斜流泵叶轮的多参数组合优化

Multi-parameter Optimizing the Impellers in Mixing-flow Pump

以比转速为392的斜流泵为研究模型,选取叶片出口安放角、叶片出口边宽度、后盖板倾角3个参数作为研究变量,采用CFD方法对斜流泵性能进行了多参数组合优化的数值计算,并根据性能试验数据验证了CFD数值方法的准确性。基于Box-Behnken试验建立3因素3水平共17组叶轮模型的试验表并对各方案设计工况下的内流进行了数值计算。采用二次响应面模型拟合效率的性能预测模型,给出目标函数,基于遗传算法对目标函数进行求解。结果表明,对效率影响最为显著的参数为叶片出口安放角,其次为叶片的后盖板倾角,且效率均随二因素的增大而减小,叶片出口边宽度对其影响最小;与原模型叶轮叶片参数相比,β2、b2大于原模型参数,T2保持不变,优化结果显示模型泵水力效率提高1.2%,扬程提高了0.8%。

Using the mixing-flow pump 392 as an example, and taking outlet width of the blade, dip angle of the back shroud, outlet angle of the blade as optimization variables, we analyzed the performance of mixing-flow pump using multi-parameter optimization and CFD simulation. The CFD model were verified by energy perfor- mance data. Three factors andthree levels were used to calculate the internal flow based on the Box-Behnken ex- periment. The predicted performance efficiency was fitted using the quadratic response-surface model. The objec- tive function was solved using the genetic algorithm. The results showed that the most significant parameter was outlet angle of the blade, followed by dip angle of the back shroud, and that the efficiency of the pump decreased as the values of these two factors increased; in contrast, the outlet width of the blade had an insignificant effect on the efficiency. The two impeller blade parameters,β2 and b：, were larger than those of original model, while T2 re- mained unchanged. The optimization results showed that the hydraulic efficiency and the head of the model pump could be improved by 1.2% and 0.8%, respectively.