LAUSR

Quantum particle swarm optimization (QPSO), inspired from the basic concept of PSO algorithm and quantum theory, is a stochastic searching algorithm. However, the algorithm may encounter a premature convergence when dealing with multimodal and complex inverse problems. Thus, some improvements are introduced. More especially, one will randomly select the best particle to take part in the current search domain. Also, a mutation strategy is added to the mean best position, and an enhancement factor (EF) is incorporated to enhance the global search capability to find the global optimum solution and to avoid premature convergence. Moreover, some parameter updating strategy is proposed to tradeoff the exploration and exploitation searches. Experiments have been conducted on well-known multimodal functions and an inverse problem. The numerical results showcase the merit and efficiency of the proposed modified quantum inspired particle swarm optimizer (MQPSO).