Power consumer feature classification method and system based on fuzzy C-means clustering algorithm

Abstract

The invention discloses a power consumer feature classification method and system based on a fuzzy C-means clustering algorithm, and relates to the technical field of power consumer feature classification in the whole industry. Comprising the steps of collecting user data of each industry, and constructing a power consumer power utilization behavior index system; calculating the power consumption behavior index weight of the power consumer; carrying out normalization and dimensionless processing on the indexes, and carrying out comprehensive evaluation; and establishing a power consumer feature portrait based on a fuzzy C-means clustering algorithm. According to the method, the scientificity and systematicness of user classification are ensured by establishing the power utilization behavior index system of the power users, and the accuracy of user portraits is improved; according to the method, data are preprocessed and normalized based on power consumption behavior indexes, comprehensive evaluation and analysis are carried out, power consumers are accurately classified into different power consumption characteristic types, and the classification accuracy is improved; the user portrait evaluation system is established through the power consumption characteristic parameters, the power user characteristic classification method is updated and optimized regularly, and the classification efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power user feature classification in the whole industry, and in particular to a power user feature classification method and system based on a fuzzy C-means clustering algorithm. Background Art

[0002] As my country's "dual carbon" goals are advanced, the power industry faces increasingly complex challenges, especially in the refinement of power user management and services. Accurately identifying and classifying the characteristics of power users is the basis for achieving coordinated development of energy, economy and environment. However, with the rapid development of my country's economy and society, there are more and more types of power-consuming equipment, and users' power consumption behavior patterns are becoming more and more diverse. In addition, emerging factors such as environmental and climate change, distributed power sources, integrated energy systems and the opening of the power sales market have introduced uncertainty to traditional users' power consumption behavior. Users' power consumption behavior patterns are seasonal, regional, industry-related, policy-related, random and adjustable. Under the current technical environment, the characteristic classification of power users can not only optimize the allocation of power resources, but also provide a basis for scientific decision-making for the government and power grid companies.

[0003] At present, the research on the classification of power user characteristics is still in its initial stage. Existing classification models mostly focus on indicators such as energy consumption and electricity usage time, and are obviously insufficient in analyzing user behavior in the market environment. This has led to the failure to fully tap the potential of power users in power dispatching, demand-side management, and energy-saving and emission reduction measures. Summary of the invention

[0004] In view of the problems existing in the prior art, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to construct a comprehensive indicator system for power users' electricity consumption behavior, calculate the weights of these indicators, normalize and dimensionlessly process the indicators, and then use the entropy weight method and the approximate ideal solution sorting method to conduct a comprehensive evaluation and create labels for users' electricity consumption behavior.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a method for classifying power user characteristics based on a fuzzy C-means clustering algorithm, which includes collecting user data from various industries and constructing a power user power consumption behavior indicator system;

[0008] Calculate the weights of electricity users' electricity consumption behavior indicators;

[0009] Normalize and dimensionless the indicators and make a comprehensive evaluation;

[0010] Based on the fuzzy C-means clustering algorithm, a characteristic profile of power users is established.

[0011] As a preferred solution of the power user characteristic classification method based on fuzzy C-means clustering algorithm described in the present invention, the power user power consumption behavior indicator system includes:

[0012] The electricity price sensitivity index refers to the load transfer rate, which is used to characterize the ratio of user load transfer;

[0013] The temperature sensitivity index refers to the relationship between the load curve change and the temperature curve change;

[0014] The electricity consumption stability index is used to measure the degree of load fluctuation of the power system within a certain period of time.

[0015] As a preferred solution of the method for classifying power user characteristics based on fuzzy C-means clustering algorithm described in the present invention, the electricity price sensitivity index includes:

[0016] Peak-valley load transfer rate refers to the ratio of the difference between the load during the peak period and the load during the valley period to the average power consumption during the peak period;

[0017] The off-peak load transfer rate refers to the proportion of users transferring part of their load from the normal period to the peak period through the electricity price incentive mechanism;

[0018] The off-peak load transfer rate refers to the proportion of users transferring part of their load from normal periods to off-peak periods through the electricity price incentive mechanism.

