User load characteristic portrait construction method and system based on branch load monitoring

Abstract

The invention discloses a user load characteristic portrait construction method and system based on branch load monitoring, and relates to the technical field of power load data analysis, and the method comprises the steps: obtaining branch load monitoring data, obtaining power network information, and obtaining user power load data according to the branch load monitoring data and the power network information. According to the method, the reliability and consistency of data are ensured through a data index smoothing coefficient and a linear interpolation method, user power load standard data are analyzed through a data principal component analysis method, user power load characteristic information is obtained through a user power load potential evaluation index, and a user power load electric appliance evaluation index is used for evaluating a user power load electric appliance. The user electric appliance load characteristic information is obtained, the user load characteristic clustering information is obtained, the user load characteristic portrait information is obtained based on user portrait construction, and the user load characteristic portrait construction efficiency and the portrait accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power load data analysis, and in particular to a method and system for constructing a user load characteristic profile based on branch line load monitoring. Background Art

[0002] With the growth of electricity demand brought about by economic and social development and the increasingly severe supply and demand situation, proper load management of electricity users is one of the important means to give full play to the flexibility and economic advantages of demand-side resources, thereby promoting the balance of electricity supply and demand and ensuring the stability of the power grid.

[0003] By fully analyzing the user's electricity consumption pattern based on the branch line data, the electricity consumption characteristics and rules of different users can be obtained, which is of great significance for accurately identifying the load characteristics of power users, providing differentiated services, and guiding the efficient implementation of relevant load management measures. By building a portrait of user load characteristics, we can better understand the user's electricity consumption behavior, needs and preferences. In this way, we can provide users with more personalized electricity services, including customized electricity consumption plans, optimized power supply, etc., to improve user satisfaction and experience.

[0004] At present, there are still problems in the construction of user load characteristic portraits, such as the inability to accurately obtain user load data based on power load data, the inability to accurately evaluate user power load potential based on user load data, and the failure to consider user energy-saving potential assessment and user appliance overload risk assessment, making it impossible to accurately construct user load characteristic portraits. Summary of the invention

[0005] In order to solve the above technical problems, a method and system for constructing a user load characteristic portrait based on branch line load monitoring are provided. This technical solution solves the problems raised in the above background technology that the user's load data cannot be accurately obtained based on the power load data, the user's power load potential cannot be accurately evaluated based on the user's load data, and the user's energy-saving potential assessment and user electrical appliance overload risk assessment are not considered, and the user's load characteristic portrait cannot be accurately constructed.

[0006] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0007] The method for constructing a user load characteristic profile based on branch line load monitoring includes:

[0008] Acquire branch road load monitoring data, wherein the branch road load monitoring data includes branch road power load time and branch road power load intensity;

[0009] Acquiring power network information, wherein the power network information includes power system circuit information and user distribution information;

[0010] Obtain user power load data based on branch line load monitoring data and power network information;

[0011] According to the user's power load data, based on data preprocessing, the user's power load standard data is obtained;

[0012] According to the user's power load standard data, based on data analysis, the user's power load characteristic data is obtained;

[0013] According to the user's power load characteristic data and based on the user's load dispatch evaluation, the user's power load potential evaluation index is obtained;

[0014] According to the user's power load potential evaluation index, obtain the user's power load characteristic information;

[0015] Acquire user load electrical appliance information, wherein the user load electrical appliance information includes user load electrical appliance power data and user load electrical appliance type information;

[0016] Obtaining a user load appliance evaluation index based on user load appliance information and user power load characteristic data;

[0017] According to the user load electrical appliance evaluation index, obtain the user electrical appliance load characteristic information;

[0018] According to the user's power load characteristic information and the user's electrical appliance load characteristic information, based on feature clustering, the user's load characteristic clustering information is obtained;

[0019] According to the user load characteristic clustering information, based on the user portrait construction, the user load characteristic portrait information is obtained.

[0020] Preferably, the obtaining of user power load standard data based on user power load data and data preprocessing specifically includes:

[0021] According to the user's power load data, obtain the user's power load time series information;

[0022] According to the user's power load time series information, based on the multiplication time series model, the user's power load seasonal decomposition information is obtained;

[0023] Obtaining user power load seasonal data based on user power load seasonal decomposition information and user power load data;

[0024] Obtain smoothing coefficient information and residual threshold according to actual analysis and statistical requirements;

[0025] Obtaining user power load residual data based on smoothing coefficient information and user power load seasonal data;

[0026] According to the user's power load residual data and the residual threshold, determine whether the user's power load residual data exceeds the residual threshold. If so, the user's power load seasonal data is abnormal data and is removed. If not, the user's power load seasonal data is normal.

