Multi-DEPSO:a DE And PSO Based Hybrid Algorithm In Dynamic Environments

The optimization problems that we confront with in real life are often dynamic, which means that the elements of these problems are time-varying. Recently many researchers and organizations are more and more interested in the dynamic optimization problems. Many conferences have the special session on evolutionary in dynamic and uncertain environments, such as CEC and GECCO and so on. Many researchers propose new dynamic optimization algorithms to solve the dynamic optimization problems. So researching dynamic optimization algorithm is a subject that has practical significance and scientific value.An effective hybrid algorithm based on Differential Evolution (DE) and Particle Swarm Optimization (PSO) is proposed in this paper for dynamic optimization problems. The multi-population strategy is used to enhance the diversity and keeps each subpopulation on a different peak. A hybrid operator based on DE and PSO (DEPSO) is designed to find and track the optima for each subpopulation. Using the DEPSO operator, each individual is sequentially carried out DE and PSO operations. An exclusion scheme is proposed that integrates the distance based exclusion scheme with the hill-valley function to keep each subpopulation on a different peak. In this exclusion scheme, the distance of two subpopulations is measured by the distance of their best individuals. If the distance of any two subpopulations is less than the threshold, then it is considered that these two subpopulations are likely to share the same peak. In this case, the hill-valley function is used to judge whether they locate on the same peak. The algorithm is applied to MPB and the set of benchmark functions provided for the CEC2009special session on evolutionary computation in dynamic and uncertain environments including Dynamic rotation peak benchmark generator and Dynamic composition benchmark generator. Experimental results show that it is significantly better in terms of overall performance of algorithm than other state-of-the-art algorithms.