[0019] As a preferred solution of the power user characteristic classification method based on fuzzy C-means clustering algorithm described in the present invention, the power consumption stability index includes:

[0020] The peak-to-valley difference rate on the maximum load day refers to the percentage of the maximum daily peak-to-valley difference of user load to the maximum load;

[0021] The load fluctuation rate on the maximum load day refers to the ratio of the standard deviation of the load curve on the day with the maximum user load to the average value;

[0022] Weekend load rate refers to the ratio of the average load of users on rest days to the average load on weekdays;

[0023] Weekly average load fluctuation rate refers to the average ratio of the standard deviation of the user's weekly load curve to the average value;

[0024] Monthly average daily load rate refers to the average value of daily load rates within a month;

[0025] The monthly load rate refers to the ratio of the average daily electricity consumption to the maximum daily electricity consumption in a month.

[0026] As a preferred solution of the power user characteristic classification method based on fuzzy C-means clustering algorithm described in the present invention, the calculation of the index weight includes:

[0027] Construct an electricity consumption behavior evaluation matrix consisting of m evaluation schemes and n indicators;

[0028] The initial calculation results of each indicator are normalized to unit values ​​to eliminate the dimensional differences between indicators;

[0029] The entropy value of each evaluation index is calculated, and the weight of each index is calculated based on the entropy value.

[0030] As a preferred solution of the method for classifying power user characteristics based on fuzzy C-means clustering algorithm described in the present invention, the comprehensive evaluation adopts TOPSIS method, which specifically includes:

[0031] Constructing a standardized decision matrix:

[0032]

[0033] Among them, r ij represents the attribute j of the i-th option after normalization;

[0034] Compute the weighted normalized decision matrix:

[0035] v ij =w i r ij ,i=1,...,n

[0036] Among them, w i is the weight of attribute j, r ij represents the normalized attribute j, v of the ith option ij is the weighted standardized decision matrix;

[0037] Determine the ideal solution A + and negative ideal solution A - , where the positive characteristic index Take the maximum value, negative characteristic index Take the minimum value;

[0038] Calculate solution i to ideal solution A + Distance:

[0039]

[0040] Calculate solution i to negative ideal solution A - Distance:

[0041]

[0042] in, represents the distance from solution i to the ideal solution, represents the distance from solution i to the negative ideal solution;

[0043] Compute the solution that is closest to the ideal solution:

[0044]

[0045] Among them, C i Represents the distance from the solution to the ideal solution.

[0046] As a preferred solution of the method for classifying power user characteristics based on fuzzy C-means clustering algorithm described in the present invention, the temperature sensitivity index is calculated using grey correlation degree:

[0047] Δmin=min i [min k (|x0(k)-x i (k)|)]

[0048] Δmax=max i [max k (|x0(k)-x i (k)|)]

[0049]

[0050] Among them, x0(k) is the highest daily temperature sequence in a certain area, x i (k) is the maximum daily load sequence of each user in the same period, ξ(k) is the sum of x0(k) and x i The correlation coefficient of the kth element in (k);

[0051] The correlation between the ith sequence and the main sequence is:

[0052]

[0053] Among them, γ i is the correlation between the ith sequence and the main sequence, ξ i (k) is the correlation coefficient of the kth element in the ith sequence.

[0054] In a second aspect, an embodiment of the present invention provides a power user feature classification system based on a fuzzy C-means clustering algorithm, which includes a power data acquisition module, a data processing and evaluation module, and a portrait optimization module;

[0055] The power data collection module is used to collect user electricity price sensitivity data, temperature sensitivity data, power consumption stability data and environmental impact data, and establish a power user power consumption behavior indicator system;

[0056] The data processing and evaluation module is used to pre-process, normalize and evaluate the data based on the electricity consumption behavior indicator system;

[0057] The portrait optimization module is used to monitor the characteristic portraits of power users, establish an evaluation system through power consumption characteristic parameters, and regularly update and optimize the power user characteristic classification method.