[0027] Based on the linear interpolation method, a small amount of missing data is filled to obtain the user's power load supplementary data:

[0028]

[0029] In the formula, X n is the power load at the nth moment in the user's power load data, X n+j is the power consumption at the n+ith moment in the user's power load data, X n+j The power consumption at the n+jth moment in the user's power load data;

[0030] According to the user's power load supplement data, based on the minimum-maximum standardization of the data, the user's power load standard data is obtained.

[0031] Preferably, the acquiring of user power load characteristic data based on data analysis according to user power load standard data specifically includes:

[0032] Obtaining user power load characteristic information based on user power load standard data;

[0033] According to the requirements for building a user load characteristic profile, the user's power load characteristic information is screened to obtain the portrait power load characteristic information;

[0034] According to the portrait power load characteristic information and the user power load standard data, based on data standardization, the user power load characteristic standardized data is obtained, wherein each characteristic data of the user power load characteristic standardized data has a mean value of 0 and a variance of 1;

[0035] According to the user's power load characteristic standardization data, the characteristic data covariance matrix is ​​obtained;

[0036] According to the covariance matrix of the characteristic data, based on the eigenvalue decomposition, the eigenvalue information and the eigenvector corresponding to the eigenvalue are obtained;

[0037] Sort the eigenvalue information from large to small to obtain principal component information, where the principal component information is the eigenvectors corresponding to the first k eigenvalues;

[0038] According to the principal component information and the user's power load standard data, the user's power load characteristic data is obtained based on the data linear combination projection.

[0039] Preferably, obtaining the user power load characteristic information according to the user power load potential evaluation index specifically includes:

[0040] According to the user's power load characteristic data, obtain the user's power load potential evaluation index;

[0041] According to the user power load potential evaluation index and based on the user power load potential evaluation model, the user power load potential evaluation index is obtained;

[0042] According to the user power load potential evaluation index, the user power load characteristic information is obtained, wherein the user power load characteristic information includes user power demand information, user power consumption time period information and user power load potential evaluation information;

[0043] Among them, the calculation formula of the user power load potential evaluation index is:

[0044]

[0045] In the formula, Q is the user's power load potential evaluation index, α1 is the weight of the user's power load peak time, T user is the user's power load peak time period, T peak is the peak time period of regional power load, α2 is the weight of the user power load response time, t wait is the user's power load response demand time, α3 is the user's power load weight, E user The power load of the user.

[0046] Preferably, obtaining the user's electrical appliance load characteristic information according to the user's load electrical appliance evaluation index specifically includes:

[0047] According to the user load appliance information and the user power load characteristic data, the user load appliance evaluation index factor is obtained;

[0048] According to the user load electrical appliance evaluation index factor, based on weight allocation, the user load electrical appliance evaluation index is obtained;

[0049] According to the user load appliance evaluation index, the user appliance load characteristic information is obtained, wherein the user appliance load characteristic information includes user energy saving potential evaluation information, user appliance risk evaluation information and user appliance overload information;

[0050] Among them, the calculation formula of the user load electrical appliance evaluation index is:

[0051]

[0052] Where R is the user load electrical appliance evaluation index, β1 is the weight of the user energy saving potential evaluation index, and P kis the useless power of the kth user appliance, t k is the ineffective working time of the kth user's electrical appliance, τ is the energy-saving potential evaluation coefficient of the user's electrical appliance, P norm (k) is the rated working power of the kth user's electrical appliance, T k is the effective working time of the kth user's electrical appliance, β2 is the weight of the user's electrical appliance risk assessment, P overload (k) is the overload power of the kth user's electrical appliance, t overload (k) is the overload time of the kth user's electrical appliance, User electrical appliance risk assessment factor.

