Partial discharge mode recognition method of deep super learning machine in combination with evidence discount

Abstract

The invention discloses a partial discharge mode identification method of a deep super learning machine (DSL) in combination with evidence discount. The method mainly comprises the following steps: (S1) signal acquisition: collecting a partial discharge signal through a sensor; (S2) data preprocessing: carrying out denoising and other processing on the original signal; (S3) carrying out primary diagnosis on the partial discharge signal by the deep super learning machine, and constructing basic probability assignment (BPA) of a plurality of evidence bodies; (S4) evidence discounting: performingevidence discounting on the BPA of each evidence body constructed in the step (S3); (S5) constructing a new feature: adding the evidence discounted in the step (S4) to original feature data, and sending the original feature data as a new feature vector to a deep super learning machine for next iteration; and (S6) repeating the step (S3) and the step (S5) until the loss function is not reduced anymore, and obtaining a final pattern recognition result output by the deep super learning machine. Compared with the prior art, the method has the advantages that the identification accuracy is higher; and the method is high in result interpretability, and has a wide market prospect and application value.

Description

technical field [0001] The invention relates to the technical field of partial discharge pattern recognition in high voltage, specifically a partial discharge diagnosis method combined with an evidence discount synthesis method and a deep super learning machine, which belongs to a core technology in the diagnosis process of electrical equipment insulation defects. A Partial Discharge Pattern Recognition Approach with Deep Super Learning Machines Incorporating Evidence Discounting. Background technique [0002] As an important part of the power system, the insulation status of electrical equipment is closely related to the safe and stable operation of the power grid. There are many reasons for electrical equipment to generate partial discharge, and the external manifestations are also different. Different types of insulation defects have very different destructive effects on electrical equipment. Therefore, it is very important to accurately identify and classify the discharg...

AI Technical Summary

Problems solved by technology

The deep neural network can dig deep into the hidden representation of data, describe the characteristics of data, and solve the problems of high dimensionality and complexity that are difficult to deal with by traditional pattern recognition methods.

But deep neural networks also have disadvantages, including a large number of hyperparameters, poor interpretability of results, etc.

Moreover, different deep learning algorithms have their own advantages and disadvantages. When a single deep learning method is used to identify and classify partial discharge signals, it is difficult to make a comprehensive diagnosis, and there will be cases where the recognition effect is not good for a certain type of discharge or the overall recognition. Will face the problem of algorithm comparison and algorithm selection

Images