[0058] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the method for classifying power user characteristics based on the fuzzy C-means clustering algorithm as described in the first aspect of the present invention are implemented.

[0059] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the method for classifying characteristics of electricity users based on the fuzzy C-means clustering algorithm as described in the first aspect of the present invention are implemented.

[0060] The beneficial effects of the present invention are as follows: the industry-wide power user characteristic classification method based on the fuzzy C-means clustering algorithm provided by the present invention ensures the scientificity and systematicity of user classification by establishing a power user power consumption behavior indicator system, improves the accuracy of user portraits, reduces classification deviations, and improves the efficiency and reliability of power resource allocation; by preprocessing and normalizing data based on power consumption behavior indicators, and conducting comprehensive evaluation and analysis, power users are accurately classified into different power consumption characteristic types, and corresponding evaluation methods are adopted, which improves the accuracy of classification, reduces management costs and resource waste, and ensures the economy and sustainability of power system operation; a user portrait evaluation system is established through power consumption characteristic parameters, and the power user characteristic classification method is regularly updated and optimized, which improves classification efficiency, optimizes user service quality, and ensures the modernization and efficiency of the classification method. The present invention achieves better results in classification accuracy, operating costs, and service efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0062] Figure 1 It is a flow chart of the classification of power user characteristics in all industries based on the power user characteristics classification method of the fuzzy C-means clustering algorithm;

[0063] Figure 2It is a flow chart of fuzzy clustering analysis algorithm for power user characteristic classification method based on fuzzy C-means clustering algorithm; DETAILED DESCRIPTION

[0064] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0065] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0066] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0067] Example 1

[0068] Reference Figure 1-2 , which is the first embodiment of the present invention, and provides a method for classifying power user characteristics based on a fuzzy C-means clustering algorithm, comprising:

[0069] S1, collect user data from various industries and build an indicator system for power consumption behavior of power users;

[0070] It should be noted that user data is divided into two categories: real-time collection data and historical operation data. The real-time collection data includes the temperature sensitivity index, electricity price sensitivity index and electricity consumption stability index of each user, which are used to monitor the real-time electricity consumption characteristics of the user; the historical operation data includes the maximum value records of various indicators and historical electricity consumption records, which are used to evaluate the overall electricity consumption characteristics of the user. The system uses standard data collection protocols to collect these data regularly.

[0071] It should be noted that the system uses a standard collection protocol to collect data every 30 minutes. Each user is equipped with an electricity data collection device, which can record different types of electricity consumption index data in real time. For historical data, the system will save nearly 60 days of data for analysis and evaluation. The collected data is processed by the preprocessing module for anomaly detection and missing value processing to ensure the integrity and accuracy of the data.

[0072] S2, calculate the weight of the power consumption behavior index of power users;

[0073] It should be noted that the indicator weight calculation adopts an evaluation method based on the entropy weight method, which mainly includes the construction and normalization of the electricity consumption behavior evaluation matrix. By recording the change values ​​of various indicators within a fixed time interval, it lays the foundation for subsequent weight calculation. The entropy weight method can effectively reflect the information content of the indicator.

[0074] It should be noted that the system selects 7 days as a complete evaluation cycle and processes the data within this cycle as a data set. When calculating weights, the system will automatically update the evaluation values ​​of various indicators to ensure the timeliness of weight calculations. For possible data fluctuations, the system uses a smoothing method to improve calculation stability.

