Title :
The Thermodynamic Particle Swarm Optimizer
Author :
Wu Yu ; Li Yuanxiang ; Xu Xing ; Peng Shen
Author_Institution :
Key Lab. of Software Eng., Wuhan Univ., Wuhan
Abstract :
This paper has presented a novel optimization algorithm - thermodynamic particle swarm optimizers (TDPSO). It combines the simplified evolutionary equation and the thermodynamically strategy.The simplified equation without the velocity variable has drastically reduced computation costs to achieve faster convergence. Inspired by the free energy principle of the thermo-dynamical theoretics, TDPSO algorithm has defined the rating-based entropy (RE)concept and a component thermodynamic replacement(CTR) rule. These definitions are applied to control the optimal process and to achieve the potential of finding a global optimum. Compared with other improved PSO techniques, the experimental results describe how-to make the TDPSO benefit from the thermodynamics.
Keywords :
entropy; evolutionary computation; particle swarm optimisation; thermodynamics; TDPSO; component thermodynamic replacement; evolutionary equation; novel optimization algorithm; rating-based entropy; thermo-dynamical theoretics; thermodynamic particle swarm optimizer; Clustering algorithms; Computer science; Equations; Optimal control; Particle swarm optimization; Process control; Simulated annealing; Software engineering; Testing; Thermodynamics;
Conference_Titel :
Computer Science and Software Engineering, 2008 International Conference on
Conference_Location :
Wuhan, Hubei
Print_ISBN :
978-0-7695-3336-0
DOI :
10.1109/CSSE.2008.1248