LS-SVM based on Chaotic Particle Swarm Optimization with simulated annealing and application

In this paper, a novel Chaotic Particle Swarm Optimization (CPSO) with simulated annealing algorithm (SACPSO) based scheme is proposed to choose the parameters of LS-SVM automatically. CPSO adopts chaotic mapping with certainty, ergodicity, and the stochastic property, possessing high search efficiency. SA algorithm employs certain probability to improve the ability of PSO to escape from a local optimum and has fast convergence and high computational precision. The hybrid algorithm is applied to a turbine heat rate modeling. The simulation results have shown that the performance of the hybrid algorithm is better than of the Particle Swarm Optimization (PSO), and the hybrid algorithm is effective and feasible for solving the problem of predicting heat rate.