[0053] Preferably, the acquiring of user load characteristic clustering information based on feature clustering according to user power load characteristic information and user electrical appliance load characteristic information specifically includes:

[0054] Step 1: According to the user's power load characteristic information and the user's electrical appliance load characteristic information, based on the load characteristic clustering requirements, obtain the initial cluster center point information;

[0055] Step 2: According to the initial cluster center point information, obtain the characteristic cluster center distance data, and assign it to the cluster center cluster with the nearest distance based on the characteristic cluster center distance data;

[0056] Step 3: Obtain the mean value of the cluster center data, and obtain the cluster center update information based on the cluster center point update;

[0057] Step 4: According to the feature clustering requirements, obtain the cluster center change threshold, and repeat steps 2 and 3 until the cluster center change degree is less than the cluster center change threshold;

[0058] Step 5: Obtain user load characteristic clustering information, wherein the user load characteristic clustering information includes cluster center point information and cluster center cluster information.

[0059] Furthermore, a user load characteristic profile building system based on branch line load monitoring is proposed, which is used to implement the above-mentioned profile building method, including:

[0060] A main control module, the main control module is used to divide user load characteristic data into different groups according to the similarity of user load characteristics, obtain user load characteristic clustering information based on feature clustering according to user power load characteristic information and user electrical appliance load characteristic information, and obtain user load characteristic portrait information based on user portrait construction according to user load characteristic clustering information;

[0061] An information acquisition module, the information acquisition module is used to acquire branch line load monitoring data, branch line power load time, branch line power load intensity, power network information, power system circuit information and user distribution information, acquire user power load data based on the branch line load monitoring data and power network information, and acquire user power load standard data based on the user power load data and data preprocessing;

[0062] An evaluation module, the evaluation module is used to obtain user power load characteristic data based on user power load standard data and data analysis, obtain user power load characteristic information based on user power load potential evaluation index, and obtain user electrical appliance load characteristic information based on user load electrical appliance evaluation index;

[0063] The display module is used to display user power load characteristic information, user electrical appliance load characteristic information and user load characteristic portrait.

[0064] Optionally, the main control module specifically includes:

[0065] A control unit, the control unit is used to obtain user load characteristic profile information based on user load characteristic clustering information and user profile construction;

[0066] An information receiving unit, which interacts with the information acquisition module and the evaluation module to acquire data, user power load characteristic information and user electrical appliance load characteristic information, and transmits the information to the feature clustering unit;

[0067] The feature clustering unit is used to divide the user load characteristic data into different groups according to the similarity of the user load characteristics, and obtain the user load characteristic clustering information based on feature clustering according to the user power load characteristic information and the user electrical appliance load characteristic information.

[0068] Optionally, the information acquisition module specifically includes:

[0069] An information acquisition unit, the information acquisition unit is used to acquire branch line load monitoring data, branch line power load time, branch line power load intensity, power network information, power system circuit information and user distribution information;

[0070] An information processing unit is used to obtain user power load data based on branch line load monitoring data and power network information, and to obtain user power load standard data based on user power load data and data preprocessing.

[0071] Optionally, the evaluation module specifically includes:

[0072] A feature extraction unit, wherein the feature extraction unit is used to obtain user power load feature data based on data analysis according to user power load standard data,

[0073] An electric load evaluation unit, the electric load evaluation unit is used to obtain a user electric load potential evaluation index according to the user electric load potential evaluation index and based on the user electric load potential evaluation model, and obtain user electric load characteristic information according to the user electric load potential evaluation index;

[0074] The load appliance evaluation unit is used to obtain the user load appliance evaluation index based on the user load appliance evaluation index and the weight distribution, and obtain the user appliance load characteristic information according to the user load appliance evaluation index.

[0075] Compared with the prior art, the present invention has the following beneficial effects:

[0076] The present invention proposes a method and system for constructing a user load characteristic portrait based on branch line load monitoring, ensures data reliability and consistency through data exponential smoothing coefficient and linear interpolation method, analyzes user power load standard data through data principal component analysis method, obtains user power load characteristic information through user power load potential evaluation index, obtains user electrical appliance load characteristic information through user load electrical appliance evaluation index, obtains user load characteristic clustering information based on user power load characteristic information and user electrical appliance load characteristic information, obtains user load characteristic portrait information based on user portrait construction, and improves the construction efficiency of user load characteristic portrait and the accuracy of portrait. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 A flow chart of a method for constructing a user load characteristic profile based on branch line load monitoring proposed in the present invention;

[0078] Figure 2 This is a flowchart of user power load data preprocessing in the present invention;

[0079] Figure 3 A flow chart for obtaining user power load characteristic data in the present invention;

[0080] Figure 4 A flow chart of clustering user power load characteristic information and user electrical appliance load characteristic information in the present invention;

