基于MATLAB遗传算法工具箱的离心泵多目标优化  

Multi-objective Optimization of Centrifugal PumpBased on MATLAB Genetic Algorithm Toolbox

作　　者：曾红[1] 王全玉 张志华[1] ZENG Hong;WANG Quan-yu;ZHANG Zhi-hua(College of Mechanical Engineering and Automation,Liaoning University of Technology,Jinzhou 121001,China)

机构地区：[1]辽宁工业大学机械工程与自动化学院,辽宁锦州121001

出　　处：《辽宁工业大学学报（自然科学版）》2020年第3期152-154,169,共4页Journal of Liaoning University of Technology(Natural Science Edition)

基　　金：辽宁省教育厅高等学校重大科技平台项目(JP2017004)。

摘　　要：为了改善离心泵运行效率低,运行过程不稳定等现状,使用MATLAB遗传算法工具箱,建立以离心泵能量损失最小、汽蚀余量最小、性能曲线无驼峰为目标函数,以离心泵叶轮和蜗壳参数为约束条件的数学模型,使用PumpLinx后处理软件对优化后的模型进行数值模拟;结果显示:与传统的离心泵优化方法相比,使用遗传算法进行优化使能量损失和汽蚀余量显著降低;优化结果证明:遗传算法在离心泵结构优化方面具有很强的实用性,优化之后的模型能量损失减小,离心泵运行稳定性提高。In order to improve the current situation of low efficiency and unstable operation of centrifugal pump,the MATLAB genetic algorithm toolbox is used to establish a mathematical model with minimum energy loss,minimum NPSH and no hump in the performance curve of centrifugal pump as the objective function,and the parameters of impeller and volute of centrifugal pump as the constraint conditions.The optimized model is numerically simulated by using PumpLinx post-processing software.The results show that compared with traditional centrifugal pump optimization methods,the genetic algorithm can significantly reduce the energy loss and NPSH.The optimization results show that genetic algorithm is very practical in centrifugal pump structure optimization.After optimization,the energy loss of the model is reduced and the operation stability of the centrifugal pump is improved.

关 键 词：离心泵 遗传算法 多目标优化 抗汽蚀 无驼峰 

分 类 号：TH311[机械工程—机械制造及自动化]