CN-122000948-A - Smart grid-oriented electrical appliance load cluster aggregation method and system

CN122000948ACN 122000948 ACN122000948 ACN 122000948ACN-122000948-A

Abstract

The invention discloses an electric appliance load cluster aggregation method and system for a smart grid, which relate to the technical field of electric appliance load management and are used for collecting multi-working-condition data of electric appliance loads and preprocessing, clustering the preprocessed multi-working-condition data by adopting an improved fuzzy C-means clustering algorithm, secondarily aggregating the clustered multi-working-condition data by adopting a Monte Carlo method, constructing a multi-working-condition classification model of the electric appliance loads, classifying working conditions of the secondarily aggregated multi-working-condition data, identifying and classifying the electric appliance load modes, aggregating electric appliance loads of a plurality of electric appliance areas and a plurality of buildings in a building based on classification results of the electric appliance load modes, participating in scheduling of the smart grid, and evaluating adjustable potential of the electric appliance loads. An independent model is established aiming at different working conditions, so that the load requirements of electric appliances under different environmental conditions can be more accurately met, and the self-adaptability and the energy-saving effect of the system are improved.

Inventors

SHI RUIJIE
LUO CONG
SHI HONGMING
WANG SHENG
LUO PENG
GONG ZHENG
LIN WUXING
OUYANG LILIN
LEI HONGWEI
HU SHI
LI CHUANG

Assignees

国网电商科技有限公司
远光能源互联网产业发展(横琴)有限公司

Dates

Publication Date: 20260508
Application Date: 20260204

Claims (10)

1. The utility model provides an electric appliance load cluster aggregation method facing a smart grid, which is characterized by comprising the following steps: collecting multi-working-condition data of an electrical appliance load, and preprocessing the multi-working-condition data of the electrical appliance load; Clustering the preprocessed multi-working condition data by adopting an improved fuzzy C-means clustering algorithm; Performing secondary aggregation treatment on the clustered multi-working condition data by adopting a Monte Carlo method; Constructing a multi-working condition classification model of the electrical load, classifying working conditions of the multi-working condition data after secondary aggregation treatment by adopting a support vector machine algorithm, identifying and classifying the electrical load mode; And based on the classification result of the electric appliance load modes, aggregating a plurality of electric appliance areas in the building and electric appliance loads of a plurality of buildings, participating in the dispatching of the intelligent power grid, and evaluating the adjustable potential of the electric appliance loads.
2. The smart grid-oriented appliance load cluster aggregation method of claim 1, wherein the evaluation of the adjustable potential of the appliance load comprises: And (3) establishing an electrical appliance aggregation response potential evaluation model considering multiple factors, and quantitatively analyzing the maximum response potential which can be achieved by the electrical appliance cluster through direct load control in a demand response event.
3. The smart grid-oriented appliance load cluster aggregation method of claim 1, wherein the improved fuzzy C-means clustering algorithm comprises: Optimizing a fuzzy C-means clustering algorithm by adopting a particle swarm optimization algorithm; The fuzzy C-means clustering algorithm randomly generates a clustering center in an initial iteration, samples with similar distances are classified into one type by measuring Euclidean distances among the samples, the clustering center is updated by solving the mean value of each type of samples, and the next iteration is carried out; The particle swarm optimization algorithm optimizes an initial cluster center generated by the fuzzy C-means clustering algorithm.
4. The smart grid-oriented appliance load cluster aggregation method as set forth in claim 3, wherein the fuzzy C-means clustering algorithm includes: Setting the pretreated Individual data samples Wherein Is the first The load vectors, the objective function J FCM is expressed as: ; Wherein, the Indicating the number of types into which the data sample is to be divided, ; Is a matrix of its similarity classifications, Is a sample For the following Membership degree of (3); in the formula, ; Represent the first Individual samples And the first The distance between class center points; Is the characteristic number of the sample; Is a weighting parameter; The sum of membership values of one sample to each cluster is 1, and the conditions are satisfied: ; Separately calculating samples Membership to A 1 And a cluster center : ; Is provided with For all of The class of the product, =0; 。
5. The smart grid-oriented electrical load cluster aggregation method of claim 3, wherein the particle swarm optimization algorithm obtains an optimal solution of the problem through information sharing among individuals in the swarm; in the particle swarm optimization algorithm, each particle records its own found optimal position; Updating the individual optimal position and the global optimal position according to the optimal positions found by all particles in the whole population; the algorithm stops when a predetermined maximum number of iterations is reached or a convergence condition is met.
6. The smart grid-oriented appliance load cluster aggregation method of claim 5, wherein the update formula for the speed and location of each particle is as follows: ; ; Wherein, the Is a particle At the time of Is used for the speed of the (c) in the (c), Is a particle At the time of Is provided in the position of (a), Is the weight of the inertia, which is the weight of the inertia, And Is a learning factor, and represents acceleration coefficients which are close to the optimal position of the individual and the global optimal position respectively; And Is at A random number between the two random numbers, Indicating particles Is the optimal location of the individual.
7. The smart grid-oriented electrical load cluster aggregation method of claim 1, wherein the clustering-processed multi-task data is subjected to secondary aggregation processing by adopting a monte carlo method, and the method comprises the following steps: Independently extracting samples from the distribution of each random variable, calculating the electrical load of each sub-area for each sampling sequence: Counting all simulation results to obtain estimated distribution, expected value and variance of the electrical appliance load; And calculating the electric load of the whole area according to the electric load sum of each sub-area.
8. An appliance load cluster aggregation system oriented to a smart grid, comprising: The data acquisition and preprocessing module is used for acquiring multi-working-condition data of the electrical appliance load and preprocessing the multi-working-condition data of the electrical appliance load; The clustering processing module is used for clustering the preprocessed multi-working condition data by adopting an improved fuzzy C-means clustering algorithm; the secondary aggregation processing module is used for carrying out secondary aggregation processing on the clustered multi-working condition data by adopting a Monte Carlo method; the classification module is used for constructing a multi-working condition classification model of the electrical load, classifying working conditions of the multi-working condition data after secondary aggregation treatment by adopting a support vector machine algorithm, identifying and classifying electrical load modes; And the intelligent power grid dispatching module is used for aggregating a plurality of electric appliance areas in the building and electric appliance loads of a plurality of buildings based on classification results of electric appliance load modes, participating in dispatching of the intelligent power grid and evaluating adjustable potential of the electric appliance loads.
9. The smart grid-oriented appliance load cluster aggregation system of claim 8, wherein the evaluation of the adjustable potential of appliance load comprises: And (3) establishing an electrical appliance aggregation response potential evaluation model considering multiple factors, and quantitatively analyzing the maximum response potential which can be achieved by the electrical appliance cluster through direct load control in a demand response event.
10. The smart grid-oriented appliance load cluster aggregation system of claim 8, wherein the improved fuzzy C-means clustering algorithm comprises: Optimizing a fuzzy C-means clustering algorithm by adopting a particle swarm optimization algorithm; The fuzzy C-means clustering algorithm randomly generates a clustering center in an initial iteration, samples with similar distances are classified into one type by measuring Euclidean distances among the samples, the clustering center is updated by solving the mean value of each type of samples, and the next iteration is carried out; The particle swarm optimization algorithm optimizes an initial cluster center generated by the fuzzy C-means clustering algorithm.

