![]() | Only 14 pages are availabe for public view |
Abstract Genetic algorithm (GA) is a branch of the so-called evolutionary computing (EC) that is inspired by the evolution of living beings in nature. GA is considered a powerful tool for solving many optimization problems.The searching ability of GA depends on different parameters such as: population size, number of generations, crossover operator and mutation operator.The proper selection of such parameters received much attention of the researchers in order to achieve fast convergence to near optimal solution and to avoid stucking into local minima.In this thesis, we propose a new method for improving the search mechanism of GA.The proposed method depends on adopting different conditional crossover operators based on Local Partitioning of the population during the selection process.The proposed method is evaluated in the experimental work through twelve benchmark optimization problems and compared with conventional real-valued coded GA.The results ensure the efficacy and competitivity of the proposed method |