FLAC^(3D) V3.0边坡稳定剪切强度折减法的多线程并行算法  

Study on the slope stability shear strength reduction method based on the multi-thread parallel computing technology in the FLAC^(3D) V3.0 software

作　　者：瞿秀祥 林杭[1] 王敏[1] 张道勇 QU Xiuxiang;LIN Hang;WANG Min;ZHANG Daoyong(School of Resources and Safety Engineering,Central South University,Changsha 410083,China;China Machinery International Engineering Design ＆ Research Institute Co.,Ltd,Changsha 410007,China)

机构地区：[1]中南大学资源与安全工程学院,湖南长沙410083 [2]中机国际工程设计研究院有限公司,湖南长沙410007

出　　处：《铁道科学与工程学报》2018年第7期1693-1699,共7页Journal of Railway Science and Engineering

基　　金：长江科学院开放研究基金资助项目(CKWV2014218/KY);中南大学"创新驱动计划"资助项目(2016CX019);国家自然科学基金资助项目(51474249);湖南省自然科学基金资助项目(2018JJ2500)

摘　　要：强度折减法将边坡的安全系数定义为边坡刚好达到临界破坏状态时的折减系数,该过程一般采用二分法对边坡内聚力和内摩擦角进行迭代计算。为优化迭代计算次数以缩短计算时间,提出多线程并行运算强度折减法的计算策略,得到二分法和多线程并行运算在不同折减系数计算范围(Kmin,Kmax)、给定误差条件η下所需最小迭代次数的理论公式。将FLAC^(3D)命令流内嵌于Python脚本中,通过Python脚本同时调用多个FLAC^(3D)应用程序,分别采用4线程并行运算和传统二分法对实例边坡进行强度折减法的计算。研究结果表明:在本文计算条件下,2种方法计算结果相同,采用4线程并行运算时迭代次数是二分法运算的0.46倍,其总的计算时间为二分法的0.55倍。采用多线程并行运算能有效减少迭代次数及总的计算时间。The safety factor of slopes was determined by the critical reduction parameter when utilizing the strength reduction method, in the process, the cohesion and internal friction angle were reduced by the bisection method usually. To reduce the iterations and computing time, the multi-thread parallel computing technology was utilized in the strength reduction method. Through theoretical analysis and computing, the minimum iteration equations of the bisection method and multi-thread parallel computing method were derived with the range（Kmin, Kmax） of reduction parameter and the given error η determined. The FLAC^（3D） command flow was embedded in Python scripts, subsequently, the multi-thread computing of FLAC^（3D） software were performed by using the Python scripts. A specific computing example was conducted by 4 threads parallel computing and the bisection method, separately. From the study of this paper, it is concluded that the final results are the same, meanwhile, the iterations of 4 threads parallel method is 0.46 times less than that of the bisection method, and the time is 0.55. The computing example suggested that the multi-parallel computing technology can reduce iterations and computing time effectively.

关 键 词：Python脚本 FLAC3D 边坡 强度折减法 多线程并行运算 

分 类 号：TU457[建筑科学—岩土工程]