In this paper, an efficient stochastic optimization algorithm is presented for parameter identification of nonlinear systems. Due to its robust performance, short running time and desirable potency to find local minimums the Lozi map-based chaotic optimization algorithm is an appropriate choice to estimate unknown parameters of nonlinear dynamic systems. To enhance the identification efficacy and in order to escape local minimum, a modified version of this algorithm with higher stability and better performance is rendered in this paper. An Improved Lozi map-based chaotic optimization algorithm (ILCOA) is employed to identify three nonlinear systems and the performance of the proposed algorithm is compared with other optimization algorithms. The simulation results of identification endorse the effectiveness of the proposed method.