set for the economic indicators (A3) include user comfort deviation and load reduction.

[0031] The above-mentioned secondary indicators include: B1: Load adjustment rate R power This represents the change in air conditioning load per unit time, reflecting the response speed of the group control system, as shown below: in, P target For the target load, P initial For the initial load, T adjust The total time required to complete the composite adjustment; B2: Adjustable capacity percentage C afjThis represents the ratio of the actual adjustable capacity to the rated capacity under current conditions, reflecting the release potential of group control, as shown below: in, P i,max For a single user, the maximum power within the permissible temperature range. P i,min This is the minimum power required for a single user within the permissible temperature range. P rated This refers to the rated capacity of the air conditioner. B3: Response Progress delta error The following table shows the degree of agreement between the actual load adjustment results and the target value: in, P actual The actual regulating power of the air conditioning cluster. P target The target power required by the power grid; B4: Load Fluctuation Rate sigma p The power fluctuation amplitude during the regulation process reflects the stability of the execution, as shown below: in, P t Let be the total power of the air conditioning cluster at time t. T To detect the length of the time window, The average power during the control period; B5: Communication Reliability C success This indicates the ratio of power grid command issuance to user confirmation of receipt, as detailed below: in, N received Confirm the number of instructions received by the user. N sent This represents the total number of instructions issued by the power grid. B6: Adjust Interruption Rate I break The percentage of time interrupted during the regulation process is as follows: in, T interrupt To calculate the cumulative interruption duration, T total Adjust the total duration according to the plan; B7: Response latency Tdelay The average delay from the issuance of a power grid command to the initiation of a user load response is shown below: in, N The number of air conditioners participating in regulation, T i,start Let the corresponding time start for the i-th user. t cmd This refers to the time when the power grid issues the instruction; B8: Data Missing Rate L loss This indicates the proportion of missing data points during the data collection process, reflecting data integrity, as shown below: in, M loss For the number of missing households, M Total number of users; B9: User Comfort Score Indicates the degree to which room temperature deviates from the user's comfort range. in, CTO This refers to the cumulative degree to which the temperature deviates from the comfortable range within the control period; CTO max Δ represents the degree to which the temperature deviates from the comfortable range during the control period. T To regulate the temperature difference during the period, delta T max The maximum temperature difference during the regulation period, T t Let t be the room temperature at the t-th time point; T 0 Set the temperature to a comfortable level; Δ t To regulate the duration; B10: Load Reduction Amount LRA-E This indicates the total electricity load reduction of air conditioning groups during the control period. in, P baseline Let t be the baseline load power; P actual (t) represents the actual operating power at time t, Δ t To regulate the duration.

[0032] S2. Based on the expert scoring method, assign relative importance to each indicator in the air conditioning load control indicator system and construct a judgment matrix. Then, calculate the basic subjective weights based on the judgment matrix using the analytic hierarchy process, as follows: Based on the Saaty scaling method, multiple experts were invited to assign relative importance to each indicator at the same level and construct a judgment matrix for the indicators at the same level. Scale the 1-9 scale (Saaty scale method): 1 indicates that the two indicators are equally important; 3 indicates that the former is slightly more important than the latter; 5 indicates that the former is significantly more important than the latter; 7 indicates that the former is strongly more important than the latter; 9 indicates that the former is extremely more important than the latter; 2, 4, 6, and 8 are intermediate values; if the importance of indicator i relative to j is... ,but .

[0033] In this embodiment, effectiveness (A1) is slightly more important than reliability (A2), denoted as 2; effectiveness (A1) is significantly more important than economy (A3), denoted as 4; and reliability (A2) is slightly more important than economy (A3), denoted as 3.

