An improved dynamic weighted particle swarm optimization (IDW-PSO) for continuous optimization problem

In this Study we found that Learning from the natural phenomenon of the social animal is the best way to learn and getting a best mechanism for adapting the dynamic nature of the environment. With the gain foothold of the research in the optimization, many methods employee to solve complex or NP-Hard problems like stuck in local optima that known as stagnation problem. Nature always act as a source of inspiration, source of generating new concept, mechanism, principles for creating artificial system for solving various of complex computation problems. In this work, proposed a variant of particle swarm optimization which is also performing better on larger search space. Hence, the proposed variant overcome the issues of ordinary particle swarm optimization (PSO) like performance goes poor for a larger search space on multimodal function environment and face the problem of stagnation. Improvement of the proposed variant is based on social nature of birds and obtained results compare with classical particle swarm optimization and latest swarm based optimization algorithm named as firefly algorithm. Furthermore, nine standard benchmark functions (both unimodal and multimodal) are used to evaluate the performance of proposed approach and compare it with other comparable algorithm based on average and standard deviation as the parameters. Experimental results show that the proposed IDW-PSO algorithm outperform the classical PSO and FA.