HAC + CRITIC-based demand response feature clustering compensation mechanism design method

The clustering and subsidy coefficient calculation of load capacity characteristics through the HAC+CRITIC method is solved, and the problem of insufficient load clustering in the existing technology is achieved, achieving more efficient load response and operating cost optimization.

CN119093386APending Publication Date: 2024-12-06GUANGXI POWER GRID CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411083526.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-08
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing load clustering method fails to effectively utilize capacity data, resulting in the compensation mechanism being not refined enough to effectively promote load participation demand response.

Method used

The HAC+CRITIC method is used to quantify load capacity characteristic indicators such as capacity size, response time and response rate, load clustering is carried out, and subsidy coefficients of different categories are calculated to achieve a more refined compensation mechanism.

Benefits of technology

It improves the precision and efficiency of load participation demand response, reduces operating costs, and increases load response capacity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119093386A_ABST
    Figure CN119093386A_ABST
Patent Text Reader

Abstract

The invention relates to the field of demand response compensation mechanisms, and provides a demand response feature clustering compensation mechanism design method based on HAC + CRITIC. The method comprises the following steps: firstly, quantizing characteristic indexes representing load capacity, extracting characteristic vectors, and clustering loads through an HAC algorithm; secondly, using a CRITIC method to calculate a feature index weight, combining the feature index weight with the feature index, and designing a compensation mechanism in a classified manner by referring to a domestic province and city demand response compensation mechanism; and finally, through comparison with other provincial and municipal compensation mechanisms in China, it is verified that the compensation mechanism provided by the invention can reduce the operation cost and improve the load response capacity. The result shows that the provided capacity characteristic quantitative index can well represent the load characteristics of the capacity data, and the HAC algorithm aggregates and classifies according to the load capacity characteristics, thereby laying a foundation for the classification design of a compensation mechanism. The compensation mechanism is higher in universality, and compared with a control group, the operation cost is reduced, the actual load response capacity is also improved, and users are promoted to actively participate in demand response.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of demand response technology, and in particular to a demand response feature clustering compensation mechanism design method based on HAC+CRITIC. Background Art

[0002] With the development of new power systems, the proportion of new energy sources has increased, and the contradiction between power supply and demand has become increasingly prominent. In order to ensure the stable operation of the power system and the reliable consumption of new energy sources, a large number of flexible load-side resources are needed to participate in demand response. The number of loads in the power system is large and the types are complex. Therefore, before studying the loads, it is necessary to consider clustering many loads. For load clustering, it is necessary to extract feature vectors as the basis for clustering. Many current methods for load feature extraction and clustering do not target capacity data. Load data needs to be processed in a certain way to obtain capacity data. Directly processing capacity data and extracting features can avoid errors in the process of processing load data, and design compensation mechanisms based on load capacity classification. At this stage, various compensation mechanisms have promoted the participation of loads in demand response to a certain extent. However, most compensation mechanisms are relatively general and not detailed enough.

[0003] In view of this, a demand response feature clustering compensation mechanism design method based on HAC+CRITIC is needed. Summary of the invention

[0004] In view of the problems existing in the prior art, the present invention provides a design method for a demand response characteristic clustering compensation mechanism based on HAC+CRITIC; the capacity size, response time, and response rate are used as indicators to quantify the load capacity characteristics, and the HAC algorithm is used to cluster the loads and classify and design the subsidy coefficient. The specific technical solution is as follows:

[0005] A design method for demand response characteristic clustering compensation mechanism based on HAC+CRITIC, including:

[0006] Obtain load capacity data;

[0007] Quantifying load capacity characteristic indicators, the capacity characteristic indicators include: capacity size, response duration and response rate;

[0008] Extracting load capacity feature vector;

[0009] The HAC algorithm is used to cluster the loads, the CRITIC method is used to calculate the objective weights of the capacity characteristic indicators, and the capacity characteristic indicators are combined with the corresponding objective weights to calculate the subsidy coefficients of different categories;

[0010] The compensation mechanism can be implemented based on the ratio of actual load response capacity to declared capacity and the subsidy coefficient.