[0034] This yields the judgment matrix for the primary indicators: For the secondary indicators B1-B10, multiple experts were invited to conduct pairwise comparisons to obtain the judgment matrix for the primary indicators: The geometric mean of each row of the judgment matrix corresponding to each level indicator is calculated based on the geometric mean method. After the geometric mean vector is normalized, the subjective weight of the first-level indicator and the local subjective weight of the second-level indicator are obtained. Calculate the geometric mean for each row of the judgment matrix of the primary indicators: Normalize the geometric mean vector to obtain the weight vector: i=1,2,…,n Substituting the values, we get: in, M i This represents the geometric mean of the i-th row of the matrix. W i This represents the subjective weight of the first-level indicator in the i-th row.

[0035] Similarly, the local subjective weights of the secondary indicators were calculated according to the above process, and the results are shown in Table 1.

[0036] Table 1. Local Subjective Weights of Secondary Indicators Converted to percentages: B1–B10 are 32.00%, 19.23%, 13.35%, 9.74%, 6.76%, 4.88%, 3.75%, 3.31%, 3.50%, and 3.48%, respectively.

[0037] Consistency checks were performed on the subjective weights of the primary indicators and the local subjective weights of the secondary indicators, as follows: The largest eigenvalue is calculated as follows: Obtain the largest eigenvalue of the first-level index matrix .

[0038] Calculate the consistency index CI and the consistency ratio CR. When CR < 0.1, the judgment matrix passes the consistency test, which means that the subjective weights of each index are credible. Wherein, RI is the average random consistency index; Similarly, the maximum eigenvalue of the secondary index is calculated using the above process. .

[0039] The judgment matrix shows good consistency, and the weights are acceptable.

[0040] The calculated subjective weights of the primary indicators are multiplied by the local subjective weights of the corresponding secondary indicators to obtain the basic subjective weights, as shown in Table 2.

[0041] Table 2 Subjective weights of indicators calculated using the analytic hierarchy process (AHP) Arrange the basic subjective weights in a certain order to form a subjective weight vector: .

[0042] S3. Collect multi-dimensional control data corresponding to indicators during the actual operation of the air conditioning load group and perform standardization processing. Based on the standardized multi-dimensional control data, calculate the information entropy value of each indicator using the entropy weight method, and determine the basic objective weights based on the entropy value. The collected data on load regulation rate, adjustable capacity percentage, response accuracy, communication reliability, and load reduction are standardized as follows: in, Let max( be the original value of the i-th data sample on the j-th indicator) x j ) represents the maximum original value of the j-th index among all samples, min( x j ) represents the minimum original value of the j-th index among all samples. The standardized value of the i-th sample data on the j-th indicator; The load fluctuation rate, regulation interruption rate, response delay time, data missing rate, and comfort deviation are standardized as follows: in, Let max( be the original value of the i-th data sample on the j-th indicator) x j ) represents the maximum original value of the j-th index among all samples, min( x j ) represents the minimum original value of the j-th index among all samples. The standardized value of the i-th sample data on the j-th indicator; Based on the standardized data, the objective weights of each indicator are calculated as follows: Calculate the weight of the i-th data sample on the j-th indicator. Where m represents the number of samples. This represents the weight of the i-th data sample on the j-th indicator; For index j, its entropy value e j Defined as: in: The entropy value is guaranteed to be in the interval [0, 1]; when all When they are completely equal, the entropy value This indicates that the indicator contributes no information; when The greater the difference, the smaller the entropy value, indicating that the indicator contributes more to the overall difference; Calculate the deviation g of i secondary indicators based on the entropy value. i : The j-th basic objective weight is defined as: in: is the objective weight of indicator j; n is the number of indicators.

[0043] In this embodiment, the calculated basic objective weights are arranged in a certain order to form an objective weight vector: .

