Abstract:
Based on industrial safety theory and related literature research, this paper establishes an industrial safety evaluation index system, uses entropy Weight-Grey correlation analysis to evaluate industrial safety from 2000 to 2018, and builds an LSTM neural network prediction model to predict the data of various evaluation indicators of industrial safety from 2019 to 2023. It combines all prediction data with historical data to establish an early warning model of industrial safety based on one-dimensional convolutional neural network, and to provide a systematic early warning on the industrial safety situation from 2019 to 2023. The results show that with the continuous industrial reforms, the safety rate has kept growing; so it predicts that the overall safety of industry in the next five years will be secure.