岩土工程弹塑性反分析的改进粒子群算法
|
摘要
为了克服常规粒子群算法(PSO)应用于岩土工程弹塑性反演时搜索效率较低、计算工作量大的缺点,通过对算法中适应值比较方式和粒子运动模式的深入分析,指出了其中存在的制约搜索效率的内在因素,并提出相应修改策略,在此基础上形成一种新的改进粒子群算法(IPSO);将新算法用于岩土材料弹塑性参数反演,结果表明,与常规粒子群算法相比,改进算法明显提高了参数的搜索效率,利用较少的迭代次数就能得到满足精度要求的结果,从而减小了岩土工程弹塑性反分析的计算量,是一种可行的参数反演方法.
In order to offset the disadvantages of low searching efficiency and large amount of calculation using traditional PSO for elastoplastic back analysis in geotechnical engineering,the comparison method of fitness value and the motion mode of individual particle of traditional PSO are analyzed and the internal factors which restrict searching efficiency pointed out.Then some corresponding improvement measures are proposed.On this basis,an improved PSO(IPSO) is put forward and used for elastoplastic parameters inverse.The results show that,compared with traditional PSO,the parameter searching efficiency and the convergence precision of IPSO have an obvious increase and the inverse results meeting accuracy requirements can be obtained with less number of iterations.Thus the calculation amount for elastplastic back analysis can be greatly reduced if using IPSO.So our proposed algorithm is a feasible method for elastoplastic back analysis in geotechnical engineering.
In order to offset the disadvantages of low searching efficiency and large amount of calculation using traditional PSO for elastoplastic back analysis in geotechnical engineering,the comparison method of fitness value and the motion mode of individual particle of traditional PSO are analyzed and the internal factors which restrict searching efficiency pointed out.Then some corresponding improvement measures are proposed.On this basis,an improved PSO(IPSO) is put forward and used for elastoplastic parameters inverse.The results show that,compared with traditional PSO,the parameter searching efficiency and the convergence precision of IPSO have an obvious increase and the inverse results meeting accuracy requirements can be obtained with less number of iterations.Thus the calculation amount for elastplastic back analysis can be greatly reduced if using IPSO.So our proposed algorithm is a feasible method for elastoplastic back analysis in geotechnical engineering.
引文
[1]杨林德.岩土工程问题的反演理论与工程实践[M].北京:科学出版社,1995:23-29.
[2]杨志法,王思敬,冯紫良,等.岩土工程反分析原理及应用[M].北京:地震出版社,2002:1-3.
[3]孙钧,黄伟.岩石力学参数弹塑性反演问题的优化方法[J].岩石力学与工程学报,1992,11(3):221-229.SUN Jun,HUANG Wei.An optimization methodfor the elastoplastic inversion of parameters in rockmechanics[J].Chinese Journal of Rock Mechanicsand Engineering,1992,11(3):221-229.
[4]KENNEDY J,EBERHART R.Particle swarm opti-mization[C]//Proceedings of IEEE InternationalConference on Neural Network.Piscataway:NJIEEE Service Center,1995:1942-1948.
[5]高玮.基于粒子群优化的岩土工程反分析研究[J].岩土力学,2006,27(5):795-798.GAO wei.Back analysis algorithm in geotechnical en-gineering based on particle swarm optimization[J].Rock and Soil Mechanics,2006,27(5):795-798.
[6]李爱国.多粒子群协同优化算法[J].复旦学报:自然科学版,2004,43(5):923-925.LI Ai-guo.Particle swarms cooperative optimizer[J].Journal of Fudan University:Natural Science Edition,2004,43(5):923-925.
[7]谭皓,沈春林,李锦.混合粒子群算法在高维复杂函数寻优中的应用[J].系统工程与电子技术,2005,27(8):1471-1474.TAN Hao,SHEN Chun-lin,LI Jin.Hybrid particleswarm optimization algorithm for high-dimensioncomplex functions[J].System Engineering and Elec-tronics,2005,27(8):1471-1474.