[0011] Preferably, the capacity size is the average value of all capacities; the response time is the ratio of valid response data points to all data points; and the response rate is the average value of the rate of change between every two data points.

[0012] Preferably, in clustering the loads using the HAC algorithm, a single-link algorithm is used to cluster the loads.

[0013] Preferably, the loads are divided into three categories after clustering: the load capacity is the largest, the response time is the longest, and the response rate is relatively high; the load capacity and the response time are both small, and the response rate is the largest; and the load capacity is large, the response rate is the largest, and the response time is the smallest.

[0014] Preferably, the objective weight is comprehensively weighed by comparing strength and conflict.

[0015] Preferably, the comparison strength index is expressed in the form of standard deviation. The larger the standard deviation of the data, the greater the fluctuation and the higher the weight.

[0016] The conflict index is expressed in the form of a correlation coefficient. The larger the correlation coefficient between the indicators, the smaller the conflict and the lower the weight.

[0017] Preferably, the subsidy coefficient calculation formula is as follows:

[0018]

[0019] Where ξ represents the load subsidy coefficient, μ′, θ′ and k′ are the standard values ​​of characteristic indicators that characterize the load capacity, response time and response rate, W1, W2 and W3 represent the weights of the three indicators, N is the number of loads of the corresponding type, and σ is the balance coefficient.

[0020] A computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the above methods.

[0021] A processor, the processor being used to run a program, wherein the program executes the method according to any one of claims 1 to 7 when running.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] The present invention uses capacity size, response time and response rate as indicators to quantify load capacity characteristics, and uses the HAC algorithm to cluster the load demand response characteristics to cluster the compensation mechanism, and classifies and designs different subsidy coefficients, so that the response compensation mechanism design is more refined. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0025] Figure 1 This is the capacity feature extraction and clustering flow chart proposed in the present invention.

[0026] Figure 2 This is a diagram of the demand response compensation process proposed by the present invention.

[0027] Figure 3 The load capacity curves of different categories obtained using the example data are shown in Figure 2.

[0028] Figure 4 This is the clustering result diagram obtained by extracting the feature vector of load capacity data. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0031] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0032] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0033] Embodiment 1:

[0034] See also Figure 1 , oneA demand response feature clustering compensation mechanism design method based on HAC+CRITIC, including:

[0035] S1. Obtain load capacity data;

[0036] S2, quantitative load capacity characteristic indicators, the capacity characteristic indicators include: capacity size, response duration and response rate;

[0037] The capacity is defined as the average value of all capacities, and its expression formula is:

[0038]

[0039] Response time definition: the ratio of effective response data points to all data points, and its expression formula is:

[0040]

[0041] Response rate definition: the average value of the rate of change between every two data points, expressed as:

[0042]

[0043] In the formula, x i 、x i+1 Represent the capacity values ​​of the i-th data point and the next data point respectively, and n represents the number of data points; the effective response data point represents the data point whose capacity value is greater than α times the average capacity value, that is, when the load capacity is greater than α times the average capacity value, it is considered to be an effective response, x represents the capacity value of the data point, N (x ≥ μ * α) represents the number of data points that meet the effective response, N a Represents the number of all data points, and α in the present invention is 0.65.

[0044] Extract the load capacity characteristic vector; specifically: obtain the capacity characteristic vector Where j represents the load number.

[0045] S3. Use the HAC algorithm to cluster the loads, use the CRITIC method to calculate the objective weights of the capacity characteristic indicators, and combine the capacity characteristic indicators with the corresponding objective weights to calculate the subsidy coefficients of different categories. Specifically include:

[0046] (1) Calculate the capacity characteristic index weight using the CRITIC method

[0047] The HAC algorithm calculates the similarity between classes through some similarity measure, sorts them from high to low by similarity, and gradually reconnects the classes. The agglomerative hierarchical clustering method uses a bottom-up strategy. At the beginning, each object is an independent class, and then merges into larger and larger classes until all objects are in one class or a certain termination condition is met.

