Several Improvements Of Harmony Search Algorithm And Their Application

Many real-world optimization problems often arise in science,engineering,economics.Studying the solution to these real-world optimization problems has important theory significance and practical application value.Swarm intelligent optimization algorithm as an efficient means and approach for solving large scale real-world optimization problems,which has became a research hotspot of the current optimization field.Many different swarm intelligent optimization algorithms have been widely applied and investigated.Therefore,the research of swarm intelligent optimization algorithm is worth further exploring and promoting.A new swarm intelligence optimization algorithm-harmony search(HS)algorithm is studied deeply in this paper.In order to strengthen the theoretical basis of HS and improve the optimization performance of HS,several modifications are taken on HS algorithm based on different perspectives,and these modified algorithms are applied to real-world optimization problems.For overcoming the existing shortage of HS algorithm,firstly,the distance bandwidth(bw)adjusting methods proposed in recent literatures are summarized,and then the exploration ability of HS improvisation is investigated.Further,the relationship between improvisation exploration and each parameter under asymmetric interval is derived,the effective of bw on the exploration ability and convergence performance is discussed,and an iterative convergence sufficiency of the iteration equation which consists of variance expectation and mean expectation is proven theoretically.Based on these analyses and proofs,a modified harmony search(MHS)algorithm is proposed.Moreover,the effects of the key parameters including HMS,PAR and HMCR on the performance of the MHS algorithm are discussed in depth.Experimental results reveal that the proposed MHS algorithm performs better than HS as well as its state-of-the-art variants and other classic excellent meta-heuristic approaches.An improved harmony search algorithm is proposed,named LHS,which mainly has three improvements:i)adaptive global pitch adjustment is designed to enhance the exploitation ability of solution space;ii)opposition-based learning technique is blended to increase the diversity of solution;iii)competition selection mechanism is established to improve solution precision and enhance the ability of escaping local optima.The performance of the LHS algorithm with respect to harmony memory size(HMS)and harmony memory considering rate(HMCR)are also analyzed in detail.To further evaluate the performance of the proposed LHS algorithm,comparison with ten state-of-the-art harmony search variants over a large number of benchmark functions with different characteristics is carried out.The numerical results confirm the superiority of the proposed LHS algorithm in terms of accuracy,convergence speed and robustness.An improved novel global harmony search(INGHS)approach is proposed to solve reliability optimization problem.This algorithm employs a stochastic position updating strategy to replace the position updating of NGHS,and thus the balance between the exploration and exploitation ability further improved.Meanwhile,in order to handle the constraint of reliability optimization problem,a new constraint relaxation method is presented and the feasibility rule is integrated.Applying the proposed approach to unconstrained global optimization problems and classic reliability optimization problems,results indicate that the INGHS algorithm can find better solution effectively and efficiently,and it also performs better than many improved harmony search algorithms.Specially,the best results obtained by INGHS algorithm are better than the optimal results provided by recent literatures for the reliability optimization problems.An amended harmony search(AHS)algorithm is proposed for solving large-scale system reliability problem?This algorithm amended the searching mechanism of HS algorithm.It taken the best-so-far solution as a study subject,then randomly selected different dimensions to conduct improvisation.Meanwhile,it modified the adjustment method of parameter bandwidth(bw)to balance global and local searching.Classical large-scale system reliability problem is solved,numerical results shown that the proposed algorithm AHS is better than all the reported 6 HS algorithms.AHS algorithm has better optimization performance compared to some excellent algorithms that reported for solving large-scale system reliability problems in the recent year.A binary modified harmony search algorithm is proposed to solve 0-1 knapsack problem(KP).In the algorithm,the best harmony is used to direct the improvisation to generate new harmony,and the improvisation process is modified.The parameter PAR is adjusted dynamically.A stochastic repair operator is developed to effectively repair infeasible harmony,moreover and enhance local search.Besides,a feasible harmony initialization method is used to guarantee initial harmony is feasible.0-1 binary model is completely used in the whole search process.14 0-1 knapsack problems to be tested,the proposed algorithm compared with the other algorithms reported recently,the statistical results demonstrate the effectiveness of the proposed algorithm.In the end,the works of the dissertation are summarized and the perspectives of further research are provided.