Description

Smart grid-oriented electrical appliance load cluster aggregation method and system Technical Field The invention relates to the technical field of power load management, in particular to an electric appliance load cluster aggregation method and system for a smart grid. Background In recent years, with the rapid development of economy and the continuous improvement of living standard of people, the electric load is continuously increased, and the electric load is particularly obvious. The peak load ratio of electric appliances in the economically developed areas such as Shanghai, jiangsu and Zhejiang is even more than 50%, and the peak load ratio shows an annual rising trend. The method has become an important demand response resource of the power system, but due to the type difference and access dispersion, the scheduling center is difficult to directly acquire the aggregate power and develop scheduling control, and the response potential exertion is limited. And stacking large-scale renewable energy sources for grid connection, so that balanced resources of the system are increasingly tensed. The intelligent power grid bidirectional communication technology and the advanced measurement system (AMI) rapidly develop to provide technical support for monitoring and controlling the load of a user side, the electric appliance load has the characteristics of strong controllability and great scheduling potential, massive electric appliance loads are aggregated through load aggregators (Load Aggregator, LA), the coordinated control participates in system adjustment, great potential is shown in aspects of peak clipping and valley filling, maintaining the stability of a power system, providing auxiliary services and the like, and the key of fully excavating and utilizing the potential is to establish an adaptive aggregation model for the electric appliance load. The existing electrical load polymerization has the following defects: 1. The traditional aggregation mode based on statistical characteristics mainly determines the general characteristics of loads, such as average loads, peak loads and the like, by statistically analyzing and sorting historical data, lacks adaptability to complex dynamic environments, cannot accurately reflect load changes under different working conditions, and therefore has limited possible effects when dealing with environments with large actual changes. 2. The general aggregation mode based on time series analysis relies on capturing the change trend of load along with time, including periodicity, seasonal and other rules, has weak response capability to sudden events, and can have errors in load change prediction under long-term steady state. Therefore, how to provide an electric appliance load cluster aggregation method and system for a smart grid, which are capable of building independent models aiming at different working conditions, more accurately adapting to electric appliance load demands under different environmental conditions, and improving the self-adaptability and energy-saving effect of the system are problems to be solved by those skilled in the art. Disclosure of Invention In view of the above, the invention provides a smart grid-oriented electrical load cluster aggregation method and system, and provides a solution based on a multi-working-condition model in order to solve the problem that building electrical loads alternately exist in multiple working conditions. The core idea of the multi-working-condition model is to cluster training data according to working condition characteristics, and then respectively establish a control model aiming at each working condition point. The effectiveness of this approach depends largely on the accuracy and efficiency of the clustering algorithm. By improving the existing clustering algorithm, different working conditions can be identified and divided more accurately, and therefore a more accurate control strategy is provided for each working condition. The self-adaptability and the energy-saving effect of the electrical system are greatly improved, and the electrical system is ensured to automatically adjust the operation mode according to the change of the environment and the demand. Through the electric appliance system after optimizing, not only can reduce the energy consumption more effectively, can also promote travelling comfort and overall efficiency. In addition, the system can respond to the requirements of the power system more flexibly, and the stability of the power grid can be improved. Therefore, by improving the clustering algorithm and combining the multi-working-condition modeling strategy, the complex building electrical load management challenges can be effectively responded, and the energy sustainable development is contributed. In order to achieve the purpose, the invention adopts the following technical scheme that the intelligent power grid-oriented electric appliance load cluster aggregation method comprise