[0048] According to different methods for calculating the distance between classes, the hierarchical clustering algorithm can also be divided into single-linkage algorithm, complete-linkage algorithm and average-linkage algorithm. The clustering of this embodiment adopts the single-linkage algorithm, which adopts the minimum distance, and the calculation formula is as follows:

[0049]

[0050] Where: represents the Euclidean distance between the feature vectors λ and λ', and C and C' are the sets to which λ and λ' belong.

[0051] In summary, by processing the original capacity data, quantifying the load capacity characteristic indicators, extracting the characteristic vectors representing the capacity size, load response duration and load response rate, calculating the Euclidean distance between the characteristic vectors of the loads, and clustering the loads with similar distances, this paper divides the clustered loads into three categories: Class I loads have the largest capacity, the longest response time, and a larger response rate; Class II loads have smaller capacity and response time, and the largest response rate; Class III loads have a larger capacity, the largest response rate, and the smallest response time.

[0052] In addition, the CRITIC method is used to calculate the capacity characteristic index weights, as follows:

[0053] The CRITIC method is an objective weighting method proposed by Diakoulaki. The idea is to use two indicators, namely the contrast intensity and the conflict index. Assuming there are m objects to be evaluated and n indicators, the data matrix is ​​m*n order.

[0054] Among them, the contrast strength index contrast strength is expressed in the form of standard deviation. The larger the data standard deviation, the greater the fluctuation and the higher the weight. Its expression is as follows:

[0055]

[0056] In the formula, x pq represents the qth index value of the pth evaluation object, represents the average value of the elements in the qth column, S q Represents the standard deviation of the qth indicator, p=1,…,m,q=1,2,3.

[0057] The conflict index is expressed in the form of correlation coefficient. The larger the correlation coefficient between the indicators, the smaller the conflict and the lower the weight. The expression is as follows:

[0058]

[0059] In the formula, r qq' Represents the correlation coefficient between the evaluation indicators q and q'. q It represents the quantitative index of the conflict between the qth indicator and other indicators.

[0060] Furthermore, the objective weight of each indicator is comprehensively measured by comparison strength and conflict. The greater the amount of information contained in the evaluation indicator, the greater the relative importance of the indicator, and the greater the objective weight. The specific expression is as follows:

[0061]

[0062] In the formula, C q represents the amount of information contained in the qth evaluation index, W q Represents the objective weight of the qth indicator.

[0063] (2) Design of demand response compensation mechanism

[0064] It should be noted that power demand response refers to the implementation of power demand response by power users in the power market, which changes their power consumption behavior according to price signals or incentive mechanisms, uses price levers to guide users to actively participate in demand response, and taps into the load regulation capacity on the demand side. It plays an important role in ensuring the balance of power supply and demand, promoting the friendly interaction of source-grid-load-storage and the consumption of renewable energy. Each entity involved in demand response must attach great importance to the importance of demand response, ensure that the response is implemented in place, and ensure the safe and stable operation of the power grid. At the same time, cities are encouraged to broaden funding sources and increase demand response compensation through various means.

[0065] The demand response compensation process mainly includes user declaration, response invitation, response execution and subsidy settlement. Figure 2 As shown:

[0066] Among the above-mentioned load capacity characteristic indicators, the response time indicator has been in a normalized form, and for the capacity size and response rate indicators, the two indicators need to be normalized first. This embodiment adopts minimum-maximum normalization.

[0067] Combining the above weights, the capacity characteristic index and the corresponding weights can be combined to obtain the subsidy coefficient. The subsidy coefficient calculation formula is as follows:

[0068]

[0069] Wherein, ξ represents the subsidy coefficient of the load, μ′, θ′ and k′ are the standard values ​​of characteristic indicators representing the load capacity, response time and response rate, W1, W2 and W3 represent the weights of the three indicators, N is the number of loads of the corresponding type, σ is the balance coefficient, and σ is taken as 0.2 in this embodiment.

