Solving the parameter identification problem using shuffled frog leaping with opposition-based initialization

The parameter identification problem can be modeled as a non-linear optimization problem. In this problem, some unknown parameters of a mathematical model presented by an ordinary differential equation using some experimental data must be estimated. This paper presents a shuffled frog leaping algorithm for solving parameter identification problem. An opposition-based initialization strategy is used to choose the fitter members as initial population. Two test cases are considered to examine the efficiency of utilized algorithm. The comparison results of shuffled frog leaping and other methods proposed in the different literature demonstrate that the shuffled frog leaping has a comparable performance than other evolutionary algorithms.