Prediction of Temperature Field in an Improved CS-LSTM Ethylene Cracking Furnace
In the process of ethylene production,it is necessary to improve the traditional temper-ature measurement method in response to the difficulty of monitoring the temperature of the ethylene cracking furnace tube.By optimizing the operation under the data model,the outlet temperature of the ethylene cracking furnace can be effectively predicted.When temperature fluctuations occur,intervention can be taken to improve product efficiency and production safe-ty.This article applies the improved cuckoo bird algorithm optimized LSTM(CS-LSTM)to real industrial data and compares it with four models.The simulation results show that using Atten-tion CS LSTM significantly improves the prediction accuracy and has good steady-state accura-cy,with a temperature prediction accuracy of 95％.
万方数据
维普
