Prediction Based on CA-GEP algorithm predicted rainfall
Sicong Huo, Chuyan Deng, Tao Lü, Xueping Zhu · 2021 IEEE 3rd International Conference on Frontiers Technology of Information and Computer (ICFTIC) · 2021
The traditional Gene Expression Programming (GEP) algorithm has the disadvantages of weak global search ability and difficult to jump out of local optimization. In order to solve these shortcomings, this paper proposes Chromosome Adaptive-Gene Expression Programming (CA-GEP) algorithm, which appropriately increases the mutation rate of the algorithm in the iterative process to enhance the population diversity, makes use of the characteristics of population diversity to jump out of the local optimization and improve the global search ability. While ensuring better population diversity, it also increases the crossover rate and reduces the mutation rate to speed up the convergence of the algorithm, so as to improve the convergence efficiency of the algorithm. According to the concentration and dispersion of individual fitness values in the population, CA-GEP algorithm adaptively adjusts the crossover rate and mutation rate of each iteration through fuzzy control, so as to make the iterative process of the algorithm more efficient. The experimental results of rainfall prediction show that CA-GEP algorithm has significantly improved the stability, global convergence ability and optimization speed, compared with BP algorithm and SVM algorithm.