典型耦合优化算法在源项反演中的对比研究
|
摘要
突发大气污染事故中,污染源的快速、准确确定是应急处置的基础.为研究有效的源项评估方法,本文基于美国草原外场SO_2释放实验,利用GA-PSO、GA-NM、PSO-NM3种耦合算法,分别与高斯点源烟羽扩散模型结合,对源强和位置等污染源参数进行反演与对比,并从算法结构与大气扩散条件方面进行反演效果差异分析.结果表明,从源强反演角度看,PSO-NM反演结果的准确性最高、稳定性最强,平均误差(11.3%)与平均标准偏差(0.7g/s)明显低于GA-NM(16.4%、13.3g/s)与GA-PSO(29.0%、26.6g/s).从位置反演角度看,PSO-NM的反演结果最为稳定,反演的平均标准偏差(0.29m)明显低于GA-NM(3.20m)与GA-PSO(3.03m)算法;在不稳定和中性扩散条件下,PSO-NM算法的位置反演准确性最高,误差为4.97m;但在稳定扩散条件下,GA-NM的位置反演误差(7.69m)最小.从反演效率角度看,PSO-NM与GA-NM反演时间最短,更适用于污染源的快速确定.
Rapid and accurate estimation of source items was the basis for environment emergency disposal on sudden air pollution accidents. In order to search for effective methods for inversing source parameters, we conducted a comparison study on the performances of three hybrid algorithms(e.g., GA-PSO, GA-NM, PSO-NM) for estimating source parameters(strength and location). Three inversion models were developed by combining GA-PSO, GA-NM, PSO-NM with Gaussian dispersion model, respectively. The study was carried out based upon SO_2 leakage tests selected from 1956 Prairie Grass emission experiment. The impacts of algorithm structure and atmospheric diffusion conditions on source term inversion were analyzed. Results showed that for source strength, the PSO-NM algorithm performed more accurate and robust, the mean error and mean standard deviation were 11.3% and 0.7 g/s, respectively, which were much lower than those of GA-NM(i.e., 16.4%, 13.3 g/s) and GA-PSO(i.e., 29.0%, 26.6 g/s). As for source location, the performance of PSO-NM was more robust, with average standard deviation of 0.29 m, which was also much lower than that of GA-NM(3.20 m) and GA-PSO(3.03 m). Under the unstable and neutral atmospheric diffusion conditions, the accuracy of PSO-NM algorithm for estimating position parameter was the best, with an error of 4.97 m; However, GA-NM method had the minimum error(7.69 m) under the stable condition. As for computational efficiency, PSO-NM and GA-PSO spent less time in source item inversion, which were more suitable for inversing source parameters for sudden air pollution.
Rapid and accurate estimation of source items was the basis for environment emergency disposal on sudden air pollution accidents. In order to search for effective methods for inversing source parameters, we conducted a comparison study on the performances of three hybrid algorithms(e.g., GA-PSO, GA-NM, PSO-NM) for estimating source parameters(strength and location). Three inversion models were developed by combining GA-PSO, GA-NM, PSO-NM with Gaussian dispersion model, respectively. The study was carried out based upon SO_2 leakage tests selected from 1956 Prairie Grass emission experiment. The impacts of algorithm structure and atmospheric diffusion conditions on source term inversion were analyzed. Results showed that for source strength, the PSO-NM algorithm performed more accurate and robust, the mean error and mean standard deviation were 11.3% and 0.7 g/s, respectively, which were much lower than those of GA-NM(i.e., 16.4%, 13.3 g/s) and GA-PSO(i.e., 29.0%, 26.6 g/s). As for source location, the performance of PSO-NM was more robust, with average standard deviation of 0.29 m, which was also much lower than that of GA-NM(3.20 m) and GA-PSO(3.03 m). Under the unstable and neutral atmospheric diffusion conditions, the accuracy of PSO-NM algorithm for estimating position parameter was the best, with an error of 4.97 m; However, GA-NM method had the minimum error(7.69 m) under the stable condition. As for computational efficiency, PSO-NM and GA-PSO spent less time in source item inversion, which were more suitable for inversing source parameters for sudden air pollution.
引文
[1]唐仙丽,杨宇静.论中国突发性环境污染事故应急立法的完善[J].重庆大学学报(社会科学版), 2005,11(6):114-117.Tang X L, Yang Y J. Improvement of emergency legislation of accidental pollution in China[J]. Journal of ChongQing University(Social Science Edition), 2005,11(6):114-117.
[2] Andrew Keats E Y F L. Bayesian inference for source determination with applications to a complex urban environment[J]. Atmospheric Environment, 2007,41:465-479.
[3] Inanc Senocak N W H M. Stochastic event reconstruction of atmospheric contaminant dispersion using Bayesian inference[J].Atmospheric Environment, 2008,42(33):7718-7727.
