Firefly Algorithm Based On Immune Multi-population And Its Application

Nowadays,optimization problems have penetrated into most fields,such as engineering,science,industry and so on.The bionic swarm intelligence optimization algorithm based on biological intelligence has many advantages,such as high efficiency,parallel,strong versatility,no special information and so on.It provides a new idea for solving the optimization problem.Firefly algorithm(FA)is a new kind of bionic swarm intelligence optimization algorithm inspired by the information of fireflies' luminous behavior.In this method,the brighter fireflies attract other individuals to move towards them in the search field,so as to change their position.In order to overcome the shortcomings of traditional firefly algorithm which includes low accuracy,weak stability and easily falling into premature convergence,in this paper,an immune multi-population firefly algorithm(IMPFA)is proposed.The method is applied to solve multimodal function optimization problems,time delay estimation for x-ray pulsar navigation(XPNAV)and wireless sensor networks(WSNs)coverage optimization problems.The main work and achievements of the paper are as follows:(1)Aiming at multimodal function optimization problem,each superiority of firefly algorithm and clonal selection algorithm,a novel serial hybrid immune multi-population firefly algorithm is proposed.In each loop iteration,the firefly algorithm with multi-population evolution mechanism(MPFA)is used for global search in the feasible region,and then the robust non-uniform mutation clonal selection algorithm(NUMCSA)is used to improve the accuracy of the sub-optimal solution which MPFA has found.The performance of IMPFA algorithm was tested by introducing complex multimodal functions.The experimental results show that IMPFA has good optimization results and stability.(2)In order to further test the effect of immune multi-population firefly algorithm,this algorithm is applied to two engineering optimization problems.The estimation of time of arrival(TOA)of the pulse is one of the important factors that affects the performance of XPNAV system,and the calculation of TOA is through time delay estimation.In this paper,a novel time delay estimation method for selecting the bispectral points based on IMPFA is presentated.This method combines bispectral algorithm and improved firefly algorithm.Experiments show that this method effec-tively reduces the computational complexity of the navigation algorithm.Network coverage is an important factor to evaluate WSNs measurement performance.We established a mathematical model to achieve the coverage of objective area.In this paper,the improved firefly algorithm is introduced to optimize the deployment of network nodes,and a new wireless sensor networks coverage optimization algorithm(IMPFA-WSNs)based on IMPFA is presentated.Simulation results show that compared with traditional firefly algorithm,the IMPFA-WSNs algorithm can expand the area coverage rate.