[0075] S3, normalize and dimensionless the indicators and make a comprehensive evaluation;

[0076] It should be noted that the comprehensive evaluation adopts a multi-attribute evaluation method based on TOPSIS. This method determines the evaluation results by analyzing the normalized data matrix, and evaluates the rationality of user classification through distance calculation and relative closeness. At the same time, the ideal solution and negative ideal solution calculation are introduced to evaluate the classification effect.

[0077] It should be noted that the evaluation matrix is ​​calculated using a standardized method, and the weights of each parameter are determined by analyzing the distribution characteristics of the indicator data. The system will regularly calibrate the evaluation method based on the deviation between the evaluation results and the actual classification to improve the classification accuracy. The comprehensive evaluation takes into account the characteristics of various indicators and the electricity consumption characteristics of users, providing a basis for dynamic classification.

[0078] S4, based on the fuzzy C-means clustering algorithm, establish the characteristic profile of power users;

[0079] It should be noted that the feature portrait is established by using a hierarchical clustering strategy, which has strong adaptability. This method can automatically select classification criteria based on the clustering analysis results and classify users by features. When the classification effect does not meet the requirements, it will be optimized and adjusted through an iterative mechanism.

[0080] Example 2

[0081] Reference Figure 1-2 , which is the second embodiment of the present invention, and provides a method for classifying power user characteristics based on a fuzzy C-means clustering algorithm, comprising:

[0082] S1, collect user data from various industries and build an indicator system for power consumption behavior of power users;

[0083] In the embodiment of the present application, the power user's power consumption behavior indicator system includes:

[0084] The electricity price sensitivity index refers to the load transfer rate, which is used to characterize the ratio of user load transfer;

[0085] The temperature sensitivity index refers to the relationship between the load curve change and the temperature curve change;

[0086] The electricity consumption stability index is used to measure the degree of load fluctuation of the power system within a certain period of time.

[0087] It should be noted that the electricity price sensitivity index refers to the load transfer rate, which is defined as the ratio of user load transfer amounts, including: peak-valley load transfer rate, off-peak load transfer rate and off-valley load transfer rate.

[0088] Furthermore, the electricity price sensitivity index includes:

[0089] Peak-valley load transfer rate refers to the ratio of the difference between the load during the peak period and the load during the valley period to the average power consumption during the peak period;

[0090] The off-peak load transfer rate, between the normal period and the peak period, refers to the proportion of users transferring part of the load in the normal period to the peak period through the electricity price incentive mechanism;

[0091] The flat-to-valley load transfer rate, between the flat period and the valley period, refers to the proportion of users transferring part of their flat-period load to the valley period through the electricity price incentive mechanism.

[0092] It should be noted that the power consumption stability index measures the degree of load fluctuation of the power system within a certain period of time. It is a set of parameters of a series of load fluctuations, which mainly include:

[0093] Peak-to-valley difference rate on the maximum load day: refers to the percentage of the maximum daily peak-to-valley difference of user load to the maximum load, which is a negative characteristic indicator;

[0094] Load fluctuation rate on the maximum load day: refers to the ratio of the standard deviation of the load curve on the day with the maximum user load to the average value, which is a negative characteristic indicator;

[0095] Weekend load rate: refers to the ratio of the average load of users on rest days to the average load on weekdays, which is a positive characteristic indicator;

[0096] Weekly average load fluctuation rate: refers to the average ratio of the standard deviation of the user's weekly load curve to the average value, which is a negative characteristic indicator;

[0097] Monthly average daily load rate: refers to the average value of daily load rate within a month (i.e. the ratio of average daily load to maximum daily load), which is a positive characteristic indicator;

[0098] Monthly load rate: refers to the ratio of the average daily electricity consumption in a month to the maximum daily electricity consumption, which is a positive characteristic indicator.

