According to the imbalance between exploration and exploitation, susceptibility to local optima, and low search efficiency of metaheuristic algorithms, a hybrid slime mould genetic algorithm based on spatial attenuation self-diffusion mechanism if presented. The algorithm uses genetic algorithm as the basic structure, and guides individuals to search in the solution space by recombining features through three operations: selection, crossover, and mutation. Firstly, it introduces oscillation-contraction mechanism with characteristics of both positive-negative feedback and random walking as crossover operators to enhance both global and local search capabilities. Secondly, a self-diffusion mechanism based on spatial decay is proposed as a mutation operator. This mechanism guides the diffusion motion using a spatial scale which decreases over the algorithm's lifecycle, promoting diversity in the early stages and effective exploration of neighborhood information in the later stages. Finally, a discriminative control strategy is introduced to adaptively adjust the algorithm's parameters based on the distribution deviation of the population fitness. This strategy helps balance the exploration and exploitation capabilities of the algorithm. To validate the algorithm's performance, experiments are conducted on two publicly available benchmark test sets: IEEE CEC2017 and IEEE CEC2021. The results demonstrate that the proposed algorithm effectively balances exploration and exploitation capabilities and exhibits superior optimization performance compared with other 23 different types of algorithms.

Most of the previous PM_2.5 prediction models present unsatisfactory performance in several aspects, including predicting accuracy and generalization ability, especially in case of the sudden change in the value of PM_2.5 situation. Therefore, we propose a method based on the U-net neural network to predict the hourly PM_2.5 concentration value on the research area, attempting to improve the prediction performance. The proposed model includes two major steps. First, based on the inverse distance interpolation of historical wind field data, discrete station PM_2.5 values are interpolated into a PM_2.5 grid map; second, the U-net neural network is applied to train the prepared spatiotemporal grid data and make predictions. The model can use the PM_2.5 concentration values of the grid map extracted at different time stamps for the PM_2.5 prediction. The PM_2.5 concentration values at all locations in the research region can be achieved. Specifically, the prediction accuracy and the generalization ability of the model in case of sudden changes are revealed. Experimental results indicate that the proposed method has a 10% improvement in the prediction accuracy of PM_2.5 concentration values in the case of sudden change.