Electric energy quality steady-state index prediction and early warning method based on LSTM neural network

Abstract

The invention relates to an electric energy quality steady-state index prediction and early warning method based on an LSTM neural network, and the method specifically comprises the following steps: S1, obtaining electric energy quality steady-state index historical data of an electric energy quality monitoring point, carrying out the standardization processing, and establishing an LSTM neural network data model; S2, inputting the standardized historical data of the power quality steady-state index as a training sample into a neural network data model for training to obtain a final neural network model parameter; S3, acquiring real-time power quality steady-state index data, and inputting the data into the neural network model to obtain power quality steady-state index prediction data; S4, comparing with a related threshold value set by an early warning level to obtain early warning information, storing the early warning information in an early warning information table, and sending the early warning information to the mobile terminal; and S5, judging whether the interval time reaches a data updating threshold value, and if yes, turning to S1 for next data updating. Compared with the prior art, the method has the advantages of high-precision prediction and accurate early warning of the steady-state indexes of the electric energy quality and the like.

Description

technical field [0001] The invention relates to the technical field of power quality monitoring and early warning, in particular to a method for predicting and early warning of power quality steady-state indicators based on LSTM neural network. Background technique [0002] With the rapid development of the national economy and modern industry, the loads in the power system are becoming increasingly complex and diverse. A large number of applications of high-power nonlinear, impact and fluctuating special loads in power grids such as new energy, electrified railways, and metallurgy have caused power quality problems such as harmonics, three-phase imbalance, voltage fluctuations, and flicker in the power grid, which have seriously affected the Power quality of grid and users. At present, a power quality monitoring system for large users including electrified railways and new energy has been built in the power grid system, which contains multiple power quality monitoring poin...

AI Technical Summary

Problems solved by technology

[0003] In order to mine useful information in the steady-state index data of power quality monitoring points, more and more algorithms are applied to the prediction of power quality steady-state index data, such as ARTXP algorithm, ARIMA time series algorithm, improved KPCA and GA-BP neural network The above algorithms play a certain role in the prediction of power quality steady-state index data, but it is difficult to provide high prediction accuracy for complex and changeable power quality problems and the data processing of a large number of monitoring points for a long time

Images