Multi-modal Function Optimization of Immune Clone Based on Uniform Design

In order to obtain all the optimal solutions of multimodal functions as much as possible, a multi peak function optimization based on uniform design is proposed. The algorithm adopts uniform design to initialize the population and ensure the uniformity and diversity of the initial antibody population distribution. Larmark learning strategy is used to search the local population in order to enhance the convergence speed and precision of the algorithm. In the immune clonal parameter setting, the parameter setting problem is described as a multi factor and multi level uniform design problem. Experimental results show that the algorithm is better in finding the best.