Statistical analysis and deep learning-based power equipment state trend prediction method

Abstract

The invention relates to a statistical analysis and deep learning-based power equipment state trend prediction method. The method comprises the steps of obtaining a power equipment state short-term trend TARIMA through an ARIMA algorithm; obtaining a power equipment state long-term trend TLSTM through an LSTM algorithm; and according to the short-term trend TARIMA and the long-term trend TLSTM ofthe power equipment, obtaining a state prediction trend T of the power equipment by adopting an exponential weighted average method, and then maintaining the state prediction trend T of the power equipment according to the state prediction trend T of the power equipment. According to the method, the state trend of the power equipment is predicted; personalized state evaluation of the power equipment, rapid detection of abnormal states, accurate prediction of state changes and intelligent diagnosis of faults can be realized, the health state of the power equipment can be comprehensively, timelyand accurately mastered, and an auxiliary decision basis is provided for intelligent operation inspection of the equipment and optimal operation of a power grid.

Description

technical field [0001] The invention belongs to the technical field of power equipment state control, and in particular relates to a method for predicting the state trend of power equipment based on statistical analysis and deep learning. Background technique [0002] The traditional state assessment mainly adopts the physical model of causality based on theoretical analysis, calculation simulation and experimental testing. However, there are many factors affecting equipment failure and the mechanism is complicated, so it is difficult to establish a perfect and accurate physical model for state assessment; and the existing methods mainly Based on a single or a small number of state parameters and unified diagnostic criteria, the determination of parameters and thresholds is mainly based on statistical analysis and subjective experience of a large number of experimental data. The analysis results are one-sided and cannot fully reflect the objective laws between fault evolution...

AI Technical Summary

Problems solved by technology

[0002] The traditional state assessment mainly adopts the physical model of causality based on theoretical analysis, calculation simulation and experimental testing. However, there are many factors affecting equipment failure and the mechanism is complicated, so it is difficult to establish a perfect and accurate physical model for state assessment; and the existing methods mainly Based on a single or a small number of state parameters and u

Images