[0099] Furthermore, the temperature sensitivity index is calculated using grey correlation:

[0100] Δmin=min i [min k (|x0(k)-x i (k)|)]

[0101] Δmax=max i [max k (|x0(k)-x i (k)|)]

[0102]

[0103] Among them, x0(k) is the highest daily temperature sequence in a certain area, x i (k) is the maximum daily load sequence of each user in the same period, ξ(k) is the sum of x0(k) and x i The correlation coefficient of the kth element in (k);

[0104] The correlation between the ith sequence and the main sequence is:

[0105]

[0106] Among them, γ i is the correlation between the ith sequence and the main sequence, ξ i (k) is the correlation coefficient of the kth element in the ith sequence.

[0107] S2, calculate the weight of the power consumption behavior index of power users;

[0108] In the embodiment of the present application, the calculation of the indicator weight includes:

[0109] Construct an electricity consumption behavior evaluation matrix consisting of m evaluation schemes and n indicators;

[0110] The initial calculation results of each indicator are normalized to unit values ​​to eliminate the dimensional differences between indicators;

[0111] The entropy value of each evaluation index is calculated, and the weight of each index is calculated based on the entropy value.

[0112] It should be noted that the electricity consumption behavior evaluation matrix consisting of m evaluation schemes and n indicators is:

[0113] X=(x ij ) m×n ,i=1,2,...,m; j=1,2,...,n

[0114] The initial calculation results of each indicator are normalized to unit values, and the indicators are standardized:

[0115]

[0116] Among them, x ij is the indicator data, P ij The standardized indicator data effectively eliminates the dimensional differences between indicators.

[0117] The entropy value of each evaluation index is:

[0118]

[0119] In particular, when P ij =0, P ij LqCy ij =0. j is the entropy value of each evaluation index;

[0120] The weight of each indicator is:

[0121]

[0122] Among them, w j It is the weight coefficient of the electricity consumption behavior of the power user, that is, the indicator weight.

[0123] S3, normalize and dimensionless the indicators and make a comprehensive evaluation;

[0124] In the examples of the present application, the comprehensive evaluation adopts the TOPSIS method, which specifically includes:

[0125] Constructing a standardized decision matrix:

[0126]

[0127] Among them, r ij represents the attribute j of the i-th option after normalization;

[0128] Compute the weighted normalized decision matrix:

[0129] v ij =w i r ij ,i=1,...,n

[0130] Among them, w i is the weight of attribute j, r ij represents the normalized attribute j, v of the ith option ij is the weighted standardized decision matrix;

[0131] Determine the ideal solution A + and negative ideal solution A - , where the positive characteristic index Take the maximum value, negative characteristic index Take the minimum value;

[0132] Calculate solution i to ideal solution A + Distance:

[0133]

[0134] Calculate solution i to negative ideal solution A - Distance:

[0135]

[0136] in, represents the distance from solution i to the ideal solution, represents the distance from solution i to the negative ideal solution;

[0137] Compute the solution that is closest to the ideal solution:

[0138]

[0139] Among them, C i Represents the distance from the solution to the ideal solution.

[0140] It should be noted that the determination of ideal solutions and negative ideal solutions is:

[0141]

[0142] Among them, A + represents the ideal solution, A - represents the negative ideal solution, A - is a positive characteristic indicator, It is a negative characteristic indicator.

[0143] S4, based on the fuzzy C-means clustering algorithm, establish the characteristic profile of power users;

[0144] It should be noted that the method for establishing a characteristic profile of power users in the entire industry with quantitative, classified and combined labels is as follows:

[0145] Fuzzy clustering analysis is used to establish a characteristic profile of power users in the entire industry with quantitative, classified and combined labels. The algorithm flow chart is as follows: Figure 2 As shown. First, introduce the number of clusters C, fuzzy index m and data set X for classification, and determine the cluster center. Then calculate the membership matrix U. Next, calculate the cluster center P based on U, and calculate U again from P, and repeat this process until the conditions are met.