[4] Ak?elik V, Biros G, Ghattas O, et al. A variational finite element method for source inversion for convective–diffusive transport[J].Finite Elements in Analysis&Design, 2003,39(8):683-705.
[5]张建文,王煜薇,郑小平,等.基于混合遗传-Nelder Mead单纯形算法的源强及位置反算[J].系统工程理论与实践, 2011,31(8):1581-1587.Zhang J W, Wang Y W, Zheng X P, et al. Back-calculation of source strength and position by a hybrid genetic-Nelder Mead simplex algorithm[J]. System Engineering-Theory&Practice, 2011,31(8):1581-1587.
[6] Ma D, Tan W, Zhang Z, et al. Parameter identification for continuous point emission source based on Tikhonov regularization method coupled with particle swarm optimization algorithm[J]. Journal of Hazardous Materials, 2017,325:239-250.
[7] Thomson L C, Hirst B, Gibson G, et al. An improved algorithm for locating a gas source using inverse methods[J]. Atmospheric Environment, 2007,41(6):1128-1134.
[8]苏芳,邵敏,蔡旭辉,等.利用逆向轨迹反演模式估算北京地区甲烷源强[J].环境科学学报, 2002,22(5):586-591.Su F, Shao M, Cai X H, et al. Estimation of methane emissions in Beijing area using backward trajectory inversion model[J]. Acta Scientiae Circumstantiae, 2002,22(5):586-591.
[9]李航,张宏升,蔡旭晖,等.日本福岛核电站泄漏事故污染物扩散的数值模拟与事故释放源项评估[J].安全与环境学报, 2013,13(5):265-270.Li H, Zhang H S, Cai X H, et al. Simulation of the pollutant dispersion and evaluation on the source release accidents in Japan's Fukushima Nuclear Power Plant[J]. Journal of Safety and Environment, 2013,13(5):265-270.
[10]张美根.多尺度空气质量模式系统及其验证II.东亚地区对流层臭氧及其前体物模拟[J].大气科学, 2005,29(6):84-94.Zhang M G. A multi-scale air quality modeling system and its evaluation II.simulation of tropospheric ozone and its precursors in east asia[J]. Chinese Jo urnal of Atmospheric Sciences, 2005,29(6):84-94.
[11]唐穗欣.标准遗传算法的原理及算例[J].软件导刊, 2007,6(1):99-101.Tang S X. Principle and example of standard genetic algorithm[J].Software Guide, 2007,6(1):99-101.
[12]张丽平,俞欢军,陈德钊,等.粒子群优化算法的分析与改进[J].信息与控制, 2004,33(5):513-517.Zhang L P, Yu H J, Chen D Z, et al. AnaIysis and improvement of particle swarm optimization algorithm[J]. Information and Control,2004,33(5):513-517.
[13]郑伟博,张纪会.基于Nelder-Mead单纯形法的改进量子行为粒子群算法[J].复杂系统与复杂性科学, 2016,13(2):97-104.Zheng W B, Zhang J H. A improved quantum behaved particular swarm optimization algorithm using nelder and mead’s simplex algorithm[J].Complex Systems and Complexity Science, 2016,13(2):97-104.
[14]潘勇,郭晓东.一种基于遗传算法改进的粒子群优化算法[J].计算机应用与软件, 2011,28(9):222-224.Pan Y, Guo X D. An improved particle swarm optimisation algorithm based on genetic algorithm[J]. Computer Applications and Software,2011,28(9):222-224.
[15]李雅琼.基于粒子群算法的遗传算法优化研究[J].兰州文理学院学报(自然科学版), 2017,31(1):55-60.Li Y Q. Research of genetic algorithm based on particular swarm optimization[J]. Journal of Lanzhou University of Arts and Science(NaturalSciences), 2017,31(1):55-60.
[16]巩永光.粒子群算法与遗传算法的结合研究[J].济宁学院学报,2008,29(6):20-22.Gong Y G. The study of PSO with GA[J]. Journal of Jining University, 2008,29(6):20-22.
[17]肖宏峰,谭冠政.单纯形搜索在遗传算法中的融合研究[J].计算机工程与应用, 2008,44(18):30-33.Xiao H F, Tan G Z. Study on fusing simplex search into genetic algorithm[J]. Computer Engineering and Applications, 2008,44(18):30-33.
[18]杜修力,崔冬,侯本伟.基于初始种群改进策略的经验遗传-单纯形算法[J].北京工业大学学报, 2014,40(12):1876-1883.Du X L, Cui D, Hou B W. Initial population improvement strategy of empirical genetic-simplex algorithm[J]. Journal of Beijing University of Technology, 2014,40(12):1876-1883.