[0081] Figure 5 This is a system structure diagram for constructing a user load characteristic profile based on branch line load monitoring proposed in the present invention. DETAILED DESCRIPTION

[0082] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0083] Reference Figure 1-4 As shown, the method for constructing a user load characteristic profile based on branch line load monitoring in an embodiment of the present invention includes:

[0084] Acquire branch road load monitoring data, wherein the branch road load monitoring data includes branch road power load time and branch road power load intensity;

[0085] Acquiring power network information, wherein the power network information includes power system circuit information and user distribution information;

[0086] Obtain user power load data based on branch line load monitoring data and power network information;

[0087] According to the user's power load data, based on data preprocessing, the user's power load standard data is obtained;

[0088] Specifically, the user's power load data is preprocessed, including:

[0089] According to the user's power load data, obtain the user's power load time series information;

[0090] According to the user's power load time series information, based on the multiplication time series model, the user's power load seasonal decomposition information is obtained;

[0091] Obtaining user power load seasonal data based on user power load seasonal decomposition information and user power load data;

[0092] Obtain smoothing coefficient information and residual threshold according to actual analysis and statistical requirements;

[0093] Obtaining user power load residual data based on smoothing coefficient information and user power load seasonal data;

[0094] According to the user's power load residual data and the residual threshold, determine whether the user's power load residual data exceeds the residual threshold. If so, the user's power load seasonal data is abnormal data and is removed. If not, the user's power load seasonal data is normal.

[0095] Based on the linear interpolation method, a small amount of missing data is filled to obtain the user's power load supplementary data:

[0096]

[0097] In the formula, X nis the power load at the nth moment in the user's power load data, X n+i is the power consumption at the n+ith moment in the user's power load data, X n+j The power consumption at the n+jth moment in the user's power load data;

[0098] According to the user's power load supplement data, based on the minimum-maximum standardization of the data, the user's power load standard data is obtained.

[0099] In this scheme, the user's power load residual data is obtained through the smoothing coefficient information and the user's power load seasonal data. According to the user's power load residual data and the residual threshold, it is judged whether the user's power load seasonal data is abnormal data. The small amount of missing data is filled through linear interpolation to ensure the reliability and consistency of the data.

[0100] According to the user's power load standard data, based on data analysis, the user's power load characteristic data is obtained;

[0101] Specifically, by analyzing the user's power load standard data, the user's power load characteristic data is obtained, including:

[0102] Obtaining user power load characteristic information based on user power load standard data;

[0103] According to the requirements for building a user load characteristic profile, the user's power load characteristic information is screened to obtain the portrait power load characteristic information;

[0104] According to the portrait power load characteristic information and the user power load standard data, based on data standardization, the user power load characteristic standardized data is obtained, wherein each characteristic data of the user power load characteristic standardized data has a mean value of 0 and a variance of 1;

[0105] Among them, according to the user power load standard data, a user power load standard data matrix is ​​constructed;

[0106]

[0107] Where D is the standard data matrix of user power load, d ij is the user power load data of the jth feature at the i-th moment, m is the total number of moments of the user power load standard data, and n is the total number of features of the user power load standard data;

[0108]

[0109] Where Z is the standardized data of user power load characteristics, μ is the mean vector of each feature, and σ is the standard deviation vector of each feature;

[0110] According to the user's power load characteristic standardization data, the characteristic data covariance matrix is ​​obtained;

[0111] Among them, the feature data covariance matrix H size is n*n, where x pq is the covariance of the p-th feature and the q-th feature;

[0112]

[0113] Where D T is the transposed matrix of the user's power load standard data matrix;

[0114] According to the covariance matrix of the characteristic data, based on the eigenvalue decomposition, the eigenvalue information and the eigenvector corresponding to the eigenvalue are obtained;

[0115] Sort the eigenvalue information from large to small to obtain principal component information, where the principal component information is the eigenvectors corresponding to the first k eigenvalues;

[0116] According to the principal component information and the user's power load standard data, the user's power load characteristic data is obtained based on the data linear combination projection.

