Optimized BP-LSTM neural network based on IGA for heat exchanger outlet temperature prediction

Abstract In this paper, an improved Genetic Algorithm (IGA) is proposed to optimize the BP-LSTM neural network for predicting heat exchanger temperature. The study introduces an enhanced genetic algorithm designed to optimize both short- and long-term memory networks (LSTM) in conjunction with the BP neural network temperature prediction model and the BP-LSTM neural network model. Simulation and comparison results demonstrate that the proposed prediction method can effectively forecast heat exchanger temperatures.