Predictive control of SCR denitrification system in thermal power plants based on GA‐BP and PSO

Abstract Aiming at the characteristics of strong non‐linearity and large inertia in the reaction process of a selective catalytic reduction (SCR) denitrification system, a predictive control algorithm based on a back propagation neural network optimized by genetic algorithm (GA‐BP) and particle swarm optimization (PSO) is proposed. First, the prediction model of a SCR denitrification system is established by GA‐BP. Second, output feedback and bias correction are used to reduce the prediction error. Third, the optimal inlet ammonia concentration is obtained by PSO. At the same time, in order to solve the problems of high dimension, large noise, and strong coupling in the original data of the SCR system in the process of establishing the prediction model, the least absolute shrinkage selection operator (LASSO) algorithm and the local outlier factor (LOF) detection algorithm are used to screen important variables and samples in the original data set of the SCR system to remove redundant variables and outliers. Finally, the simulation results show that the prediction model has good prediction accuracy and that the proposed predictive control method can achieve accurate control of ammonia injection concentration. This method improves the denitrification efficiency and reduces the NO x emission concentration, which can provide good guidance for on‐site production.