[19]何大阔,李延强,王福利.基于单纯形算子的混合遗传算法[J].信息与控制, 2001,30(3):276-278.He D K, Li Y Q, Wang F L. Hybrid genetic algorithm based on the operator of pattern search[J]. Information and Control, 2001,30(3):276-278.
[20]王芳.粒子群算法的研究[D].西南大学, 2006.Wang F. Research on particle swarm algorithm[D]. Southwest University, 2006.
[21] Fan S S, Zahara E. A hybrid simplex search and particle swarm optimization for unconstrained optimization[J]. European Journal of Operational Research, 2007,181(2):527-548.
[22] Vakil Baghmisheh M T, Peimani M, Sadeghi M H, et al. A hybrid particle swarm–Nelder–Mead optimization method for crack detection in cantilever beams[J]. Applied Soft Computing, 2012,12(8):2217-2226.
[23] Ma D, Deng J, Zhang Z. Comparison and improvements of optimization methods for gas emission source identification[J].Atmospheric Environment, 2013,81:188-198.
[24] Zhang C, Ning J, Lu S, et al. A novel hybrid differential evolution and particle swarm optimization algorithm for unconstrained optimization[J]. Operations Research Letters, 2009,37(2):117-122.
[25] Cantelli A, D'Orta F, Cattini A, et al. Application of genetic algorithm for the simultaneous identification of atmospheric pollution sources[J].Atmospheric Environment, 2015,115:36-46.
[26] Cervone G, Franzese P. Non-Darwinian evolution for the source detection of atmospheric releases[J]. Atmospheric Environment, 2011,45(26):4497-4506.
[27] Barad M L. Project prairie grass, a field program in diffusion,[J].Geophysical Research, 1958,1(59).
[28] Wang Y, Huang L, Huang H, et al. Evaluation of Bayesian source estimation methods with Prairie Grass observations and Gaussian plume model:A comparison of likelihood functions and distance measures[J]. Atmospheric Environment, 2017,152:519-530.
[29]章颖,梁漫春,黎岢,等.基于遗传-模拟退火算法的源项反演方法研究[J].核电子学与探测技术, 2014,34(4):451-455.Zhang Y, Liang M C, Li K, et al. Research of source term inversion based on genetic simulated annealing algorithms[J]. Nuclear Electronics&Detection Technology, 2014,34(4):451-455.
[2] Andrew Keats E Y F L. Bayesian inference for source determination with applications to a complex urban environment[J]. Atmospheric Environment, 2007,41:465-479.
[3] Inanc Senocak N W H M. Stochastic event reconstruction of atmospheric contaminant dispersion using Bayesian inference[J].Atmospheric Environment, 2008,42(33):7718-7727.
[4] Ak?elik V, Biros G, Ghattas O, et al. A variational finite element method for source inversion for convective–diffusive transport[J].Finite Elements in Analysis&Design, 2003,39(8):683-705.
[5]张建文,王煜薇,郑小平,等.基于混合遗传-Nelder Mead单纯形算法的源强及位置反算[J].系统工程理论与实践, 2011,31(8):1581-1587.Zhang J W, Wang Y W, Zheng X P, et al. Back-calculation of source strength and position by a hybrid genetic-Nelder Mead simplex algorithm[J]. System Engineering-Theory&Practice, 2011,31(8):1581-1587.
[6] Ma D, Tan W, Zhang Z, et al. Parameter identification for continuous point emission source based on Tikhonov regularization method coupled with particle swarm optimization algorithm[J]. Journal of Hazardous Materials, 2017,325:239-250.
[7] Thomson L C, Hirst B, Gibson G, et al. An improved algorithm for locating a gas source using inverse methods[J]. Atmospheric Environment, 2007,41(6):1128-1134.
[8]苏芳,邵敏,蔡旭辉,等.利用逆向轨迹反演模式估算北京地区甲烷源强[J].环境科学学报, 2002,22(5):586-591.Su F, Shao M, Cai X H, et al. Estimation of methane emissions in Beijing area using backward trajectory inversion model[J]. Acta Scientiae Circumstantiae, 2002,22(5):586-591.
[9]李航,张宏升,蔡旭晖,等.日本福岛核电站泄漏事故污染物扩散的数值模拟与事故释放源项评估[J].安全与环境学报, 2013,13(5):265-270.Li H, Zhang H S, Cai X H, et al. Simulation of the pollutant dispersion and evaluation on the source release accidents in Japan's Fukushima Nuclear Power Plant[J]. Journal of Safety and Environment, 2013,13(5):265-270.
[10]张美根.多尺度空气质量模式系统及其验证II.东亚地区对流层臭氧及其前体物模拟[J].大气科学, 2005,29(6):84-94.Zhang M G. A multi-scale air quality modeling system and its evaluation II.simulation of tropospheric ozone and its precursors in east asia[J]. Chinese Jo urnal of Atmospheric Sciences, 2005,29(6):84-94.
