A Robust Evolutionary Algorithm for Large Scale Optimization

In this paper we present a new cooperative coevolution hybrid Taguchi-genetic algorithm (CCHTGA) to solve high-dimensional optimization problems with continuous variables which are typical in large scale systems. The CCHTGA employs cooperative coevolution frame to decompose the problem into several subsystems and adopts a novel decomposition technique “random grouping strategy”. Subsystem optimizer is an improved hybrid Taguchi-genetic algorithm (HTGA) which is effective, more robust and rapidly convergent. The performance of the CCHTGA compared to a state-of-the-art cooperative coevolution particle swarm algorithm (CCPSO2). The simulation results obtained from the application of this method to function optimization indicate that CCHTGA can be a competitive algorithm to achieve better and robust results.