大型洞室群软岩置换方案优化的并行实现  

Parallel computing for scheme optimization of soft-rock replacement in a large cavern group

作　　者：李邵军[1] 冯夏庭[1] 安红刚[2] 

机构地区：[1]中科院武汉岩土力学研究所岩土力学重点实验室,湖北武汉430071 [2]同济大学地下建筑与工程系,上海200092

出　　处：《岩土力学》2003年第3期335-338,共4页Rock and Soil Mechanics

基　　金：国家重点基础研究发展规划项目(编号:2002CB412708);自然科学基金项目(编号:50179034);中科院知识创新工程重要方向项目(编号:KGCX2-SW-302-02)

摘　　要：并行计算已成为求解大规模岩土工程问题的一种强大趋势。以水布垭大型洞室群软岩置换方案优化为例,探讨了方案优化中的并行计算问题,分析了并行计算中的编程模式、任务划分、负载平衡和编程方法等问题,在Windows环境下的PC机群上成功实现了软岩置换方案优化的并行计算,并获得了近乎线性的加速比,从而大大提高了方案优化的计算速度和效率,为岩土工程计算并行化思路提供了重要参考。Parallel computing is becoming the strong trend of solving large scale geotechnical problems. Taking the scheme optimization of the soft-rock replacement in Shuibuya large cavern group for example, parallel computing for the scheme of optimization is discussed. The following problems such as the parallel programming modes, programming method, tasks dividing, data communication and load equilibrium are analyzed. The parallel computing for the scheme optimization of soft-rock replacement is successfully accomplished in the personal computers which run in the windows operating system. The final results indicate that the linear speedup ratio is almost achieved, accordingly the computing speed and efficiency of the scheme optimization is promoted greatly. This paper provides an important reference for the application of parallel computing in the geotechnical engineering.

关 键 词：软岩置换 方案优化 并行计算 线性加速比 

分 类 号：TU457[建筑科学—岩土工程]