长短叶片混流式水轮机水力性能预测精度验证及改进  

Validation and Improvement of Prediction Accuracy in Hydraulic Performance of Francis Turbine with Splitter Blades

作　　者：张杰[1] 周大庆[1] 陈会向[2] 刘敏[1] 

机构地区：[1]河海大学能源与电气学院,江苏南京211100 [2]河海大学水利水电学院,江苏南京211100

出　　处：《水电能源科学》2016年第10期161-164,109,共5页Water Resources and Power

基　　金：国家自然科学基金重点项目(51339005);中央高校基本科研业务费(2015B34314)

摘　　要：针对长短叶片混流式水轮机选择不同湍流模型、各部件网格划分方案对流量、转矩、效率预测相对偏差值较大、预测精度较低问题,利用ICEM CFD划分软件对各工况(小流量、设计流量、大流量)下的水轮机全流道及其过流部件(蜗壳及导叶、转轮、尾水管)各划分出4种网格分配方案,采用收敛较快的S-A模型初步预测了水轮机的水力效率与所划分的网格规模无关,进而选择方案3分析水轮机的水力性能;选取S-A模型、标准κ-ε模型、RNGκ-ε模型、标准κ-ω模型、SSTκ-ω模型5种湍流模型计算了流量、转矩、水力效率的预测值,发现各模型预测的水力效率预测精度较高,但流量及转矩的预测精度偏低。为此,重新划分过流部件网格规模,并再划分出5种网格方案,仍采用S-A湍流模型进行初步分析,发现虽可提高流量及转矩的预测精度,但提升幅度有限;进而采用RNGκ-ε、SSTκ-ω模型、Realizableκ-ε模型、RSM模型预测流量及转矩,可进一步提高流量、转矩的预测精度,但仍低于效率的预测精度。所提出的预测方法可为水力机械设计和性能优化提供重要参考。To solve the problem of large relative deviation and low prediction accuracy of discharge,torque and efficiency of the Francis turbine with splitter blades,which influenced by turbulence models and meshing schemes,ICEM CFD software was used in each operating condition（low flow,designed flow,high flow）and all components（spiral and guide vanes,runner,draft tube）to put forward four kinds of grid schemes.The S-A model with a faster convergence was adopted to preliminarily predict the hydraulic efficiency of water turbine for proving the independence of grid partition.Then,the third scheme was selected for further analysis.Five kinds of turbulence models（S-A model,standardκ-εmodel,RNGκ-εmodel,standardκ-ωmodel,SSTκ-ωmodel）were selected to calculate the predicted value of the flow rate,torque and hydraulic efficiency,which finds hydraulic efficiency with high prediction precision in all models but the flow rate and torque with a relative lower prediction accuracy.So,five other grid schemes were meshed again and S-A turbulence model was also used for preliminary analysis.It finds that this method can improve the prediction accuracy of the flow and torque,but the amplitude of enhancement is limited.Furthermore,RNGκ-ε,SSTκ-ω model,Realizableκ-εmodel and RSM model were used to predict flow rate and torque,the further improvement of flow prediction accuracy and torque appears,but still lower than the prediction accuracy of efficiency.The proposed prediction method can provide an important reference for the design and performance optimization of hydraulic machinery.

关 键 词：长短叶片混流式水轮机 CFD数值模拟 网格划分 湍流模型 预测精度 

分 类 号：TK733.1[交通运输工程—轮机工程] TK730[动力工程及工程热物理—流体机械及工程]