An improved grey wolf optimizer(MGWO) is used to optimize BiLSTM to predict water wall temperature. The improved algorithm adopts nonlinear factor adjustment strategy

adaptive position update strategy and dynamic weight modification strategy to improve the global optimization ability of the GWO. The improved grey wolf optimizer is used to optimize the number of hidden layers

learning rate and regularization parameters of the BiLSTM model to improve the prediction accuracy of the model. The data of a power plant in Xinjiang are used for prediction simulation. The results show that

the improved optimizer has higher prediction accuracy

and can predict the change trend of wall temperature when the unit is lifting and lowering load. Compared with the LSTM and BiLSTM models

the average root mean square error of the model reduces by 9.86% and 3.69%

respectively

and the overtemperature of water wall temperature can be predicted in advance

which is of great significance for the prevention of overtemperature of water wall.