深層学習を用いた動的機器モニタリング信号による予知保全
概要
A time-series data future prediction algorithm using Long-Short Term Memory (LSTM), which is a kind of Regression Neural Network (RNN), has been proposed for the purpose of detecting early abnormality of monitoring signals of dynamic devices. An improvement for applying this algorithm to actual dynamic equipment monitoring signals was proposed, and an anomaly judgment method was also proposed.