[8]田明俊,周晶.岩土工程参数反演的一种新方法[J].岩石力学与工程学报,2005,24(9):1492-1496.TIAN Ming-jun,ZHOU Jing.New algorithm for pa-rameter inversion in geotechnical engineering[J].Chinese Journal of Rock Mechanics and Engineering,2005,24(9):1492-1496.
[9]刘建华,樊晓平,瞿志华.一种惯性权重动态调整的新型粒子群算法[J].计算机工程与应用,2007,43(7):68-70.LIU Jian-hua,FAN Xiao-ping,QU Zhi-hua.Newparticle swarm optimization algorithm with dynamicchange of inertia weights[J].Computer Engineeringand Application,2007,43(7):68-70.
[10]郭文忠,陈国龙.粒子群优化算法中惯性权值调整的一种新策略[J].计算机工程与科学,2007,29(1):70-72,75.GUO Wen-zhong,CHEN Guo-long.A new strate-gy of inertia weight adjustment for particle swarmoptimization[J].Computer Engineering and Sci-ence,2007,29(1):70-72,75.
[2]杨志法,王思敬,冯紫良,等.岩土工程反分析原理及应用[M].北京:地震出版社,2002:1-3.
[3]孙钧,黄伟.岩石力学参数弹塑性反演问题的优化方法[J].岩石力学与工程学报,1992,11(3):221-229.SUN Jun,HUANG Wei.An optimization methodfor the elastoplastic inversion of parameters in rockmechanics[J].Chinese Journal of Rock Mechanicsand Engineering,1992,11(3):221-229.
[4]KENNEDY J,EBERHART R.Particle swarm opti-mization[C]//Proceedings of IEEE InternationalConference on Neural Network.Piscataway:NJIEEE Service Center,1995:1942-1948.
[5]高玮.基于粒子群优化的岩土工程反分析研究[J].岩土力学,2006,27(5):795-798.GAO wei.Back analysis algorithm in geotechnical en-gineering based on particle swarm optimization[J].Rock and Soil Mechanics,2006,27(5):795-798.
[6]李爱国.多粒子群协同优化算法[J].复旦学报:自然科学版,2004,43(5):923-925.LI Ai-guo.Particle swarms cooperative optimizer[J].Journal of Fudan University:Natural Science Edition,2004,43(5):923-925.
[7]谭皓,沈春林,李锦.混合粒子群算法在高维复杂函数寻优中的应用[J].系统工程与电子技术,2005,27(8):1471-1474.TAN Hao,SHEN Chun-lin,LI Jin.Hybrid particleswarm optimization algorithm for high-dimensioncomplex functions[J].System Engineering and Elec-tronics,2005,27(8):1471-1474.
[8]田明俊,周晶.岩土工程参数反演的一种新方法[J].岩石力学与工程学报,2005,24(9):1492-1496.TIAN Ming-jun,ZHOU Jing.New algorithm for pa-rameter inversion in geotechnical engineering[J].Chinese Journal of Rock Mechanics and Engineering,2005,24(9):1492-1496.
[9]刘建华,樊晓平,瞿志华.一种惯性权重动态调整的新型粒子群算法[J].计算机工程与应用,2007,43(7):68-70.LIU Jian-hua,FAN Xiao-ping,QU Zhi-hua.Newparticle swarm optimization algorithm with dynamicchange of inertia weights[J].Computer Engineeringand Application,2007,43(7):68-70.
[10]郭文忠,陈国龙.粒子群优化算法中惯性权值调整的一种新策略[J].计算机工程与科学,2007,29(1):70-72,75.GUO Wen-zhong,CHEN Guo-long.A new strate-gy of inertia weight adjustment for particle swarmoptimization[J].Computer Engineering and Sci-ence,2007,29(1):70-72,75.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心