[0117] According to the user's power load characteristic data and based on the user's load dispatch evaluation, the user's power load potential evaluation index is obtained;

[0118] According to the user's power load potential evaluation index, obtain the user's power load characteristic information;

[0119] Specifically, according to the user's power load potential evaluation index, the user's power load characteristic information is obtained, including:

[0120] According to the user's power load characteristic data, obtain the user's power load potential evaluation index;

[0121] According to the user power load potential evaluation index and based on the user power load potential evaluation model, the user power load potential evaluation index is obtained;

[0122] According to the user power load potential evaluation index, the user power load characteristic information is obtained, wherein the user power load characteristic information includes user power demand information, user power consumption time period information and user power load potential evaluation information;

[0123] Among them, the calculation formula of the user power load potential evaluation index is:

[0124]

[0125] In the formula, Q is the user's power load potential evaluation index, α1 is the weight of the user's power load peak time, T useris the user's power load peak time period, T peak is the peak time period of regional power load, α2 is the weight of the user power load response time, t wait is the user's power load response demand time, α3 is the user's power load weight, E user The power load of the user.

[0126] In this solution, the user power load potential evaluation index is obtained based on the user power load potential evaluation model through the user power load potential evaluation index, and the user power load characteristic information is obtained according to the user power load potential evaluation index. By analyzing the user power load potential evaluation index, information about the user power load characteristics is obtained, such as the user's regular power consumption time, frequently used power equipment, etc. This information can help the power system better predict and dispatch loads.

[0127] Acquire user load electrical appliance information, wherein the user load electrical appliance information includes user load electrical appliance power data and user load electrical appliance type information;

[0128] Obtaining a user load appliance evaluation index based on user load appliance information and user power load characteristic data;

[0129] According to the user load electrical appliance evaluation index, obtain the user electrical appliance load characteristic information;

[0130] Specifically, according to the user load electrical appliance evaluation index, the user electrical appliance load characteristic information is obtained, including:

[0131] According to the user load appliance information and the user power load characteristic data, the user load appliance evaluation index factor is obtained;

[0132] According to the user load electrical appliance evaluation index factor, based on weight allocation, the user load electrical appliance evaluation index is obtained;

[0133] According to the user load appliance evaluation index, the user appliance load characteristic information is obtained, wherein the user appliance load characteristic information includes user energy saving potential evaluation information, user appliance risk evaluation information and user appliance overload information;

[0134] Among them, the calculation formula of the user load electrical appliance evaluation index is:

[0135]

[0136] Where R is the user load electrical appliance evaluation index, β1 is the weight of the user energy saving potential evaluation index, and P k is the useless power of the kth user appliance, t k is the ineffective working time of the kth user's electrical appliance, τ is the energy-saving potential evaluation coefficient of the user's electrical appliance, Pnorm (k) is the rated working power of the kth user's electrical appliance, T k is the effective working time of the kth user's electrical appliance, β2 is the weight of the user's electrical appliance risk assessment, P overload (k) is the overload power of the kth user's electrical appliance, t overload (k) is the overload time of the kth user's electrical appliance, User electrical appliance risk assessment factor.

[0137] In this scheme, based on the user load appliance information and the user power load characteristic data, the user load appliance evaluation index factor is obtained, and based on the user load appliance evaluation index factor, the user load appliance evaluation index is obtained based on weight distribution. Based on the user load appliance evaluation index, the user appliance load characteristic information is obtained, and the characteristics and evaluation information of the user load appliance are accurately described. This can find room for energy-saving improvements, provide a basis for energy management and policy formulation, analyze user appliance overload, and provide a reference for the safe operation of the power system.

[0138] According to the user's power load characteristic information and the user's electrical appliance load characteristic information, based on feature clustering, the user's load characteristic clustering information is obtained;

[0139] Specifically, according to the user power load characteristic information and the user electrical appliance load characteristic information, based on feature clustering, the user load characteristic clustering information is obtained, and the specific steps are as follows:

[0140] Step 1: According to the user's power load characteristic information and the user's electrical appliance load characteristic information, based on the load characteristic clustering requirements, obtain the initial cluster center point information;

[0141] Step 2: According to the initial cluster center point information, obtain the characteristic cluster center distance data, and assign it to the cluster center cluster with the nearest distance based on the characteristic cluster center distance data;

[0142] Step 3: Obtain the mean value of the cluster center data, and obtain the cluster center update information based on the cluster center point update;

[0143] Step 4: According to the feature clustering requirements, obtain the cluster center change threshold, and repeat steps 2 and 3 until the cluster center change degree is less than the cluster center change threshold;

[0144] Step 5: Obtain user load characteristic clustering information, wherein the user load characteristic clustering information includes cluster center point information and cluster center cluster information.

