A knowledge-guided and manual intervention-based gene expression programming for PM2.5 concentration prediction

In view of the lack interpretation and inability to know occurrence mechanism PM2.5 concentration by deep learning algorithm in solving prediction problem, this paper adopts a knowledge-guided manual intervention-based gene expression programming (KMGEP) solve it. The KMGEP not only has strong model ability, but also can obtain explicit function relationship between its influencing factors. process implementation, knowledge guidance intervention are introduced GEP for predicting so as improve global optimization ability convergence speed. paper, daily winter (from December February) Xi’an region is taken an example, compared with artificial neural network back propagation (BP-ANN) convolutional long short-term memory combined (CNN-LSTM). Experimental results show that high accuracy prediction, obtained reveal