Model for Predicting NO_x Emission from Boilers Based on MWOA-LSSVM Integration

Abstract

Accurate, reliable prediction of NO_x emission in flue gas is of great significance for operation of power station boilers with low nitrogen emissions. To improve the accuracy of a prediction model, a method for predicting NO_x emission from boilers based on integration of the whale optimization algorithm and least squares support vector machine (MWOA-LSSVM) is proposed in this paper. First, the sample space is divided, and a segmentation logistic chaotic map is then used to initialize the population. The nonlinear adaptive parameters are improved, and quadratic interpolation update position is used to improve the whale optimization algorithm (WOA) by broadening the global exploration ability of the algorithm. The MWOA is used to globally optimize the kernel function width and penalty factor of the LSSVM sub-model in each subspace, yielding the sub-model as an output. Finally, the sub-model output is integrated using the least squares regression, yielding the output from the integrated model. The simulation results show that the MWOA-LSSVM integrated model has stable, high-precision simulation performance compared with other selected prediction models and can provide more accurate predictions of NO_x emissions from boilers.