[0146] Furthermore, the present embodiment also provides a power user feature classification system based on a fuzzy C-means clustering algorithm, comprising a power data acquisition module, a data processing and evaluation module, and a portrait optimization module;

[0147] The power data collection module is used to collect user electricity price sensitivity data, temperature sensitivity data, power consumption stability data and environmental impact data, and establish a power user power consumption behavior indicator system;

[0148] The data processing and evaluation module is used to pre-process, normalize and evaluate the data based on the electricity consumption behavior indicator system;

[0149] The portrait optimization module is used to monitor the characteristic portraits of power users, establish an evaluation system through power consumption characteristic parameters, and regularly update and optimize the power user characteristic classification method.

[0150] This embodiment also provides a computer device, which is suitable for the case of a method for classifying power user characteristics based on a fuzzy C-means clustering algorithm, and includes a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the method for classifying power user characteristics based on a fuzzy C-means clustering algorithm as proposed in the above embodiment.

[0151] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0152] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for classifying characteristics of power users based on a fuzzy C-means clustering algorithm as proposed in the above embodiment is implemented.

[0153] In summary, the industry-wide power user feature classification method based on fuzzy C-means clustering algorithm provided by the present invention ensures the scientificity and systematicness of user classification by establishing a power user power consumption behavior indicator system, improves the accuracy of user portraits, reduces classification deviations, and improves the efficiency and reliability of power resource allocation; by preprocessing and normalizing data based on power consumption behavior indicators, and conducting comprehensive evaluation and analysis, power users are accurately classified into different power consumption feature types, and corresponding evaluation methods are adopted, which improves the accuracy of classification, reduces management costs and resource waste, and ensures the economy and sustainability of power system operation; a user portrait evaluation system is established through power consumption feature parameters, and the power user feature classification method is regularly updated and optimized, which improves classification efficiency, optimizes user service quality, and ensures the modernization and efficiency of the classification method. The present invention achieves better results in classification accuracy, operating costs, and service efficiency.

[0154] Example 3

[0155] This is the third embodiment of the present invention. This embodiment provides a method for classifying power user characteristics based on a fuzzy C-means clustering algorithm. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0156] In order to verify the effectiveness of the method of the present invention, 50 enterprise users in a provincial industrial park were selected as test subjects, and the test period was from June 1, 2023 to August 31, 2023 (a total of 92 days). The test used high-precision smart meters for data collection, and the sampling frequency was 5 minutes / time. The test was divided into a control group (traditional K-means clustering method) and an experimental group (the method of the present invention).

[0157] First, data collection and preprocessing are performed. In the collection stage, the smart meter collection system records the electricity consumption data of each user during the test period, including the 15-minute average load, peak and valley power consumption, and maximum demand. At the same time, the hourly temperature data during the test period is obtained from the meteorological department. In the data preprocessing stage, the median filling method is used to handle missing values, and the 3σ criterion is used to eliminate outliers, and finally a valid data set is obtained.

[0158] Then, we constructed an index system for electricity consumption behavior. Based on the data during the test period, we calculated the electricity price sensitivity index (including peak-valley load transfer rate, flat-peak load transfer rate, and flat-valley load transfer rate), temperature sensitivity index (using the grey correlation calculation method), and electricity consumption stability index (including peak-valley difference rate, load fluctuation rate, etc.) of each user. To ensure the comparability of the data, July was selected as the key analysis month.

[0159] The improved entropy weight method is used to calculate the indicator weights. First, a 10×12 evaluation matrix (10 typical users, 12 evaluation indicators) is constructed. After normalizing the matrix, the entropy value of each indicator is calculated to obtain the weight coefficient. On this basis, the TOPSIS method is used for comprehensive evaluation to calculate the relative closeness of each user to the ideal solution.

[0160] Finally, the traditional K-means method and the fuzzy C-means clustering algorithm of the present invention are used to classify users. The number of clusters is set to 4, corresponding to high-sensitivity and high-volatility type, high-sensitivity and low-volatility type, low-sensitivity and high-volatility type, and low-sensitivity and low-volatility type users. The number of iterations is set to 100, and the convergence threshold is set to 0.001.