[0044] S4. Construct a comprehensive weight optimization model based on the game theory combinatorial weighting method. Obtain a subjective weight vector by combining each basic subjective weight with each basic objective weight to obtain an objective weight vector. The objective is to minimize the sum of the comprehensive deviation between the comprehensive weight and the subjective weight vector and the comprehensive deviation between the comprehensive weight and the objective weight vector. Solve the comprehensive weight optimization model to obtain the optimal combination coefficients of the subjective and objective weights. Based on the optimal combination coefficients, the subjective weight vector, and the objective weight vector, obtain the comprehensive weights of each indicator, as detailed below: The comprehensive weight optimization model is constructed as follows: in, α 1 Subjective weighting coefficient; α 2 For objective weighting coefficients, For comprehensive weighting, Weights for subjective indicators; Weights for objective indicators; The objective function is defined as minimizing the overall deviation between the subjective weight, objective weight, and comprehensive weight of each indicator, as shown below: Will W Substituting the expression into the optimization objective function and obtaining the solution through mathematical transformation, we get the solution. α 1 and α 2 The system of linear equations is as follows: in, for The transpose of , for The transpose of ; Calculate matrix parameters: AHP autocorrelation(a): Entropy weight autocorrelation (c): Cross-correlation (b): Substitute into the system of equations: Calculate the coefficient determinant D: calculate Auxiliary determinant : calculate Auxiliary determinant : Solve using Cramer's rule : Based on the obtained optimal combination coefficients ,according to Calculate the overall weight: .

[0045] S5. Obtain the operational data of the air conditioning load control process to be evaluated and perform standardization processing (similar to S3) to obtain the standardized values ​​of each indicator. Then, perform a weighted summation of the standardized values ​​of each indicator and their corresponding comprehensive weights to calculate the comprehensive evaluation result of the air conditioning load control effect.

[0046] Example 2 This embodiment provides an air conditioning load control evaluation system based on an improved AHP-entropy weighting method, including: The air conditioning load control index system construction module is used to construct an air conditioning load control index system based on the control characteristics of air conditioning load groups. The subjective weight setting module is used to assign relative importance to each indicator in the air conditioning load control index system based on the expert scoring method and construct a judgment matrix. The basic subjective weights are calculated based on the judgment matrix using the analytic hierarchy process. The objective weight setting module is used to collect multi-dimensional control data corresponding to indicators during the actual operation of the air conditioning load group and perform standardized processing. Based on the standardized multi-dimensional control data, the information entropy value of each indicator is calculated by the entropy weight method, and the basic objective weights are determined based on the entropy value. The final comprehensive weight generation module is used to construct a comprehensive weight optimization model based on the game theory combination weighting method. It obtains a subjective weight vector through each basic subjective weight and an objective weight vector through each basic objective weight. With the goal of minimizing the comprehensive deviation between the comprehensive weight and the subjective weight vector plus the comprehensive deviation between the comprehensive weight and the objective weight vector, it solves the comprehensive weight optimization model to obtain the optimal combination coefficient of subjective and objective weights. Based on the optimal combination coefficient, the subjective weight vector, and the objective weight vector, it obtains the comprehensive weight of each indicator. The comprehensive evaluation result generation module is used to acquire the operation data of the air conditioning load control process to be evaluated and perform standardization processing to obtain the standardized values ​​of each indicator. The standardized values ​​of each indicator are then weighted and summed with their corresponding comprehensive weights to calculate the comprehensive evaluation result of the air conditioning load control effect.

[0047] Example 3 This embodiment provides a computer storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps of the air conditioning load control evaluation method based on the improved AHP-entropy weight combination weighting method as described in Embodiment 1.

[0048] Example 4 This embodiment 4 provides an electronic device, including a memory and a processor: the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the air conditioning load control evaluation method based on the improved AHP-entropy weight combination weighting method as described in embodiment 1.

[0049] The processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0050] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device.

[0051] Memory can be volatile memory, such as random-access memory (RAM); memory can also be non-volatile memory, such as read-only memory, flash memory, hard disk drive (HDD), or solid-state drive (SSD); or memory can be any other medium capable of carrying or storing desired program code in the form of instructions or data structures that can be accessed by a computer, but is not limited thereto. Memory can be a combination of the above-mentioned types of memory.

