Research On Cuckoo Search And Its Applications

Cuckoo Search is a bionic optimization algorithm inspired by the habits of the cuckoos.As one of the swarm intelligence algorithms,Cuckoo Search has the advantages of self-organization,good parallelism,strong global search ability,easy integration with other optimization algorithms,and is widely used in complex optimization problems of scientific research and engineering.Based on the previous research work,this thesis studies and tries to propose the improvements of Cuckoo Search.The proposed algorithms are used to solve optimization problems of practical engineering background.The main contents of this paper are as follows:(1)The modified adaptive cuckoo search algorithm is proposed.From the perspective of parameter control,the proposed algorithm adopts logarithmic adaptive step size and adds random traction component in step of Levy flights to improve the performance.By the experiment of benchmark functions and comparing with other algorithms,this algorithm has better search performance.The proposed algorithm is applied to the parameter estimation of the heavy oil thermal cracking model,and the experimental results verify its effectiveness.(2)The cuckoo search algorithm with explosion operator is proposed.This algorithm uses the exponential adaptive step size and fuses the explosion operator of the fireworks algorithm to enhance the local search ability.In the optimizing experiment of benchmark functions,the effectiveness of the proposed algorithm is demonstrated by comparison with the other algorithms.The proposed algorithm is applied to the four cases of proton exchange membrane fuel cell model.The experimental results show that the proposed algorithm has good performance.(3)The gravity search based cuckoo search algorithm is proposed.The algorithm introduces the gravitational operator while retaining the adaptive step size,and adopts the adaptive discovery probability to improve the search performance.Through the optimizing experiment of benchmark functions,the experimental results verify the good search ability of the proposed algorithm.The proposed algorithm is applied to the parameter estimation of the hydrogenation reaction model.The simulation results show that the proposed model has higher accuracy.