提出引入自适应因子的改进型差分进化算法(IDE),并应用于优化在线支持回归机(OSVR)的核参数和惩罚参数,建立岩土体位移和岩土力学参数之间复杂非线性关系的动态最优IDE-OSVR模型,输入岩土体位移值直接输出岩土力学参数实现参数反演。通过均匀设计方法与ABAQUS正计算构建初始训练集,然后逐次反演并进行验算误差,将误差未达到预定阈值前的验算样本增添至训练集,使得IDE-OSVR模型不断在线学习,提高参数反演精度。将IDE-OSVR-ABAQUS反演方法应用于工程算例,并同几种典型方法对比。结果表明该方法的岩土力学参数反演速度很快,反演精度很高,是一种合理的岩土力学参数反演方法。
Abstract
An improved differential evolution algorithm (IDE) is proposed by introducing adaptive factor and is applied to optimizing the kernel parameter and penalty parameter of online support vector regression (OSVR). A dynamic optimal IDE-OSVR model which reflects the complex nonlinear relationship between rock and soil mass displacements and geotechnical parameters is established. The inversion of parameters can be accomplished by inputting the soil mass displacements in the IDE-OSVR model. The initial training set is designed with uniform design method and ABAQUS calculation, then the errors of successive inversed parameters are checked, and the checked sample will be added to training set if the error is greater than the predetermined threshold. Through the continuous online learning process, the precision of parameters inversion by IDE-OSVR model can be enhanced. The inversion method based on IDE-OSVR-ABAQUS is applied to an engineering example and the result is compared with those of typical methods. The comparison result shows that the IDE-OSVR-ABAQUS inversion method is very fast with high accuracy, hence is a reasonable method for the inversion of geotechnical mechanics parameters.
关键词
改进型差分进化算法 /
在线支持回归机 /
ABAQUS /
参数反演 /
岩土体位移
Key words
IDE /
OSVR /
ABAQUS /
parameters inversion /
soil mass displacement
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 李 枫. 岩体力学参数的非线性随机反演优化方法的研究.raybet体育在线
院报,2006,23(3):51-54.
贾善坡. 基于遗传算法的岩土力学参数反演及其在ABAQUS中的实现.水文地质工程地质,2012,39(1):31-35.
韩 峰,徐 磊,张太俊. 坝基岩体力学参数的PSO-ABAQUS 联合反演.河海大学学报,2013,41(4):321-325.
曹明杰,曹 鑫,徐政治. 量子遗传算法在混凝土重力坝综合弹性模量反演中的应用.raybet体育在线
院报,2016,33(4):111-114.
冯夏庭,张治强,杨成祥,等. 位移反分析的进化神经网络方法研究.岩石力学与工程学报,1999,18(5):497-502.
文辉辉,尹健民,秦志光,等. BP神经网络在隧道围岩力学参数反演中的应用.raybet体育在线
院报,2013,30(2):47-51,56.
赵洪波. 基于微粒群优化的智能位移反分析研究用.岩土工程学报,2006,28(11):2035-2308.
杨云浩,徐卫亚,聂卫平. 基于ε-SVR与PSO-DE的岩层弹塑性参数反演及应用.中国矿业大学学报,2009,40(1):95-102.
漆祖芳,姜清辉,周创兵,等. 基于ε-SVR和MVPSO算法的边坡位移反分析方法及其应用.岩石力学与工程学报,2013,32(6):1185-1196.
STORN R, PRICE K. Differential Evolution-A Simple and Efficient Heuristic for Global Optimization over Continuous Spaces. Journal of Global Optimization, 1997, 11(4):341-359.
刘 琛,林 盈,胡晓敏.差分演化算法各种更新策略的对比分析.计算机科学与探索,2013,7(11):983-993.
许小健,黄小平,钱德玲.自适应加速差分进化算法.复杂系统与复杂性科学,2008,5(1):87-92.
卢远富,包腾飞,李涧鸣,等. 基于改进型混合蛙跳算法的支持回归机大坝变形预测模型.三峡大学学报(自然科学版),2015,37(5):14-18.
王定成,姜 斌.支持向量机控制与在线学习方法研究进展.系统仿真学报,2007,19(6):1177-1512.
王定成.支持向量机建模预测与控制.北京:气象出版社,2009.
PLATT J. Using Analytic QP and Sparseness to Speed Training of Support Vector Machines∥Neural Information Processing Systems Foundation. Proceedings of Advances in Neural Information Processing Systems. November 29-December 4,1999:557-563.
向 文,张强勇,张建国.坝区岩体蠕变参数解析——智能反演方法及其工程应用 .岩土力学,2015,36(5):1505-1512.
贾善坡. Boom Clay泥岩渗流应力损伤耦合流变模型、参数反演与工程应用.武汉:中国科学院武汉岩土力学研究所,2009.
方开泰.均匀设计与均匀设计表.北京:科学出版社,1994.