基于人工神经网络模型的混流式水轮机转轮多目标优化

• Published in: Guanʻgai paishui xuebao Vol. 42; no. 9; pp. 46 - 52
• Main Authors: WANG, Rongtao, LAI Xide, CHEN, Xiaoming
• Format: Journal Article
• Language: Chinese
• Published: Xinxiang City Chinese Academy of Agricultural Sciences (CAAS) Farmland Irrigation Research Institute Editorial Office of Journal of Irrigation and Drainage 01.01.2023; 西华大学能源与动力工程学院,成都 610039
• Subjects: Angles (geometry); Artificial neural networks; Blades; Cavitation; Conical inlets; Design optimization; Draft tubes; Efficiency; Gene mapping; Genetic algorithms; Guide vanes; Hydroelectric plants; Hydroelectric power; Hydroelectric power stations; Multiple objective analysis; Neural networks; Objective function; Optimization; Parameters; Pressure distribution; Random variables; Stability; Turbines; genetic algorithm; artificial neural network; 遗传算法; 转轮; 人工神经网络; multi-objective optimization design; 混流式水轮机; 多目标优化; runner; francis turbine
• ISSN: 1672-3317
• DOI: 10.13522/j.cnki.ggps.2022588

【目的】探究一种同时提升混流式水轮机运行效率、空化性能及运行稳定性的优化方法，为混流式水轮机转轮的多目标优化提供技术途径。【方法】以转轮叶片进出口安放角、安装角为优化变量，通过对叶片几何参数随机离散抽样获取样本数据库，基于CFD数值计算获取各样本的性能参数，进而建立同时考虑混流式水轮机转轮效率、出口旋流数以及空化系数的多目标函数；基于人工神经网络建立优化变量与多目标函数的映射关系，最后采用遗传算法对转轮叶片的18个几何参数进行全局寻优，并对优化前后的转轮叶片性能进行对比分析。【结果】在导叶开度为112°且运行水头分别为160、175、180 m的3个工况下，优化后的转轮效率相较优化前分别提高了0.22%、0.56%、0.60%；叶片压力分布情况得到有效改善；转轮无叶区与尾水管锥管段处压力脉动幅值显著降低。【结论】叶片进口安放角的优化程度越大，混流式水轮机综合性能的提升幅度越大。