[0052] This invention can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented in whole or in part as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0053] It will be readily understood by those skilled in the art that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, combinations, substitutions, improvements, etc., made under the spirit and principles of the present invention are included within the protection scope of the present invention.

Claims

1. A method for evaluating air conditioning load control based on an improved AHP-entropy weight combination method, characterized in that, include: Based on the characteristics of air conditioning load group regulation, an air conditioning load regulation index system is constructed; Based on the expert scoring method, the relative importance of each indicator in the air conditioning load control index system is assigned to each level of the index system and a judgment matrix is ​​constructed. The basic subjective weights are calculated based on the judgment matrix using the analytic hierarchy process. Collect multi-dimensional control data corresponding to indicators during the actual operation of the air conditioning load group and perform standardization processing. Based on the standardized multi-dimensional control data, calculate the information entropy value of each indicator using the entropy weight method, and determine the basic objective weights based on the entropy value. A comprehensive weight optimization model is constructed based on the game theory combinatorial weighting method. Subjective weight vectors are obtained through each basic subjective weight, and objective weight vectors are obtained through each basic objective weight. The comprehensive weight optimization model is solved with the goal of minimizing the comprehensive deviation between the comprehensive weight and the subjective weight vector plus the comprehensive deviation between the comprehensive weight and the objective weight vector. The optimal combination coefficients of subjective and objective weights are obtained, and the comprehensive weights of each indicator are obtained based on the optimal combination coefficients, the subjective weight vector, and the objective weight vector. The operation data of the air conditioning load control process to be evaluated is obtained and standardized to obtain the standardized values ​​of each indicator. The standardized values ​​of each indicator are then weighted and summed with their corresponding comprehensive weights to calculate the comprehensive evaluation result of the air conditioning load control effect.

2. The air conditioning load control evaluation method based on the improved AHP-entropy weight combination weighting method according to claim 1, characterized in that, The air conditioning load control index system includes a primary index system, which includes effectiveness indicators, reliability indicators, and economic indicators.

3. The air conditioning load control evaluation method based on the improved AHP-entropy weight combination method according to claim 2, characterized in that, Each indicator in the primary indicator system corresponds to at least one secondary indicator; The secondary indicators corresponding to the effectiveness indicators include load regulation rate, adjustable capacity ratio, response accuracy, and load fluctuation rate; the secondary indicators corresponding to the reliability indicators include communication reliability, regulation interruption rate, response delay time, and data loss rate; and the secondary indicators corresponding to the economic indicators include user comfort deviation and load reduction amount.

4. The air conditioning load control evaluation method based on the improved AHP-entropy weight combination weighting method according to claim 3, characterized in that, The secondary indicators include: Load regulation rate R power The change in air conditioning load per unit time is shown below: in, P target For the target load, P initial For the initial load, T adjust The total time required to complete the composite adjustment; Adjustable capacity percentage C afj The ratio of the actual adjustable capacity to the rated capacity under current conditions is shown below: in, P i,max For a single user, the maximum power within the permissible temperature range. P i,min This is the minimum power required for a single user within the permissible temperature range. P rated This refers to the rated capacity of the air conditioner. Response progress δ error The following table shows the degree of agreement between the actual load adjustment results and the target value: in, P actual The actual regulating power of the air conditioning cluster. P target The target power required by the power grid; Load volatility σ p The power fluctuation amplitude during the regulation process reflects the stability of the execution, as shown below: in, P t Let be the total power of the air conditioning cluster at time t. T To detect the length of the time window, The average power during the control period; Communication reliability C success This indicates the ratio of power grid command issuance to user confirmation of receipt, as detailed below: in, N received Confirm the number of instructions received by the user. N sent This represents the total number of instructions issued by the power grid. Adjusting the interrupt rate I break The percentage of time interrupted during the regulation process is as follows: in, T interrupt To accumulate the interruption duration, T total Adjust the total duration according to the plan; Response latency T delay The average delay from the issuance of a power grid command to the initiation of a user load response is shown below: in, N The number of air conditioners participating in regulation, T i,start Let the corresponding time start for the i-th user. t cmd This refers to the time when the power grid issues the instruction; Data missing rate L loss The percentage of missing data points during the data collection process is shown below: in, M loss For the number of missing households, M Total number of users; User comfort Score This indicates the degree to which the room temperature deviates from the user's comfort range, as detailed below: in, CTO This refers to the cumulative degree to which the temperature deviates from the comfortable range within the control period; CTO max Δ represents the degree to which the temperature deviates from the comfortable range during the control period. T To regulate the temperature difference during the regulation period, ΔT max The maximum temperature difference during the regulation period, T t Let t be the room temperature at the t-th time point; T 0 The initial room temperature; Δ t To regulate the duration; Load reduction LRA-E This indicates the total electricity load reduction of the air conditioning group during the control period, as detailed below: in, P baseline Let t be the baseline load power; P real (t) represents the actual operating power at time t, Δ t To regulate the duration.

