基于Isight的液力变矩器泵叶优化设计

Isight-Based Optimization Design of Hydrodynamic Torque Converter Pump Vanes

针对液力变矩器复杂的内流场仿真计算与泵轮叶片优化设计这两类非线性且孤立的计算问题,提出了一种新的叶片优化设计方法。这种方法是在Isight平台上,集成参数化的三维流道模型文件,网格划分文件和流体分析文件,以内置的多岛遗传算法为工具,进行自动优化设计。该方法实现了将LB46液力变矩器泵轮叶片几何形状的优化转化为控制其叶片形状的参数数值优化。将优化结果与实验数据进行对比,证明了该方法的可靠性。

A new blade optimization design method is proposed,which aims at two types of non-linear and isolated calculation problems,the complex internal flow field simulation calculation of the hydraulic torque converter and the optimal design of the pump impeller blades.It integrates parameterized 3D flow channel model files,meshing files and fluid analysis files on Isight platform,and uses built-in multi-island genetic algorithm as a tool to automatically optimize the design.The geometry optimization of the pump blade of the LB46 hydraulic torque converter is transformed into the numerical optimization of parameters controlling its blade shape.The comparison of experiment data with optimization result proves the reliability of this method.