Research On Cuckoo Search And Its Applications

Optimization problem is a kind of mathematical programming problem which can bealways found in science and engineering area. It mainly includes unconstrained continuousproblems, constrained optimization problems and discrete optimization problems. The designof optimization algorithm has been a hot topic in the field of computing. Because of its highefficiency, swarm intelligence has become one of the core theories in computationaloptimization. Cuckoo Search (CS), a new swarm optimization algorithm is introduced in thispaper, the modification of the algorithm and its applications in the function optimizationproblems, constrained optimization problems and scheduling problems are mainly studied.Firstly, the essence of optimization algorithm is introduced, the developments andapplications of CS are summarized. Based on the framework of CS, the teaching-learningmechanism is introduced and a new algorithm, Teaching-learning-based Cuckoo Search(TLCS), is proposed. In order to verify the effectiveness of the proposed algorithm,40famousbenchmark problems was selected and tested on the proposed algorithm. The experimentalresults show the TLCS outperforms other two algorithms.Secondly, on the basis of the TLCS, a constraint handling strategy is constructed and13constrained optimization problems are used to test the performance of TLCS. The results ofthe experiment study are compared with those of the previous algorithms. The experimentalresults show that the proposed approach outperforms other algorithms and has achievedsignificant improvement.Thirdly, the TLCS is applied on the machining parameter optimization and structuraldesign optimization problems. And the proposed method is applied for several well-knownengineering optimization problems. Experimental results show that the TLCS obtains somesolutions better than those previously reported in the literature, which reveals that theproposed TLCS is a very effective and robust approach for these engineering problems.Finally, through the random key representation, the TLCS is applied to solve the flowshop scheduling problem. Several standard benchmarks are used to test the proposed methodand the experimental results showed the TLCS has successfully applied on the solving ofpermutation flow shop problem.