Research on Air Quality Prediction Based on the Hybrid Algorithm of GWO and LSSVM
The study proposes an air quality prediction model based on Grey Wolf Optimizer(GWO)optimizing Least Squares Support Vector Machine(LSSVM),aiming at the problem that air quality is influenced by multiple factors and the prediction is difficult.This model uses the GWO to iteratively optimize the kernel function and penalty factor of the LSSVM,reduce the blindness of parameter selection and improve the prediction accuracy.Meanwhile,to verify the superiority of this model,its prediction results are compared with those of LSSVM,Genetic Algorithm optimizing LSSVM and BP Neural Network.The results show that:① Compared with traditional single prediction models,the goodness of fit of the hybrid prediction model is better.② Among the hybrid models,the goodness of fit of GWO optimizing LSSVM is higher than that of Genetic Algorithm optimizing LSSVM.Therefore,GWO optimizing LSSVM has practical application value for the prediction of air quality index.
ZhengzhouAir qualityGrey Wolf AlgorithmLeast Squares Support Vector Machine
万方数据
维普
