An Improved QPSO Algorithm based on Student’s t-Distribution

Inspired by the classical PSO algorithm and quantum mechanics theories, this paper presents a novel QPSO algorithm with student’s t-distribution strategy as mutation operator (HTDQ-PSO). The quantum behavior theory is introduced to change the updating mode of the particles, while the mutation strategy is introduced to improve the population diversity and enhance the global search ability. The incorporation of dynamic variation scale into QPSO algorithm not only enhances the particle’ s ability of jumping out of local optima, but also increases the convergence rate, thus the performance of QPSO can be improved in preventing premature convergence and increasing solution precision. For validation, five high- dimensional complex nonlinear benchmark functions, including two unimodal and three multimodal, are used to compare the proposed algorithm with three other PSO variants. The simulation results show that the proposed HTDQ-PSO algorithm is superior to the other algorithms with a better astringency and stability.