Temporal Feature Selection for Multi-Step Ahead Reheater Temperature Prediction

Abstract

Accurately predicting the reheater steam temperature over both short and medium time periods is crucial for the efficiency and safety of operations. With regard to the diverse temporal effects of influential factors, the accurate identification of delay orders allows effective temperature predictions for the reheater system. In this paper, a deep neural network (DNN) and a genetic algorithm (GA)-based optimal multi-step temporal feature selection model for reheater temperature is proposed. In the proposed model, DNN is used to establish a steam temperature predictor for future time steps, and GA is used to find the optimal delay orders, while fully considering the balance between modeling accuracy and computational complexity. The experimental results for two ultra-super-critical 1000 MW power plants show that the optimal delay orders calculated using this method achieve high forecasting accuracy and low computational overhead. Moreover, it is argued that the similarities of the two reheater experiments reflect the common physical properties of different reheaters, so the proposed algorithms could be generalized to guide temporal feature selection for other reheaters.