Research On Improved Artificial Fish Swarm Algorithm And Its Application In Logistics Location Optimization

Artificial fish swarm algorithm(AFSA)is a new global optimization search intelligent algorithm based on animal behavior.Its basic idea is to simulate the fish behaviors,including foraging,rear-end and clusters and achieve global optimization search through the cooperation and competition between artificial fish.This algorithm has the characteristics that are simple,strong parallel capability and weak demand for initial value and it has been successfully applied in signal processing,neural network optimization,image processing,economic system optimization,biological information processing and other fields.However,there are some disadvantages,such as poor diversity,easy to fall into local optimum,slow convergence and reduced search efficiency in the late stage of the algorithm.Therefore,this dissertation proposed a new improved algorithm based on summarizing artificial fish swarm algorithm and its application status and applied the improved algorithm in logistics location optimization.Main works accomplished in this dissertation are as following:(1)Proposed an improved artificial fish swarm algorithm based on DNA computation.Applying the crossover and mutation operation in basic artificial fish swarm algorithm to increase the fish diversity in the late stage of the algorithm,so that the artificial fish can escape from the local extreme points and approach to the global extreme points.This dissertation theoretically analyzed the convergence capability,time complexity and space complexity of the improved algorithm.Besides,test function verified the superiority of the algorithm and applied new improved algorithm in logistics location optimization.The computer simulation result shows that the improved artificial fish swarm algorithm based on DNA computation is more efficient in solving logistics location optimization and can find the lower cost center sites.(2)Proposed an improved artificial fish swarm algorithm based on bacterial foraging.Combined artificial fish swarm algorithm with bacterial foraging algorithm and used the advantage of bacterial foraging algorithm that is chemokine operations have expanded local optimization search capacity.Besides,it also embedded chemokine operator in artificial fish swarm algorithm and improved the local search capability in the late stage of the algorithm.This dissertation analyzed the convergence capability,time complexity and space complexity of the improved algorithm and test function verified the effectiveness of the algorithm.In the same time,this algorithm is also applied in logistics location optimization and simulation verified that the superiority of the improved algorithm is better than the basic artificial fish swarm algorithm and genetic algorithm.(3)Proposed an adaptive Levy distributive mixing mutation artificial fish swarm algorithm.According to the basic idea of that mutation can increase biodiversity.This dissertation introduced Levy mutation and chaos mutation into artificial fish swarm algorithm.Levy mutation can guide artificial fish swarm algorithm escape from local optima and maintain the diversity of artificial fish,chaos mutation enhanced local search capability and ensures the convergence rate of the algorithm in late stage.Theoretical analysis and test function verified the effectiveness of the improved algorithm and applied the improved algorithm in logistics location optimization.Simulation results show that improved algorithm has good optimization performance.