The Research And Applications Of Hybrid Evolutionary Computation

Evolutionary Algorithms(EAs) are a cluster of statistical optimizationtechnologies inspired by nature. They have the feature of self-organization, self-adaptive and self-learning. With the internal parallel mechanism, EAs isspecial suitable for large complex problem. There exists many branches in EAfields such as genetic Algorithm(GA), Gene Expression Programming(GEP), ParticleSwarm Optimization(PSO), Quantum Inspired Evolutionary Algorithms(QEA), Estimation of Distribution Algorithm(EDA). Every one of these algorithmshas its only advantages over others. In this paper we study the combinationof these algorithms and focus on the EDA, QEA and GuoTao algorithms. Themajor contributions and creative work is as follows:1. Propose the relax complemental concept and apply it into EDA.2. Present a hybrid algorithm and call it Quantum-inspired Estimation ofDistribution Algorithm(QEDA)3. Combine the linear programming with EDA for multi knapsack problem.4. Improved multi-parent crossover with chaos optimization in Guotao algo-rithm.The work of this paper enriches the content of EAs. The results of EDAand Guotao algorithm obtained here have certain theoretical and applicationvalues.