摘要
为平衡种群的探索与开发,提出一种改进的竞争粒子群算法(CGCSO)。通过柯西高斯变异更新胜利者的位置,提高种群的开发能力;利用环形拓扑结构信息传递速度慢的特点,将其用于胜利者的学习过程,增强种群的多样性;采用可行解优先的约束处理技术,使该算法能够处理约束优化问题。进行8个标准测试函数的仿真实验,并研究比较其它算法,该算法在优化精度和收敛性上表现较好。将该算法应用于处理汽油调和配方在线优化问题,仿真取得了较好的结果,进一步验证了该算法的有效性。
To balance the exploration and exploitation of the population,an improved competitive swarm optimizer was proposed(CGCSO).The Winners' positions were updated by Cauchy and Gaussian mutation,which improved the exploitation capability of the population.The ring topology with slow transmission of information was applied to the Winners' learning process,which enhanced the diversity of the population.Feasibility rules were adopted as constraint technique to deal with constrained optimization problems.According to the experiments on 8 benchmark functions,and compared with the other algorithms,the proposed algorithm shows better performance,especially on the optimization accuracy and convergence.The CGCSO algorithm was applied to deal with the real-time optimization of gasoline blending recipe,and the simulation results also verify the effective performance of the proposed algorithm.
引文
[1]Cheng R,Jin Y.A competitive swarm optimizer for large scale optimization[J].IEEE Transaction on Cybernetics,2015,45(2):191-204.
[2]Zhang WX,Chen WN,Zhang J. A dynamic competitive swarm optimizer based on entropy for large scale optimization[C]//8th International Conference on Advance Computational Intelligence.Chiang Mai:IEEE,2016:365-371.
[3]Mohapatra P,Das KN,Roy S.A modified competitive swarm optimizer for large scale optimization problems[J].Applied Soft Computing,2017,59:340-362.
[4]Hu Z,Cai X,Fan Z.An improved memetic algorithm using ring neighborhood topology for constrained optimization[J].Soft Computing,2014,18(10):2023-2041.
[5]Hu M,Wu T,Weir JD.An adaptive particle swarm optimization with multiple adaptive methods[J].IEEE Transactions on Evolutionary Computation,2013,17(5):705-720.
[6]KANG Lanlan,DONG Wenyong,TIAN Jiangsen.Opposition-based particle swarm optimization with adaptive Cauchy mutation[J].Computer Science,2015,42(10):226-231(in Chinese).[康岚兰,董文永,田降森.一种自适应柯西变异的反向学习粒子群优化算法[J].计算机科学,2015,42(10):226-231.]
[7]Lim W, Mat Isa,Nor Ashidi.Particle swarm optimization with adaptive time-varying topology connectivity[J].Applied Soft Computing,2014,24:623-642.
[8]Wang Y,Wang BC,Yen GG.Incorporating objective function information into the feasibility rule for constrained evolutionary optimization[J].IEEE Transaction on Cybernetics,2016,46(12):2938-2952.
[9]Elsayed SM,Sarker RA,Essam DL.Adaptive configuration of evolutionary algorithms for constrained optimization[J].Applied Mathematics and Computation,2013,222:680-711.
[10]Debchoudhury S,Biswas S.Modified estimation of distribution algorithm with differential mutation for constrained optimization[C]//IEEE Congress on Evolutionary Computation.Cancun:IEEE,2013:1724-1731.
[11]He K,Qian Feng,Cheng Hui,et al.A novel adaptive algorithm with near-infrared spectroscopy and its application in online gasoline blending processes[J].Chemometrics and Intelligent Laboratory Systems,2015,140:117-125.
[12]YUAN Qi,CHENG Hui,ZHONG Weimin,et al.Improved group search optimizer and application on gasoline blending process[J].Journal of Chemical Industry and Engineering,2013,64(12):4427-4433(in Chinese).[袁奇,程辉,钟伟民,等.全局群搜索优化算法及其在汽油调合中的应用[J].化工学报,2013,64(12):4427-4433.]
[13]MIAO Zhenhua,SUN Xudong,SHAO Cheng.An adaptive genetic algorithm with parallel mutation and its performance evaluation[J].Information and Control,2016,45(2):142-150(in Chinese).[苗振华,孙旭东,邵诚.一种并行变异自适应遗传算法及其性能分析[J].信息与控制,2016,45(2):142-150.]