Parameter estimation of chaotic system based on multi-mechanism hybrid Elephant herding algorithm

To estimate the unknown parameters of chaotic system model, a multi-mechanism hybrid elephant herding algorithm is proposed. For the slow convergence and low precision of elephant herding algorithm, it is difficult to jump out of the shortcomings of local optimum. Using chaotic initialization can increase the diversity of the population and avoid falling into local extremum. Z-type scaling factor and global optimum solution are used to enhance its search ability, and opposition-basedlearningis used to increase the diversity of the evolutionary population of the algorithm. Improve the local search ability to approach the global optimal individual with higher accuracy. By test functions, the improved algorithm is superior to the elephant optimization, sine consine algorithm, whale optimization and multi-verse optimizer.Finally, the parameters of two chaotic systems are estimated, which verifies that the improved algorithm has strong optimization speed, accuracy and stability.