[0145] According to the user load characteristic clustering information, based on the user portrait construction, the user load characteristic portrait information is obtained.

[0146] Reference Figure 5As shown, further, in combination with the above-mentioned method for constructing a user load characteristic profile based on branch road load monitoring, a user load characteristic profile construction system based on branch road load monitoring is proposed, including:

[0147] A main control module, the main control module is used to divide user load characteristic data into different groups according to the similarity of user load characteristics, obtain user load characteristic clustering information based on feature clustering according to user power load characteristic information and user electrical appliance load characteristic information, and obtain user load characteristic portrait information based on user portrait construction according to user load characteristic clustering information;

[0148] An information acquisition module, the information acquisition module is used to acquire branch line load monitoring data, branch line power load time, branch line power load intensity, power network information, power system circuit information and user distribution information, acquire user power load data based on the branch line load monitoring data and power network information, and acquire user power load standard data based on the user power load data and data preprocessing;

[0149] An evaluation module, the evaluation module is used to obtain user power load characteristic data based on user power load standard data and data analysis, obtain user power load characteristic information based on user power load potential evaluation index, and obtain user electrical appliance load characteristic information based on user load electrical appliance evaluation index;

[0150] The display module is used to display user power load characteristic information, user electrical appliance load characteristic information and user load characteristic portrait.

[0151] Main control module, specifically including:

[0152] A control unit, the control unit is used to obtain user load characteristic profile information based on user load characteristic clustering information and user profile construction;

[0153] An information receiving unit, which interacts with the information acquisition module and the evaluation module to acquire data, user power load characteristic information and user electrical appliance load characteristic information, and transmits the information to the feature clustering unit;

[0154] The feature clustering unit is used to divide the user load characteristic data into different groups according to the similarity of the user load characteristics, and obtain the user load characteristic clustering information based on feature clustering according to the user power load characteristic information and the user electrical appliance load characteristic information.

[0155] Information acquisition module, specifically including:

[0156] An information acquisition unit, the information acquisition unit is used to acquire branch line load monitoring data, branch line power load time, branch line power load intensity, power network information, power system circuit information and user distribution information;

[0157] An information processing unit is used to obtain user power load data based on branch line load monitoring data and power network information, and to obtain user power load standard data based on user power load data and data preprocessing.

[0158] Assessment modules include:

[0159] A feature extraction unit, wherein the feature extraction unit is used to obtain user power load feature data based on data analysis according to user power load standard data,

[0160] An electric load evaluation unit, the electric load evaluation unit is used to obtain a user electric load potential evaluation index according to the user electric load potential evaluation index and based on the user electric load potential evaluation model, and obtain user electric load characteristic information according to the user electric load potential evaluation index;

[0161] The load appliance evaluation unit is used to obtain the user load appliance evaluation index based on the user load appliance evaluation index and the weight distribution, and obtain the user appliance load characteristic information according to the user load appliance evaluation index.

[0162] In summary, the advantages of the present invention are: outliers in user power load data are removed by using a data exponential smoothing coefficient, missing data are supplemented by a linear interpolation method, the reliability and consistency of the data are ensured, errors in data risks are avoided, and the construction of user load characteristic portraits is affected; user power load standard data is analyzed by a data principal component analysis method to obtain user power load characteristic data; user power load characteristic information is obtained by using a user power load potential evaluation index; user electrical appliance load characteristic information is obtained by using a user load electrical appliance evaluation index; user load characteristic clustering information is obtained based on feature clustering according to user power load characteristic information and user electrical appliance load characteristic information; user load characteristic portrait information is obtained based on user load characteristic clustering information and user portrait construction according to user load characteristic clustering information, thereby improving the construction efficiency of user load characteristic portraits and the accuracy of portraits.

[0163] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for constructing a user load characteristic profile based on branch line load monitoring, characterized in that: include: Acquire branch road load monitoring data, wherein the branch road load monitoring data includes branch road power load time and branch road power load intensity; Acquiring power network information, wherein the power network information includes power system circuit information and user distribution information; Obtain user power load data based on branch line load monitoring data and power network information; According to the user's power load data, based on data preprocessing, the user's power load standard data is obtained; According to the user's power load standard data, based on data analysis, the user's power load characteristic data is obtained; According to the user's power load characteristic data and based on the user's load dispatch evaluation, the user's power load potential evaluation index is obtained; According to the user's power load potential evaluation index, obtain the user's power load characteristic information; Acquire user load electrical appliance information, wherein the user load electrical appliance information includes user load electrical appliance power data and user load electrical appliance type information; Obtaining a user load appliance evaluation index based on user load appliance information and user power load characteristic data; According to the user load electrical appliance evaluation index, obtain the user electrical appliance load characteristic information; According to the user's power load characteristic information and the user's electrical appliance load characteristic information, based on feature clustering, the user's load characteristic clustering information is obtained; According to the user load characteristic clustering information, based on the user portrait construction, the user load characteristic portrait information is obtained.

