In this paper, a novel evolutionary learning algorithm is proposed by hybridizing the Taguchi method, chaos disturbance operation, multilevel immune algorithm (MIA), and artificial bee colony algorithm (ABC). The algorithm is thus called HTCMIABC to estimate the parameter of chaotic systems. The HTCMIABC comprises two main different phases. First, we use the MIA as the recognition phase to balance local and global searches and accelerate the search speed to enhance the evolutionary phase. Second, the evolutionary phase is built on the ABC and chaos disturbance operation to have the capabilities of exploration and exploitation. Moreover, the Taguchi method and crossover operation are inserted between the recognition phase and evolutionary phase for the recombination and diversification of several antibodies to improve the searching ability. Finally, the HTCMIABC algorithm is examined by parameter identification of the nonlinear chaotic system. Simulation results show that the proposed algorithm is more efficient than some typical existing algorithms. The effects of noise and population size are investigated as well.
All Science Journal Classification (ASJC) codes
- Modelling and Simulation
- Computational Theory and Mathematics
- Computational Mathematics