A Modified Cultural Algorithm based on Genetic Algorithm for Solving Global Optimization Problems

This paper introduce one innovations into the world of multimodal function optimization: a new Cultural Algorithm (CA) based on Genetic Algorithm (GA) with a new adaptive mutation (AM-CGA). The idea is to use knowledge about local optima found during the search with other techniques utilized for escape from local minimum in multimodal functions. The search knowledge was maintained using a Cultural Algorithm structure-, which is updated by behaviors of individuals and is used to actively guide the search. The test results show that the algorithm can get comparable or superior results to that of some current well-known unconstrained numerical optimization. The results also point to the potential of introducing of techniques such as Hill Climbing and Simulated Annealing for improving the search space of algorithm. Simulation results on benchmark indicate that AM-CGA improved the benchmark results.