5. The air conditioning load control evaluation method based on the improved AHP-entropy weight combination method according to claim 2, characterized in that, Based on expert scoring, the relative importance of each indicator in the air conditioning load control indicator system is assigned at each level, and a judgment matrix is ​​constructed. The analytic hierarchy process (AHP) is then used to calculate the basic subjective weights based on the judgment matrix, including: Based on the Saaty scaling method, several experts were invited to assign relative importance to each indicator at the same level and construct a judgment matrix for the indicators at the same level. The geometric mean of each row of the judgment matrix corresponding to each level indicator is calculated based on the geometric mean method. After the geometric mean vector is normalized, the subjective weight of the first-level indicator and the local subjective weight of the second-level indicator are obtained. Perform consistency checks on each judgment matrix, as follows: Calculate the largest eigenvalue γ max As shown below: In the formula, n is the dimension of the judgment matrix, i.e., the number of indicators at the same level; A is the judgment matrix. w i Let be the subjective weight of the i-th indicator; Based on the largest eigenvalue γ max Calculate the consistency index CI and the consistency ratio CR. When CR < 0.1, the judgment matrix passes the consistency test, which means that the subjective weight of the first-level indicator or the local subjective weight of the second-level indicator corresponding to the judgment matrix is ​​reliable. The basic subjective weights are obtained by multiplying the subjective weights of the credible primary indicators by the local subjective weights of the corresponding secondary indicators.

6. The air conditioning load control evaluation method based on the improved AHP-entropy weight combination weighting method according to claim 3, characterized in that, The process involves collecting multi-dimensional control data corresponding to indicators from the actual operation of the air conditioning load group and standardizing it. Based on the standardized multi-dimensional control data, the information entropy value of each indicator is calculated using the entropy weight method, and the basic objective weights are determined based on the entropy values, including: The collected data samples corresponding to load regulation rate, adjustable capacity percentage, response accuracy, communication reliability, and load reduction amount are standardized as follows: The standardized value of the i-th data sample corresponding to the j-th indicator. y ij It equals the original data of the i-th sample of the indicator minus the minimum value among all the original data of the indicator, and then divided by the maximum value among all the original data of the indicator minus the minimum value among all the original data of the indicator. Standardization was applied to load fluctuation rate, regulation interruption rate, response delay time, data missing rate, and comfort deviation, as detailed below: The standardized value of the i-th data sample corresponding to the j-th indicator. y ij It equals the original data of the i-th sample of the indicator minus the maximum value among all the original data of the indicator, and then divided by the maximum value among all the original data of the indicator minus the minimum value among all the original data of the indicator. Based on standardized values, the objective weights of each basis are calculated as follows: Calculate the weight of the i-th data sample on the j-th secondary indicator: Where m represents the number of samples. This represents the weight of the i-th data sample on the j-th secondary indicator; For index j, its entropy value e j Defined as: in: Ensure that the entropy value is within the interval [0, 1]. Calculate the deviation g of i secondary indicators based on the entropy value. i : The j-th basic objective weight is defined as: in: is the j-th basic objective weight; n is the number of secondary indicators.

