Parameter Study on Improved Quantum-behaved Particle Swarm Optimizations
Abstract
Inspired by concepts in quantum mechanics and particle swarm optimization (PSO) algorithm, quantum-behaved particle swarm optimization (QPSO) algorithm was proposed as a variant of PSO algorithm with better global search ability. At the same time, some improved QPSO algorithms are also presented. In order to determine whether the performance of the algorithm is affected by the location of the parameter, this paper compares four variants of QPSO algorithm. The operator is exerted on the mean best position and the particle’s previous position to improve the search ability of the QPSO algorithm, respectively. Finally, some empirical studies on popular benchmark functions are performed in order to make a full performance evaluation and comparison among four variants of QPSO algorithms. The experimental results show that the new parameter based on individual particles evolutionary process which located in the mean best position algorithm (IEQPSO-1) is more effective approach than others in most cases.
Keywords
Quantum-behaved particle swarm optimization, Swarm intelligence, Optimization algorithms
DOI
10.12783/dtetr/amsm2017/14812
10.12783/dtetr/amsm2017/14812
Refbacks
- There are currently no refbacks.