PCA-Kmeans clustering method for classified management of power quality of traction power supply system

Abstract

The invention discloses a PCA-Kmeans clustering method for power quality classification management of a traction power supply system, and the method comprises the steps: carrying out the dimension reduction of an electric energy quality evaluation index into two features through employing a PCA principal component analysis method, and carrying out the successful division evaluation of the electric energy quality of each electric power segment through a Kmeans clustering algorithm. And finally, through comparison with other electric energy quality division modes, it is verified that the method has better practicability. The method comprises the following steps: firstly, preprocessing three-phase voltage and current data output by a rail transit substation, then respectively carrying out symmetric decomposition and Fourier transform, then extracting six characteristic values for calculating and evaluating the electric energy quality, and finally carrying out dimension reduction simulation by utilizing a PCA-Kmeans clustering algorithm through the six characteristic values to obtain the electric energy quality of the rail transit substation. The reliability of the traction power supply system is graded, evaluated and verified, and the reliability of the traction power supply system is improved comprehensively to evaluate and determine the electric energy quality.

Description

technical field [0001] The invention relates to the technical field of electric energy quality evaluation of a traction power supply system, in particular to a PCA-Kmeans clustering method for classified management of electric energy quality of a traction power supply system. Background technique [0002] The power quality assessment of electrified railway is to calculate the system frequency deviation, voltage total harmonic distortion rate, voltage fluctuation and flicker, voltage unbalance and other electric energy by analyzing the basic power quality data such as voltage, current and frequency on the electrified railway system side. Quality indicators, and then judge its power quality status through the relevant technical standards of our country. [0003] However, in the actual power quality analysis, the characteristic parameters often contain a large number of irrelevant harmonics and the data dimension is high, which makes the evaluation and analysis of power quality...

AI Technical Summary

Problems solved by technology

[0003] In reality, the characteristic parameters in power quality analysis often contain a large number of irrelevant harmonics and the data dimension is high, which makes the evaluation and analysis of power quality very difficult, and its evaluation and classification will be wrong, and the effect is not good.

At present, the existing evaluation algorithms such as: matter-element analysis method, BP neural network analysis method, etc., the calculation process is complicated, the cost is high, and the result is not accurate.

Images