Parallel Multi-memetic Global Optimization Algorithm for Optimal Control of Polyarylenephthalide’s Thermally-Stimulated Luminescence

This paper presents a modification of the parallel multi-memetic global optimization algorithm based on the Mind Evolutionary Computation algorithm which is designed for loosely coupled computing systems. The algorithm implies a two-level adaptation strategy based on the proposed landscape analysis procedure and utilization of multi-memes. It is also consistent with the architecture of loosely coupled computing systems due to the new static load balancing procedure that allows to allocate more computational resources for promising search domain’s sub-areas while maintaining approximately equal load of computational nodes. The new algorithm and its software implementation were utilized to solve a computationally expensive optimal control problem for a model of chemical reaction’s dynamic for thermally-stimulated luminescence of polyarylenephtalides. Results of the numerical experiments are presented in this paper.