Studies And Application Of Firefly Algorithm

Firefly algorithm(FA)is a novel optimization algorithm in the field of swarm intelligence,which mimics the flashing and communication behavior of fireflies and is a kind of stochastic algorithm.Because of its simple structure,fewer parameters to tune and preferable search capability,many researchers pay attention to it.Now,FA has been applied widely in many fields such as engineering,computer,management,economics and biology,et al.However,similar to other population based stochastic algorithms,FA has inherent defects.For instances,the algorithm converges slowly at later period,easy to premature convergence and fall into the local optima.Thus may result to the low precision.In order to solve these problems,this paper improves the initialization,population evolution and population diversity of firefly optimization algorithm,and applies it to clustering.The main work arrangements are as follows:(1)The following improvements are proposed based on the traditional FA: the traditional FA using random method to generate the initial population,likely to cause uneven initial population,which is not conducive to solving the optimal value.In order to make the initial population more homogeneous and improve the population diversity,the chaotic self-mapping optimization in chaotic optimization strategy is used to generate chaotic sequences to initialize the location of fireflies.Secondly,in the process of algorithm evolution,inertia weight is introduced to control the influence of the previous generation on the offspring,and the best individual guidance is used to enhance information sharing among individuals.Then,the Gauss mutation operation is added to disturb the individuals who may fall into the local optimum region at the later stage of the iteration,which helps the population jump out of the local optimum value and increase the population diversity.Finally,the dynamic step length and the symmetric boundary variation operation are introduced,and the global search and local optimization are balanced to solve the individual cross boundary problem.The improved FA is compared with traditional FA and related algorithms in the 6 standard functions,the results show that the improved FA has higher accuracy and faster convergence speed,and the comprehensive performance of the improved algorithm is obviously improved.(2)The improved FA is applied to practical problems: To further evaluate the performance of the improved algorithms,we apply them to integration into K-means clustering algorithm.Although K-means clustering algorithm is simple and popular,it has a fundamental drawback of falling into local optima that depends on the randomly generatedinitial centroid values.Because of the better global search ability and the faster convergence rate of FA,we apply three of the modified FA to integration into K-means clustering algorithm to overcome its defect and obtain the globally optimal solutions.Experiments on UCI datasets show that the modified FA is effective optimization tool and can obtain the better clustering results.