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.