一种改进的自适应混合型蝙蝠算法
|
摘要
针对基本蝙蝠算法(BA)存在的寻优精度不高,易出现早熟收敛等缺陷,本文提出了一种改进的自适应混合型蝙蝠算法(YSBA).首先,该算法舍弃了速度这一参数,简化了计算;其次,加入位置收缩因子β,用来控制与约束蝙蝠的位置,平衡蝙蝠算法中局部和全局搜索,提高算法的寻优精度,最后,重新设置了响度A和脉冲频率r的计算方法,此方法可以也可以避免陷入局部最优.最后通过11个典型的基准函数优化试验,与基本蝙蝠算法(BA)以及采用机动飞行的蝙蝠算法(MFBA)相比,发现改进的自适应混合型蝙蝠算法能够解决局部过分搜索的问题,避免陷入局部最优值,具有较高的计算精度.
Aiming at the existence of basic bat algorithm(BA)optimization accuracy is not high,traps into local optima easily.This paper presents a new improved bat algorithm,which is named YSBA.In this algorithm,firstly,to simplify the calculation and improve the convergence speed,a new search equation is proposed in generate new solutions.Secondly,location constrict factor is added,which can be used to control with the location of the bats,balance the global and local search of bats and improve the optimization precision of the algorithm.Finally,reset the method of the calculation of loudness and rate,which can also be used to avoid trapping into local search.To verify the performance of our algorithm,11 typical experiments are employed.The experimental results show that the new algorithm(YSBA)is significantly improved,which includes optimization accuracy,convergence speed,and they can also avoid falling into a local optimum.
Aiming at the existence of basic bat algorithm(BA)optimization accuracy is not high,traps into local optima easily.This paper presents a new improved bat algorithm,which is named YSBA.In this algorithm,firstly,to simplify the calculation and improve the convergence speed,a new search equation is proposed in generate new solutions.Secondly,location constrict factor is added,which can be used to control with the location of the bats,balance the global and local search of bats and improve the optimization precision of the algorithm.Finally,reset the method of the calculation of loudness and rate,which can also be used to avoid trapping into local search.To verify the performance of our algorithm,11 typical experiments are employed.The experimental results show that the new algorithm(YSBA)is significantly improved,which includes optimization accuracy,convergence speed,and they can also avoid falling into a local optimum.
引文
[1]Yang X-S.A new metaheuristic bat-inspired algorithm.Nature inspired cooperative strategies for optimization[M].Berlin:Springer,2010:65-74.
[2]徐华,张庭,包哲人,等.求解柔性作业车间调度问题的改进蝙蝠算法[J].信息与控制,2016,45(6):722-728.
[3]GolmaryamiM,SalehS,ArdekaniA,et al.A new modified bat algorithm to solve optimal management of multi-objective reconfiguration problem[J].Journal Of Intelligent&Fuzzy Systems,2014,27(3):1567-1573.
[4]Gandomi,A,Yang,X,Alavi,A,et al.Bat algorithm for constrained optimization tasks[J].Neural Computing&Applications,2013,22(6):1239-1255.
[5]张勇凯,李芳,包晓晓.改进蝙蝠算法在云制造供应链中的应用[J].数学理论与应用,2015,5(2):83-94.
[6]Clerc M,Kennedy J.The Particle swarm-explosion,stability,and convergence in a multidimensional complex space[J].IEEE Transactions on Evolutionary Comput7tion,2002,6(l):58-73.
[7]王文,王勇,王晓伟.采用机动飞行的蝙蝠算法[J].计算机应用研究,2014,31(10):2962-2964.
[2]徐华,张庭,包哲人,等.求解柔性作业车间调度问题的改进蝙蝠算法[J].信息与控制,2016,45(6):722-728.
[3]GolmaryamiM,SalehS,ArdekaniA,et al.A new modified bat algorithm to solve optimal management of multi-objective reconfiguration problem[J].Journal Of Intelligent&Fuzzy Systems,2014,27(3):1567-1573.
[4]Gandomi,A,Yang,X,Alavi,A,et al.Bat algorithm for constrained optimization tasks[J].Neural Computing&Applications,2013,22(6):1239-1255.
[5]张勇凯,李芳,包晓晓.改进蝙蝠算法在云制造供应链中的应用[J].数学理论与应用,2015,5(2):83-94.
[6]Clerc M,Kennedy J.The Particle swarm-explosion,stability,and convergence in a multidimensional complex space[J].IEEE Transactions on Evolutionary Comput7tion,2002,6(l):58-73.
[7]王文,王勇,王晓伟.采用机动飞行的蝙蝠算法[J].计算机应用研究,2014,31(10):2962-2964.