[11]唐穗欣.标准遗传算法的原理及算例[J].软件导刊, 2007,6(1):99-101.Tang S X. Principle and example of standard genetic algorithm[J].Software Guide, 2007,6(1):99-101.
[12]张丽平,俞欢军,陈德钊,等.粒子群优化算法的分析与改进[J].信息与控制, 2004,33(5):513-517.Zhang L P, Yu H J, Chen D Z, et al. AnaIysis and improvement of particle swarm optimization algorithm[J]. Information and Control,2004,33(5):513-517.
[13]郑伟博,张纪会.基于Nelder-Mead单纯形法的改进量子行为粒子群算法[J].复杂系统与复杂性科学, 2016,13(2):97-104.Zheng W B, Zhang J H. A improved quantum behaved particular swarm optimization algorithm using nelder and mead’s simplex algorithm[J].Complex Systems and Complexity Science, 2016,13(2):97-104.
[14]潘勇,郭晓东.一种基于遗传算法改进的粒子群优化算法[J].计算机应用与软件, 2011,28(9):222-224.Pan Y, Guo X D. An improved particle swarm optimisation algorithm based on genetic algorithm[J]. Computer Applications and Software,2011,28(9):222-224.
[15]李雅琼.基于粒子群算法的遗传算法优化研究[J].兰州文理学院学报(自然科学版), 2017,31(1):55-60.Li Y Q. Research of genetic algorithm based on particular swarm optimization[J]. Journal of Lanzhou University of Arts and Science(NaturalSciences), 2017,31(1):55-60.
[16]巩永光.粒子群算法与遗传算法的结合研究[J].济宁学院学报,2008,29(6):20-22.Gong Y G. The study of PSO with GA[J]. Journal of Jining University, 2008,29(6):20-22.
[17]肖宏峰,谭冠政.单纯形搜索在遗传算法中的融合研究[J].计算机工程与应用, 2008,44(18):30-33.Xiao H F, Tan G Z. Study on fusing simplex search into genetic algorithm[J]. Computer Engineering and Applications, 2008,44(18):30-33.
[18]杜修力,崔冬,侯本伟.基于初始种群改进策略的经验遗传-单纯形算法[J].北京工业大学学报, 2014,40(12):1876-1883.Du X L, Cui D, Hou B W. Initial population improvement strategy of empirical genetic-simplex algorithm[J]. Journal of Beijing University of Technology, 2014,40(12):1876-1883.
[19]何大阔,李延强,王福利.基于单纯形算子的混合遗传算法[J].信息与控制, 2001,30(3):276-278.He D K, Li Y Q, Wang F L. Hybrid genetic algorithm based on the operator of pattern search[J]. Information and Control, 2001,30(3):276-278.
[20]王芳.粒子群算法的研究[D].西南大学, 2006.Wang F. Research on particle swarm algorithm[D]. Southwest University, 2006.
[21] Fan S S, Zahara E. A hybrid simplex search and particle swarm optimization for unconstrained optimization[J]. European Journal of Operational Research, 2007,181(2):527-548.
[22] Vakil Baghmisheh M T, Peimani M, Sadeghi M H, et al. A hybrid particle swarm–Nelder–Mead optimization method for crack detection in cantilever beams[J]. Applied Soft Computing, 2012,12(8):2217-2226.
[23] Ma D, Deng J, Zhang Z. Comparison and improvements of optimization methods for gas emission source identification[J].Atmospheric Environment, 2013,81:188-198.
[24] Zhang C, Ning J, Lu S, et al. A novel hybrid differential evolution and particle swarm optimization algorithm for unconstrained optimization[J]. Operations Research Letters, 2009,37(2):117-122.
[25] Cantelli A, D'Orta F, Cattini A, et al. Application of genetic algorithm for the simultaneous identification of atmospheric pollution sources[J].Atmospheric Environment, 2015,115:36-46.
[26] Cervone G, Franzese P. Non-Darwinian evolution for the source detection of atmospheric releases[J]. Atmospheric Environment, 2011,45(26):4497-4506.
[27] Barad M L. Project prairie grass, a field program in diffusion,[J].Geophysical Research, 1958,1(59).
[28] Wang Y, Huang L, Huang H, et al. Evaluation of Bayesian source estimation methods with Prairie Grass observations and Gaussian plume model:A comparison of likelihood functions and distance measures[J]. Atmospheric Environment, 2017,152:519-530.
[29]章颖,梁漫春,黎岢,等.基于遗传-模拟退火算法的源项反演方法研究[J].核电子学与探测技术, 2014,34(4):451-455.Zhang Y, Liang M C, Li K, et al. Research of source term inversion based on genetic simulated annealing algorithms[J]. Nuclear Electronics&Detection Technology, 2014,34(4):451-455.