Power quality disturbance classification method based on sparse automatic coding depth neural network

Abstract

The invention discloses a power quality disturbance classification method based on a sparse automatic coded depth neural network. The method comprises the following steps of preprocessing a power quality disturbance original data set, inputting a preprocessed data sample into a stacked sparse automatic encoder to carry out automatic feature learning to obtain a feature expression, a step of transmitting an extracted feature expression to a softmax classifier for fine adjustment and further adjusting network parameters to obtain a trained classification model, and a step of inputting the a testsample into the classification model for power quality disturbance classification. According to the method, nine power quality disturbance signals including two kinds of composite disturbances can beaccurately identified, and the method has good robustness.

Description

technical field [0001] The invention relates to a method for classifying power quality disturbances, in particular to a method for classifying power quality disturbances based on a sparse automatic encoding deep neural network. Background technique [0002] In recent years, with the rapid development of smart grids and the connection of a large number of new energy power generation devices and impact, nonlinear, and fluctuating loads to the grid, power quality problems in power systems have become increasingly prominent. Various power quality disturbance events have serious impact on industrial production and residents. Lives have had a very serious impact. On the other hand, with the continuous development of science and technology, more and more electrical equipment put forward higher requirements for power quality, and the accurate identification and effective classification of power quality disturbances can provide assistance for the management and governance of power qu...

AI Technical Summary

Problems solved by technology

ANN has a simple structure, strong problem-solving ability, and can handle noisy data well, but the training time is long and the accuracy needs to be improved

[0004] The above feature extraction methods have achieved good results in the identification of power quality disturbances, but the implementation of these methods requires relevant prior knowledge, and the process of determining the implementation details usually requires repeated modifications and adjustments, which is time-consuming and difficult to obtain. The Deep Essential Characteristics of Fault Data

Images