A temperature prediction method of electric power equipment based on depth neural network

Abstract

The invention discloses a temperature prediction method of electric power equipment based on a depth neural network. The method comprises the following steps: firstly, collecting the sample data of field parameters and the corresponding temperature data of actual equipment after the time period t, transmitting the data to the background monitoring system, preprocessing the data, and training the temperature prediction model through the depth neural network; then, the field parameter data are collected regularly to predict the temperature of the equipment after the time period t, and the equipment with high temperature is warned at different levels. Finally, in the routine patrol inspection, the electric power patrol robot collects the actual equipment temperature data, compares the equipment temperature data with the predicted equipment temperature data, and checks whether the temperature prediction model can adapt to the latest equipment state. The invention effectively prevents the high-temperature situation of the equipment, reserves the emergency time for the emergency repair work of the substation, and improves the working safety of the substation.

Description

technical field [0001] The invention relates to the technical field of power equipment fault monitoring, in particular to a method for predicting the temperature of power equipment based on a deep neural network. Background technique [0002] With the progress of society, the development of all walks of life is inseparable from the safe operation of the power system, so the position of the power industry in production and life is very important. Due to the problems of high labor intensity, harsh working environment, and low work efficiency in manual inspections, electric power inspection robots emerged as the times require, which can realize the unattended mode of substations. Due to the complex structure of power equipment, it is easy to fail during operation. Therefore, in order to truly achieve unattended mode, it is necessary to predict and analyze equipment failures. Since most substation accidents are caused by equipment fires, there is usually a process of continuous...

AI Technical Summary

Problems solved by technology

[0003] The traditional temperature prediction analysis mainly uses the temperature data of some equipment accumulated in the past, and uses the linear regression method to perform curve fitting to obtain a number of temperature curves that can reflect the rate of change of temperature, but the flexibility is poor, and environmental factors and equipment are not considered. The influence of factors on the surface temperature of the equipment, especially in some outdoor substations, the environmental factors are easy to change and affect the temperature of the equipment, and the traditional temperature monitoring method generally requires the staff to use the infrared sensor to obtain the infrared temperature information of the equipment, the workload is heavy and complicated

Images