2. The method for constructing a user load characteristic profile based on branch line load monitoring according to claim 1 is characterized in that: The step of obtaining user power load standard data based on user power load data and data preprocessing specifically includes: According to the user's power load data, obtain the user's power load time series information; According to the user's power load time series information, based on the multiplication time series model, the user's power load seasonal decomposition information is obtained; Obtaining user power load seasonal data based on user power load seasonal decomposition information and user power load data; Obtain smoothing coefficient information and residual threshold according to actual analysis and statistical requirements; Obtaining user power load residual data based on smoothing coefficient information and user power load seasonal data; According to the user's power load residual data and the residual threshold, determine whether the user's power load residual data exceeds the residual threshold. If so, the user's power load seasonal data is abnormal data and is removed. If not, the user's power load seasonal data is normal. Based on the linear interpolation method, a small amount of missing data is filled to obtain the user's power load supplementary data: Where, X n is the power load at the nth moment in the user's power load data, X n+i is the power consumption at the n+ith moment in the user's power load data, X n+j The power consumption at the n+jth moment in the user's power load data; According to the user's power load supplement data, based on the minimum-maximum standardization of the data, the user's power load standard data is obtained.

3. The method for constructing a user load characteristic profile based on branch line load monitoring according to claim 1 is characterized in that: The obtaining of user power load characteristic data based on data analysis according to user power load standard data specifically includes: Obtaining user power load characteristic information based on user power load standard data; According to the requirements for building a user load characteristic profile, the user's power load characteristic information is screened to obtain the portrait power load characteristic information; According to the portrait power load characteristic information and the user power load standard data, based on data standardization, the user power load characteristic standardized data is obtained, wherein each characteristic data of the user power load characteristic standardized data has a mean value of 0 and a variance of 1; According to the user's power load characteristic standardization data, the characteristic data covariance matrix is ​​obtained; According to the covariance matrix of the characteristic data, based on the eigenvalue decomposition, the eigenvalue information and the eigenvector corresponding to the eigenvalue are obtained; Sort the eigenvalue information from large to small to obtain principal component information, where the principal component information is the eigenvectors corresponding to the first k eigenvalues; According to the principal component information and the user's power load standard data, the user's power load characteristic data is obtained based on the data linear combination projection.

4. The method for constructing a user load characteristic profile based on branch line load monitoring according to claim 1 is characterized in that: The obtaining of user power load characteristic information according to the user power load potential evaluation index specifically includes: According to the user's power load characteristic data, obtain the user's power load potential evaluation index; According to the user power load potential evaluation index and based on the user power load potential evaluation model, the user power load potential evaluation index is obtained; According to the user power load potential evaluation index, the user power load characteristic information is obtained, wherein the user power load characteristic information includes user power demand information, user power consumption time period information and user power load potential evaluation information; Among them, the calculation formula of the user power load potential evaluation index is: In the formula, Q is the user's power load potential evaluation index, α1 is the weight of the user's power load peak time, T user is the user's power load peak time period, T peak is the peak time period of regional power load, α2 is the weight of the user power load response time, t wait is the user's power load response demand time, α3 is the user's power load weight, E user The power load of the user.

5. The method for constructing a user load characteristic profile based on branch line load monitoring according to claim 1 is characterized in that: The step of obtaining the user's electrical appliance load characteristic information according to the user's load electrical appliance evaluation index specifically includes: According to the user load appliance information and the user power load characteristic data, the user load appliance evaluation index factor is obtained; According to the user load electrical appliance evaluation index factor, based on weight allocation, the user load electrical appliance evaluation index is obtained; According to the user load appliance evaluation index, the user appliance load characteristic information is obtained, wherein the user appliance load characteristic information includes user energy saving potential evaluation information, user appliance risk evaluation information and user appliance overload information; Among them, the calculation formula of the user load electrical appliance evaluation index is: Where R is the user load electrical appliance evaluation index, β1 is the weight of the user energy saving potential evaluation index, and P k is the useless power of the kth user appliance, t k is the ineffective working time of the kth user's electrical appliance, τ is the energy-saving potential evaluation coefficient of the user's electrical appliance, P norm (k) is the rated working power of the kth user's electrical appliance, T k is the effective working time of the kth user's electrical appliance, β2 is the weight of the user's electrical appliance risk assessment, P overload (k) is the overload power of the kth user's electrical appliance, t overload (k) is the overload time of the kth user's electrical appliance, User electrical appliance risk assessment factor.

