Filter Simulated Annealing (FSA) Method

The FSA method is a simulated-annealing-based method proposed to deal with the constrained global optimization problem. The considered problem is reformulated so as to take the form of optimizing two functions; the objective function and the constraint violation function. Then, the FSA method is applied to solve the reformulated problem. The FSA method invokes a multi-start diversification scheme in order to achieve an efficient exploration process. To deal with the considered problem, a filter-set-based procedure is built in the FSA structure. Finally, an intensification scheme is applied as a final stage of the proposed method in order to overcome the slow convergence of SA-based methods.

Source Reference: A. Hedar and M. Fukushima, Derivative-free filter simulated annealing method for constrained continuous global optimization, Journal of Global Optimization, to appear. [pdf]

Source Codes: MATLAB codes for FSA are downloadable here FSA_ver_0.1.