差分变异和交叉迁移的生物地理学优化算法
|
摘要
为了增强生物地理学优化(BBO)算法的优化性能,提出了一种差分变异和交叉迁移的BBO算法(DCBBO).首先用差分扰动操作替换BBO算法的变异操作,形成差分变异算子,强化了探索能力;其次用基于维度的垂直交叉操作取代BBO算法的迁移操作,形成交叉迁移算子,提升开采能力的同时又注重了探索能力;最后,为平衡算法的探索和开采,将启发式水平交叉操作融入交叉迁移算子中,形成混合交叉迁移算子,进一步提升开采能力.在不同维度的一组常用基准函数上进行了大量实验,结果表明,与其他state-of-the-art算法相比,DCBBO优化能力显著,稳定性更强,运行速度更快.
In order to enhance the optimization performance of the biogeography-based optimization( BBO) algorithm,an improved BBO algorithm with differential mutation and cross migration( DCBBO)was proposed. Firstly,BBO's mutation operation was replaced by a differential disturbance operation to form a differential mutation operator. It could improve the exploration. Secondly,a dimension-based vertical crossover operation was used instead of BBO's original migration operation to generate a cross migration operator. It could improve the exploitation and emphasize the exploration. Finally,to balance the exploration and exploitation,a heuristic horizontal crossover operation was merged into the cross operator to obtain a hybrid cross migration operator. It could further improve the exploitation. A large number of experiments were made on a set of common benchmark functions with different dimensions. The results showed that DCBBO could obtain more significant optimization ability,stronger stability and faster running speed compared with other state-of-the-art algorithms.
In order to enhance the optimization performance of the biogeography-based optimization( BBO) algorithm,an improved BBO algorithm with differential mutation and cross migration( DCBBO)was proposed. Firstly,BBO's mutation operation was replaced by a differential disturbance operation to form a differential mutation operator. It could improve the exploration. Secondly,a dimension-based vertical crossover operation was used instead of BBO's original migration operation to generate a cross migration operator. It could improve the exploitation and emphasize the exploration. Finally,to balance the exploration and exploitation,a heuristic horizontal crossover operation was merged into the cross operator to obtain a hybrid cross migration operator. It could further improve the exploitation. A large number of experiments were made on a set of common benchmark functions with different dimensions. The results showed that DCBBO could obtain more significant optimization ability,stronger stability and faster running speed compared with other state-of-the-art algorithms.
引文
[1]SIMON D.Biogeography-based optimization[J].IEEE transactions on evolutionary computation,2008,12(6):702-713.
[2]MA H.An analysis of the equilibrium of migration models for biogeography-based optimization[J].Information sciences,2010,180(18):3444-3464.
[3]FENG Q,LIU S,ZHANG J,et al.Improved biogeography-based optimization with random ring topology and Powell's method[J].Applied mathematical modelling,2017,41:630-649.
[4]XIONG G,SHI D,DUAN X.Enhancing the performance of biogeography-based optimization using polyphyletic migration operator and orthogonal learning[J].Computers and operations research,2014,41(1):125-139.
[5]张新明,康强,涂强,等.融合细菌觅食趋化算子的生物地理学优化算法[J].郑州大学学报(理学版),2016,48(4):44-53.
[6]STORN R,PRICE K.Differential evolution:a simple and efficient heuristic for global optimization over continuous spaces[J].Journal of global optimization,1997,11(4):341-359.
[7]张新明,尹欣欣,涂强.动态迁移和椒盐变异融合生物地理学优化算法的高维多阈值分割[J].光学精密工程,2015,23(10):2943-2951.
[8]张新明,涂强,尹欣欣.混合迁移的高效BBO算法及其在图像分割中的应用[J].计算机科学与探索,2016,10(10):1459-1468.
[9]ZHENG Y J,LING H F,XUE J Y.Ecogeography-based optimization:enhancing biogeography-based optimization with ecogeographic barriers and differentiations[J].Computers and operations research,2014,50(10):115-127.
[10]夏学文,刘经南,高柯夫,等.具备反向学习和局部学习能力的粒子群算法[J].计算机学报,2015,38(7):1397-1407.
[11]周晓根,张贵军,郝小虎,等.一种基于局部Lipschitz下界估计支撑面的差分进化算法[J].计算机学报,2016,39(12):2631-2651.
[12]NIU Q,ZHANG L,LI K.A biogeography-based optimization algorithm with mutation strategies for model parameter estimation of solar and fuel cells[J].Energy conversion and management,2014,86:1173-1185.
[13]SIMON D,OMRAN M G H,CLERC M.Linearized biogeography-based optimization with re-initialization and local search[J].Information sciences,2014,267:140-157.
[14]ZHANG J,SANDERSON A C.JADE:adaptive differential evolution with optional external archive[J].IEEE transactions on evolutionary computation,2009,13(5):945-958.
[2]MA H.An analysis of the equilibrium of migration models for biogeography-based optimization[J].Information sciences,2010,180(18):3444-3464.
[3]FENG Q,LIU S,ZHANG J,et al.Improved biogeography-based optimization with random ring topology and Powell's method[J].Applied mathematical modelling,2017,41:630-649.
[4]XIONG G,SHI D,DUAN X.Enhancing the performance of biogeography-based optimization using polyphyletic migration operator and orthogonal learning[J].Computers and operations research,2014,41(1):125-139.
[5]张新明,康强,涂强,等.融合细菌觅食趋化算子的生物地理学优化算法[J].郑州大学学报(理学版),2016,48(4):44-53.
[6]STORN R,PRICE K.Differential evolution:a simple and efficient heuristic for global optimization over continuous spaces[J].Journal of global optimization,1997,11(4):341-359.
[7]张新明,尹欣欣,涂强.动态迁移和椒盐变异融合生物地理学优化算法的高维多阈值分割[J].光学精密工程,2015,23(10):2943-2951.
[8]张新明,涂强,尹欣欣.混合迁移的高效BBO算法及其在图像分割中的应用[J].计算机科学与探索,2016,10(10):1459-1468.
[9]ZHENG Y J,LING H F,XUE J Y.Ecogeography-based optimization:enhancing biogeography-based optimization with ecogeographic barriers and differentiations[J].Computers and operations research,2014,50(10):115-127.
[10]夏学文,刘经南,高柯夫,等.具备反向学习和局部学习能力的粒子群算法[J].计算机学报,2015,38(7):1397-1407.
[11]周晓根,张贵军,郝小虎,等.一种基于局部Lipschitz下界估计支撑面的差分进化算法[J].计算机学报,2016,39(12):2631-2651.
[12]NIU Q,ZHANG L,LI K.A biogeography-based optimization algorithm with mutation strategies for model parameter estimation of solar and fuel cells[J].Energy conversion and management,2014,86:1173-1185.
[13]SIMON D,OMRAN M G H,CLERC M.Linearized biogeography-based optimization with re-initialization and local search[J].Information sciences,2014,267:140-157.
[14]ZHANG J,SANDERSON A C.JADE:adaptive differential evolution with optional external archive[J].IEEE transactions on evolutionary computation,2009,13(5):945-958.