6. The method for constructing a user load characteristic profile based on branch line load monitoring according to claim 1 is characterized in that: The step of obtaining user load characteristic clustering information based on feature clustering according to user power load characteristic information and user electrical appliance load characteristic information specifically includes: Step 1: According to the user's power load characteristic information and the user's electrical appliance load characteristic information, based on the load characteristic clustering requirements, obtain the initial cluster center point information; Step 2: According to the initial cluster center point information, obtain the characteristic cluster center distance data, and assign it to the cluster center cluster with the nearest distance based on the characteristic cluster center distance data; Step 3: Obtain the mean value of the cluster center data, and obtain the cluster center update information based on the cluster center point update; Step 4: According to the feature clustering requirements, obtain the cluster center change threshold, and repeat steps 2 and 3 until the cluster center change degree is less than the cluster center change threshold; Step 5: Obtain user load characteristic clustering information, wherein the user load characteristic clustering information includes cluster center point information and cluster center cluster information.

7. A user load characteristic profile building system based on branch line load monitoring, used to implement the profile building method according to any one of claims 1 to 6, characterized in that: include: A main control module, the main control module is used to divide user load characteristic data into different groups according to the similarity of user load characteristics, obtain user load characteristic clustering information based on feature clustering according to user power load characteristic information and user electrical appliance load characteristic information, and obtain user load characteristic portrait information based on user portrait construction according to user load characteristic clustering information; An information acquisition module, the information acquisition module is used to acquire branch line load monitoring data, branch line power load time, branch line power load intensity, power network information, power system circuit information and user distribution information, acquire user power load data based on the branch line load monitoring data and power network information, and acquire user power load standard data based on the user power load data and data preprocessing; An evaluation module, the evaluation module is used to obtain user power load characteristic data based on user power load standard data and data analysis, obtain user power load characteristic information based on user power load potential evaluation index, and obtain user electrical appliance load characteristic information based on user load electrical appliance evaluation index; The display module is used to display user power load characteristic information, user electrical appliance load characteristic information and user load characteristic portrait.

8. The user load characteristic profile construction system based on branch line load monitoring according to claim 7 is characterized in that: The main control module specifically includes: A control unit, the control unit is used to obtain user load characteristic profile information based on user load characteristic clustering information and user profile construction; An information receiving unit, which interacts with the information acquisition module and the evaluation module to acquire data, user power load characteristic information and user electrical appliance load characteristic information, and transmits the information to the feature clustering unit; The feature clustering unit is used to divide the user load characteristic data into different groups according to the similarity of the user load characteristics, and obtain the user load characteristic clustering information based on feature clustering according to the user power load characteristic information and the user electrical appliance load characteristic information.

9. The user load characteristic profile construction system based on branch line load monitoring according to claim 7 is characterized in that: The information acquisition module specifically includes: An information acquisition unit, the information acquisition unit is used to acquire branch line load monitoring data, branch line power load time, branch line power load intensity, power network information, power system circuit information and user distribution information; An information processing unit is used to obtain user power load data based on branch line load monitoring data and power network information, and to obtain user power load standard data based on user power load data and data preprocessing.

10. The user load characteristic profile construction system based on branch line load monitoring according to claim 7 is characterized in that: The evaluation module specifically includes: A feature extraction unit, wherein the feature extraction unit is used to obtain user power load feature data based on data analysis according to user power load standard data, An electric load evaluation unit, the electric load evaluation unit is used to obtain a user electric load potential evaluation index according to the user electric load potential evaluation index and based on the user electric load potential evaluation model, and obtain user electric load characteristic information according to the user electric load potential evaluation index; The load appliance evaluation unit is used to obtain the user load appliance evaluation index based on the user load appliance evaluation index and the weight distribution, and obtain the user appliance load characteristic information according to the user load appliance evaluation index.

Images