Hybrid genetic algorithm based on quantum computing for numerical optimization and parameter estimation

Quantum computing is applied to genetic algorithm (GA) to develop a class of quantum-inspired genetic algorithm (QGA) characterized by certain principles of quantum mechanisms for numerical optimization. Furthermore, a framework of hybrid QGA, named RQGA, is proposed by reasonably combining the Q-bit search of quantum algorithm in micro-space and classic genetic search of real-coded GA (RGA) in macro-space to achieve better optimization performances. Simulation results based on typical functions demonstrate the effectiveness of the hybridization, especially the superiority of RQGA in terms of optimization quality, efficiency as well as the robustness on parameters and initial conditions. Moreover, simulation results about model parameter estimation also demonstrate the effectiveness and efficiency of the RQGA.