基于Kriging模型与NLPQL算法的高比转数离心泵参数化设计

Parametric design of high-specific-speed centrifugal pump based on Kriging model and NLPQL algorithm

针对目前采用传统设计方法无法有效解决高比转数离心泵效率低的问题,以型号为100-80-125的高比转数离心泵为研究对象,基于Kriging模型对其叶轮子午流道盖板控制点参数和叶片型线参数进行了全面的敏感性筛选,并采用NLPQL算法对高比转数离心泵进行参数优化,以期提高其运行效率.对优化后的模型进行了数值计算,并通过试验对比了优化前后模型泵的运行效率.研究结果表明:在额定流量工况下,优化后泵模型的实际效率提高了2.6%,扬程提高了2.3 m;优化后未出现不利于铸造和加工的几何形状,优化效果较为显著.研究结果可为提高高比转数离心泵的效率提供了一种行之有效的方法.

The traditional design method cannot effectively solve the problem of low efficiency of a high-specific-speed centrifugal pump.Hence,a 100-80-125 high-specific-speed centrifugal pump was taken as the research object,a comprehensive sensitivity screening on the impeller cover control points and blade profile parameters were carried out based on the Kriging model.The NLPQL algorithm was used to optimize the parameters of high-specific-speed centrifugal pump to improve its operation efficiency.The optimized model was numerically calculated,and the operation efficiency of the model pump before and after optimization was compared through experiments.The results show that under the rated flow condition,the actual efficiency of the optimized pump is increased by 2.6%,and the head is increased by 2.3 m.Besides,there is no structure difficult to cast and process,and the optimization effect is significant.This research can provide an effective method for improving the efficiency of high-specific-speed centrifugal pump.