Abstract: An novel chaos harmony search algorithms(CHS) are proposed to overcome the shortcoming of lacking the local searching ability.Firstly,in order to distribute more evenly in the domain space,the initialized population is generated based on the good point set.Secondly,an adaptive mechanism,in which candidate harmony is generated,is introduced to accelerate the convergence speed.Thirdly,a local search mechanism based on the chaos sequence is designed to improve the local search capability of CHS The simulation results for five benchmarks show that compared with the basic harmony search algorithm,the new algorithms performance better.