Lightning trip type identification method

Abstract

The invention discloses a lightning trip type identification method, which comprises the following steps: extracting lightning trip transient voltage signal feature image data, inputting the feature image data into a deep residual neural network model, and outputting an identification result by the deep residual neural network model. The step of extracting the feature image data of the lightning trip transient voltage signal comprises the steps of performing multi-scale generalized S transformation on the extracted one-dimensional time domain voltage signal, drawing time-frequency distribution images of different-scale generalized S transformation, and taking the time-frequency distribution image data as the feature image data of the lightning trip transient voltage signal. According to the method, the one-dimensional time domain voltage signal is subjected to multi-scale generalized S transformation, so that the extraction range of lightning trip transient voltage signal feature image data is expanded, the problem that transient feature quantities are not sufficiently utilized by a one-dimensional signal analysis method in neural network discrimination is solved, and the accuracy of lightning trip type discrimination is improved.

Description

technical field [0001] The invention belongs to the technical field of lightning intelligent protection, and in particular relates to a lightning trip type identification method. Background technique [0002] The power industry is an important part of the national economy, and the transmission line is the main artery of the power industry and an important part of the power system. Transmission lines, especially EHV and UHV transmission lines, have long lengths, wide coverage, complex natural and geographical conditions, and are easily affected by various natural disasters. The reliability of their operation plays a vital role in the safe and stable operation of the entire power system. effect. [0003] Lightning damage is one of the most common natural disasters faced by transmission lines, and lightning strikes have become one of the most important factors that cause transmission line failures, trips, and affect the safe operation of power grids. With the development of E...

AI Technical Summary

Problems solved by technology

[0006] (1) The existing transient signal processing and analysis of lightning tripping mainly use analysis methods such as wavelet decomposition, empirical mode decomposition, and Hilbert change. Make full use of rich feature quantities;

[0007] (2) The characteristic distribution range of lightning tripping transient signal in time domain and frequency domain is relatively large, and a single time domain signal analysis method can only have a good analysis effect on signals in a certain range of time domain or frequency domain. The range is narrow, and in practical applications, there will be cases where the recognition accuracy of some lightning tripping transient signals is low;

[0008] (3) Existing studies mostly use ATP-EMTP or PSCAD simulation software to simulate the lightning tripping transient signal, but there is a big difference between the simulated signal and the actually monitored signal. If the classifier is trained using More simulation signals will cause the identification to fall into a local optimal solution, and the real lightning tripping transient signal cannot be accurately identified; but if the simulation signal is not used, the currently measured lightning tripping transient signal obtained is less, and oversimulation will occur. The combination phenomenon leads to poor training results of existing classifiers

Images