[0161] The test results are shown in the following table:

[0162]

[0163] The following conclusions can be drawn from the analysis of the test data:

[0164] In terms of classification accuracy: the classification accuracy of the method of the present invention is significantly higher than that of the traditional method, with the accuracy of Class A users increasing by 11.8 percentage points (from 82.5% to 94.3%) and that of Class B users increasing by 12.8 percentage points (from 80.8% to 93.6%). This is mainly due to the fact that the fuzzy C-means clustering algorithm adopted by the present invention can better handle the ambiguity and uncertainty of user electricity consumption behavior, while taking into account the comprehensive impact of the three dimensions of temperature sensitivity, electricity price sensitivity and electricity consumption stability.

[0165] Computational efficiency: The computational time of the method of the present invention is reduced by about 15% compared with the traditional method. This is because the present invention adopts the improved entropy weight method to calculate the index weight, simplifies the weight calculation process, optimizes the convergence conditions of the clustering algorithm, and reduces the number of unnecessary iterations.

[0166] In terms of stability index: the stability index of the method of the present invention reaches above 0.88, which is about 20% higher than that of the traditional method. This shows that the classification results obtained by the method of the present invention are more stable and reliable, and are not easily affected by data fluctuations. This is mainly because the present invention fully considers the relationship between various types of electricity consumption characteristics when constructing the index system, and effectively captures the correlation between these characteristics through methods such as gray correlation.

[0167] In terms of peak load potential assessment accuracy: the assessment accuracy of the method of the present invention exceeds 90%, which is about 15 percentage points higher than that of the traditional method. This shows that the method of the present invention can more accurately identify the user's power consumption characteristics and load regulation capabilities, providing a reliable basis for demand-side response and precise load management.

[0168] User satisfaction: The method of the present invention has achieved a high user satisfaction (over 92%), which is about 8-9 percentage points higher than the traditional method. This is because the classification results of the method of the present invention are more in line with the actual power consumption characteristics of users, and can better provide users with personalized power services and energy-saving suggestions.

[0169] In summary, the method of the present invention is significantly superior to traditional methods in terms of classification accuracy, computational efficiency, result stability, peak load potential assessment, and user satisfaction, and has obvious technical advantages and application value. In particular, when processing large-scale, multi-dimensional electricity consumption data, the method of the present invention exhibits stronger adaptability and reliability. These advantages provide strong technical support for the refined management of power users and the implementation of demand-side response.

[0170] 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 preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for classifying power user characteristics based on fuzzy C-means clustering algorithm, characterized by: include, Collect user data from various industries and build an indicator system for power users' electricity consumption behavior; Calculate the weights of electricity users' electricity consumption behavior indicators; Normalize and dimensionless the indicators and make a comprehensive evaluation; Based on the fuzzy C-means clustering algorithm, a characteristic profile of power users is established.

2. The method for classifying power user characteristics based on fuzzy C-means clustering algorithm according to claim 1, characterized in that: The power user electricity consumption behavior indicator system includes: The electricity price sensitivity index refers to the load transfer rate, which is used to characterize the ratio of user load transfer; The temperature sensitivity index refers to the relationship between the load curve change and the temperature curve change; The electricity consumption stability index is used to measure the degree of load fluctuation of the power system within a certain period of time.

3. The method for classifying power user characteristics based on fuzzy C-means clustering algorithm according to claim 2, characterized in that: The electricity price sensitivity index includes: Peak-valley load transfer rate refers to the ratio of the difference between the load during the peak period and the load during the valley period to the average power consumption during the peak period; The off-peak load transfer rate refers to the proportion of users transferring part of their load from the normal period to the peak period through the electricity price incentive mechanism; The off-peak load transfer rate refers to the proportion of users transferring part of their load from normal periods to off-peak periods through the electricity price incentive mechanism.