7. The air conditioning load control evaluation method based on the improved AHP-entropy weight combination weighting method according to claim 3, characterized in that, A comprehensive weight optimization model is constructed based on the game theory combinatorial weighting method. A subjective weight vector is obtained by combining each basic subjective weight with each basic objective weight to obtain an objective weight vector. The objective is to minimize the sum of the comprehensive deviation between the comprehensive weight and the subjective weight vector and the comprehensive deviation between the comprehensive weight and the objective weight vector. The comprehensive weight optimization model is solved to obtain the optimal combination coefficients of the subjective and objective weights. Based on these optimal combination coefficients, the subjective weight vector, and the objective weight vector, the comprehensive weights of each indicator are obtained, including: The comprehensive weight optimization model is constructed as follows: in, α 1 Subjective weighting coefficient; α 2 For objective weighting coefficients, For comprehensive weighting, For subjective weight vectors; For objective weight vectors; The objective function is defined as minimizing the sum of the combined deviation between the combined weights and the subjective weights vector, and the combined deviation between the combined weights and the objective weights vector, as shown below: Will W Substituting the expression into the objective function and obtaining the coefficients through mathematical transformation, we can solve for the coefficients. α 1 Sum of coefficients α 2 The system of linear equations is as follows: in, for The transpose of , for The transpose of ; Solve the system of linear equations to obtain the coefficients. α 1 Sum of coefficients α 2 For coefficients α 1 Sum of coefficients α 2 Normalization is performed to obtain the optimal combination coefficients. Based on the optimal combination coefficients, subjective weight vector, and objective weight vector, the comprehensive weights of each indicator are obtained.

8. An air conditioning load control evaluation system based on an improved AHP-entropy weight combination method, characterized in that, include: The air conditioning load control index system construction module is used to construct an air conditioning load control index system based on the control characteristics of air conditioning load groups. The subjective weight setting module is used to assign relative importance to each indicator in the air conditioning load control index system based on the expert scoring method and construct a judgment matrix. The basic subjective weights are calculated based on the judgment matrix using the analytic hierarchy process. The objective weight setting module is used to collect multi-dimensional control data corresponding to indicators during the actual operation of the air conditioning load group and perform standardized processing. Based on the standardized multi-dimensional control data, the information entropy value of each indicator is calculated by the entropy weight method, and the basic objective weights are determined based on the entropy value. The final comprehensive weight generation module is used to construct a comprehensive weight optimization model based on the game theory combination weighting method. It obtains a subjective weight vector through each basic subjective weight and an objective weight vector through each basic objective weight. With the goal of minimizing the comprehensive deviation between the comprehensive weight and the subjective weight vector plus the comprehensive deviation between the comprehensive weight and the objective weight vector, it solves the comprehensive weight optimization model to obtain the optimal combination coefficient of subjective and objective weights. Based on the optimal combination coefficient, the subjective weight vector, and the objective weight vector, it obtains the comprehensive weight of each indicator. The comprehensive evaluation result generation module is used to acquire the operation data of the air conditioning load control process to be evaluated and perform standardization processing to obtain the standardized values ​​of each indicator. The standardized values ​​of each indicator are then weighted and summed with their corresponding comprehensive weights to calculate the comprehensive evaluation result of the air conditioning load control effect.

9. A computer storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the air conditioning load control evaluation method based on the improved AHP-entropy weight combination weighting method as described in any one of claims 1-7.

10. An electronic device, comprising a memory and a processor: the memory for storing computer-executable instructions, the processor for executing the computer-executable instructions, wherein the computer-executable instructions, when executed by the processor, implement the steps of the air conditioning load control evaluation method based on the improved AHP-entropy weight combination weighting method as described in any one of claims 1-7.

Images