Adaptive Non-uniform Distribution of Quantum Particles in mQSO

This paper studies properties of quantum particles rules of movement in particle swarm optimization (PSO) for non-stationary optimization tasks. A multi-swarm approach based on two types of particles: neutral and quantum ones is a framework of the experimental research. A new method of generation of new location candidates for quantum particles is proposed. Then a set of experiments is performed where this method is verified. The test-cases represent different situations which can occur in the search process concerning different numbers of moving peaks respectively to the number of sub-swarms. To obtain the requested circumstances in the testing environment the number of sub-swarms is fixed. The results show high efficiency and robustness of the proposed method in all of the tested variants.