4. The method for classifying power user characteristics based on fuzzy C-means clustering algorithm according to claim 3, characterized in that: The electricity consumption stability index includes: The peak-to-valley difference rate on the maximum load day refers to the percentage of the maximum daily peak-to-valley difference of user load to the maximum load; The load fluctuation rate on the maximum load day refers to the ratio of the standard deviation of the load curve on the day with the maximum user load to the average value; Weekend load rate refers to the ratio of the average load of users on rest days to the average load on weekdays; Weekly average load fluctuation rate refers to the average ratio of the standard deviation of the user's weekly load curve to the average value; Monthly average daily load rate refers to the average value of daily load rates within a month; The monthly load rate refers to the ratio of the average daily electricity consumption to the maximum daily electricity consumption in a month.

5. The method for classifying power user characteristics based on fuzzy C-means clustering algorithm according to claim 4, characterized in that: The calculation of the indicator weights includes: Construct an electricity consumption behavior evaluation matrix consisting of m evaluation schemes and n indicators; The initial calculation results of each indicator are normalized to unit values ​​to eliminate the dimensional differences between indicators; The entropy value of each evaluation index is calculated, and the weight of each index is calculated based on the entropy value.

6. The method for classifying power user characteristics based on fuzzy C-means clustering algorithm according to claim 5, characterized in that: The comprehensive evaluation adopts the TOPSIS method, which specifically includes: Constructing a standardized decision matrix: Among them, r ij represents the attribute j of the i-th option after normalization; Compute the weighted normalized decision matrix: v ij =w i r ij ,i=1,...,n Among them, w i is the weight of attribute j, r ij represents the normalized attribute j, v of the ith option ij is the weighted standardized decision matrix; Determine the ideal solution A + and negative ideal solution A - , where the positive characteristic index Take the maximum value, negative characteristic index Take the minimum value; Calculate solution i to ideal solution A + Distance: Calculate solution i to negative ideal solution A - Distance: in, represents the distance from solution i to the ideal solution, represents the distance from solution i to the negative ideal solution; Compute the solution that is closest to the ideal solution: Among them, C i Represents the distance from the solution to the ideal solution.

7. The method for classifying power user characteristics based on fuzzy C-means clustering algorithm according to claim 6, characterized in that: The temperature sensitivity index is calculated using grey correlation: Δmin=min i [my k (|x0(k)-x i (k)|)] Δmax=max i [max k (|x0(k)-x i (k)|)] Among them, x0(k) is the highest daily temperature sequence in a certain area, x i (k) is the maximum daily load sequence of each user in the same period, ξ(k) is the sum of x0(k) and x i The correlation coefficient of the kth element in (k); The correlation between the ith sequence and the main sequence is: Among them, γ i is the correlation between the ith sequence and the main sequence, ξ i (k) is the correlation coefficient of the kth element in the ith sequence.

8. A power user characteristic classification system based on fuzzy C-means clustering algorithm, based on the power user characteristic classification method based on fuzzy C-means clustering algorithm according to any one of claims 1 to 7, characterized in that: It also includes a power data acquisition module, a data processing and evaluation module, and a portrait optimization module; The power data collection module is used to collect user electricity price sensitivity data, temperature sensitivity data, power consumption stability data and environmental impact data, and establish a power user power consumption behavior indicator system; The data processing and evaluation module is used to pre-process, normalize and evaluate the data based on the electricity consumption behavior indicator system; The portrait optimization module is used to monitor the characteristic portraits of power users, establish an evaluation system through power consumption characteristic parameters, and regularly update and optimize the power user characteristic classification method.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for classifying electric power user characteristics based on the fuzzy C-means clustering algorithm described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for classifying electric power user characteristics based on the fuzzy C-means clustering algorithm described in any one of claims 1 to 7 are implemented.

Images