Based on a prediction method for improving WOA-Elman air quality prediction

Abstract

Aiming at the problem that the Elman neural network is easy to fall into the local optimal solution when predicting air quality indicators, the prediction accuracy is low. A prediction model combining the PCA of meteorological factors and the improved whale optimization algorithm IWOA Elman neural network is proposed. Use PCA to extract the main components that affect the air quality index as the input variables of the Elman neural network, use the initial population optimization and the introduction of inertial weights to optimize WOA, enhance the global search ability and convergence speed, and then proceed to get the weight and value of the Elman neural network and optimize the threshold. The results show that the prediction error of this model is better than the single Elman model, PCA-Elman model, IWOA-Elman model and BP model. The model is based on Chongqing air quality data and meteorological data for experiments, which provides a realistic reference for air quality index research.