Adaptive Cuckoo Search Algorithm for Parameter Estimation of Heavy Oil Thermal Cracking Model

Abstract

To overcome the shortcomings of cuckoo search, an adaptive cuckoo search (ACS) algorithm is proposed. In the ACS algorithm, the logarithmic adaptive step size is adopted and the random traction component is added to the Levy flight formula to update the population. The effectiveness of the ACS algorithm is verified with four benchmark functions. The ACS algorithm is applied to estimate the parameters of the heavy oil thermal cracking model. The experimental results show that the established model has higher accuracy than the models obtained by the other optimization algorithms.