[0070] The compensation mechanism can be implemented based on the proportion of actual load response capacity to declared capacity and the subsidy coefficient, and its effectiveness can be verified.

[0071] It should also be noted that after the design of the demand response compensation mechanism is completed, it is necessary to select evaluation indicators to verify the effectiveness of the compensation mechanism. In order to better verify the effectiveness of the compensation mechanism proposed in the present invention, this embodiment takes the power grid as the main body and verifies the effectiveness of the mechanism by measuring the operating cost of the power grid. In the process of calculating the operating cost, after considering the load participating in the demand response, the power grid needs to provide economic subsidies to the load that effectively participates in the demand response. On the contrary, if the load fails to participate in the demand response according to the declared capacity, it will also be subject to certain economic penalties; at the same time, considering that users participate in the power auxiliary service market, they will also receive a part of the subsidy.

[0072] In summary, the grid operation cost formula is as follows:

[0073]

[0074] 0≤X i ≤P

[0075] Where: R represents the grid operation cost; P represents the demand response capacity; X i represents the actual response capacity of the i-th load; ρ c represents the compensation benchmark price for loads participating in demand response; ξ(X i / P) indicates that the actual load response capacity accounts for the proportion of the declared capacity. i Subsidy coefficient when / P; ρ a and ρ p They respectively represent the electricity sales price for the load participating in ancillary services and the penalty price for the load not fully meeting the declared capacity.

[0076] In order to implement the demand response classification compensation mechanism, the data from Guangxi, China were used for verification. The load data involved four types of loads: industrial, agricultural, commercial, and residential loads. The total number of loads was 22, all of which were weekly content data. The representative capacity curve of the load was selected, such as Figure 3 As shown:

[0077] One point is taken every 15 minutes, that is, each load has 672 data points. After extracting and clustering the characteristic vector of load capacity data, the clustering results are as follows: Figure 4 As shown:

[0078] Finally, the loads are divided into three categories: Figure 4In the figure, the orange dot represents Class I load, the green dot represents Class II load, and the red dot represents Class III load. Class I load has the largest capacity, the longest response time, and the largest response rate; Class II load has a smaller capacity and response time, and the largest response rate; Class III load has a larger capacity, the largest response rate, and the smallest response time.

[0079] The subsidy coefficient of the load is calculated by the formula for calculating the subsidy coefficient in the above content. Since there is only one Class II load, its corresponding subsidy coefficient cannot be calculated, and the Class II load characteristics are at a medium level. Therefore, the average of the subsidy coefficients of Class I and Class III loads is considered as the subsidy coefficient of Class II load, and the subsidy coefficients corresponding to the three types of loads are 0.49, 0.415 and 0.34 respectively.

[0080] At the same time, combined with the current power market demand response compensation mechanism in various provinces and cities in China, according to the proportion of actual response capacity to declared capacity, this paper selects the proportion with higher universality, namely <60%, 60%-80% and >80%. The load response capacity ratio and the corresponding compensation coefficient are shown in Table 1:

[0081] Table 1 Compensation coefficient design

[0082] Response capacity ratio <60% 60%-80% >80% Class I load 0 0.49 1 Class II load 0 0.415 1 Class III load 0 0.34 1

[0083] Combined with the "Implementation Plan for Market-oriented Demand Response of Guangxi Electricity (Trial)", the subsidy base price for the three types of loads participating in demand response is set at the upper limit of the response price of 2.5 yuan / kWh, the response time is set at the lower limit of the response time of 1 hour, the penalty price is set at 3 yuan / kW, and the subsidy price for participating in ancillary services is set at 4.5 yuan / kW. At the same time, when calculating the classified compensation mechanism proposed in this paper, a step-by-step compensation method is adopted. The grid-side operating cost and the actual capacity of the load participating in demand response are selected as evaluation indicators.

