Longitudinal time axis clustering method in generalized load modeling on basis of seasonality

Abstract

The invention discloses a longitudinal time axis clustering method in generalized load modeling on the basis of seasonality. According to the method, root bus data formed by wind power and loads of the whole year are obtained; transverse time axis pre-clustering is carried out, single day time re-clustering is carried out, longitudinal time unit clustering is carried out, longitudinal time axis clustering results of the whole year are output, a longitudinal time axis clustering method is utilized for analyzing the data of the whole year, and categorical data considering the seasonality are obtained for accurate modeling. An AP algorithm and a longitudinal time axis clustering strategy considering the seasonality are utilized, the large sample data actually measured can be partitioned reasonably, and the simulation result shows that compared with a traditional modeling method, generalized load modeling carried out after clustering analysis makes a model practical on the basis of meeting the requirement for accuracy, and is beneficial for improving the simulation accuracy and the simulation effectiveness of an electrical power system.

Description

technical field [0001] The invention relates to a seasonality-based longitudinal time axis clustering method in generalized load modeling. Background technique [0002] As an intermittent energy source, wind power has a great impact on the security and stability of the power system due to its randomness and volatility, and also brings great challenges to generalized load modeling. With the increase of wind power capacity, the generalized load node sometimes presents power source characteristics and sometimes load characteristics. Different characteristics correspond to different models, which will produce qualitative changes in power system simulation calculation. Power system analysis is very important. [0003] In measurement-based load modeling, time-variation is the biggest obstacle to its application. However, the uncertainty brought about by wind power access increases the difficulty of the original load modeling problem. Research shows that classification and synthe...

AI Technical Summary

Problems solved by technology

The above-mentioned literature can solve the time-varying problem well through a reasonable classification and synthesis method, but the clustering method used in some literature needs to artificially set the number of clusters, cluster centers, etc., which cannot exclude subjective factors, and is not universal in the new situation. Applicability; In addition, due to the small sample size of the analysis object, the clustering strategy is relatively simple. It is only necessary to determine the appropriate clustering method and feature vector for clustering to be divided into categories with clear boundaries. In the face of the whole year (or longer time ) for large sample data of wind power and load, the simple clustering strategy cannot reasonably classify all sample data

Images