[0084] In order to verify the effectiveness of the compensation mechanism proposed in this paper, two control groups are set up, namely the compensation mechanisms of Fujian Province and Ningxia Hui Autonomous Region. The subsidy mechanisms of the two provinces and cities are shown in Table 2:

[0085] Table 2 Subsidy mechanisms in two provinces and cities

[0086]

[0087] Combined with the grid operation cost formula, the operation cost and actual load response capacity are calculated according to the response capacity ratio and subsidy coefficient of the two provinces and cities. Under the three subsidy mechanisms, the corresponding operation costs and actual load response capacity are shown in Table 3:

[0088] Table 3 Comparison of operating costs under three compensation mechanisms

[0089]

[0090] As shown in Table 3, compared with the compensation mechanisms of the control groups of Fujian Province and Ningxia Hui Autonomous Region, the classified subsidy mechanism proposed in this paper reduced the total operating costs by 1081.707 yuan and 567.457 yuan respectively, and the load response capacity increased by 1025.7kW and 684.9kW respectively. For each type of load, the operating cost has been reduced to a certain extent, and at the same time, the load response capacity has also increased to a certain extent. Therefore, it can be proved that the classified compensation mechanism proposed in this paper is more economical and improves the load participation in demand response capacity.

[0091] Example 2

[0092] A computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method described.

[0093] Example 3

[0094] A processor is used to run a program, wherein the method described is executed when the program is run.

[0095] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0096] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0097] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0098] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A design method for demand response feature clustering compensation mechanism based on HAC+CRITIC, characterized in that: include: Obtain load capacity data; Quantifying load capacity characteristic indicators, the capacity characteristic indicators including: capacity size, response duration and response rate; extracting load capacity characteristic vectors; The HAC algorithm is used to cluster the loads, the CRITIC method is used to calculate the objective weights of the capacity characteristic indicators, and the capacity characteristic indicators are combined with the corresponding objective weights to calculate the subsidy coefficients of different categories; The compensation mechanism can be implemented based on the ratio of actual load response capacity to declared capacity and the subsidy coefficient.

2. According to claim 1, a method for designing a demand response feature clustering compensation mechanism based on HAC+CRITIC, characterized in that: The capacity size is the average value of all capacities; the response time is the ratio of valid response data points to all data points; and the response rate is the average value of the rate of change between every two data points.

3. According to claim 1, a method for designing a demand response feature clustering compensation mechanism based on HAC+CRITIC, characterized in that: In the process of clustering the loads using the HAC algorithm, a single connection algorithm is used to cluster the loads.

4. The method for designing a demand response feature clustering compensation mechanism based on HAC+CRITIC according to claim 3 is characterized in that: The loads are clustered and divided into three categories: the load capacity is the largest, the response time is the longest, and the response rate is relatively high; the load capacity and the response time are both small, and the response rate is the highest; and the load capacity is large, the response rate is the highest, and the response time is the lowest.

5. According to claim 1, a method for designing a demand response feature clustering compensation mechanism based on HAC+CRITIC, characterized in that: The objective weight is comprehensively measured by comparison strength and conflict.

6. The method for designing a demand response feature clustering compensation mechanism based on HAC+CRITIC according to claim 5 is characterized in that: The comparison strength index is expressed in the form of standard deviation. The larger the standard deviation of the data, the greater the fluctuation and the higher the weight. The conflict index is expressed in the form of a correlation coefficient. The larger the correlation coefficient between the indicators, the smaller the conflict and the lower the weight.

7. The method for designing a demand response feature clustering compensation mechanism based on HAC+CRITIC according to claim 1 is characterized in that: The subsidy coefficient calculation formula is as follows: Where ξ represents the load subsidy coefficient, μ′, θ′ and k′ are the standard values ​​of characteristic indicators that characterize the load capacity, response time and response rate, W1, W2 and W3 represent the weights of the three indicators, N is the number of loads of the corresponding type, and σ is the balance coefficient.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

9. A processor, characterized in that: The processor is used to run a program, wherein the program executes the method according to any one of claims 1 to 7 when running.