| 基于神经网络的离心泵汽蚀性能预测 |
| |
| 引用本文： | LIU Hou-Lin,王勇,TAN Ming-gao,陆斌斌,FENG Yan-hua.基于神经网络的离心泵汽蚀性能预测[J].排灌机械,2008,26(3):15-18. |
| |
| 作者姓名： | LIU Hou-Lin 王勇 TAN Ming-gao 陆斌斌 FENG Yan-hua |
| |
| 作者单位： | 江苏大学,流体机械工程技术研究中心,江苏,镇江212013 |
| |
| 基金项目： | 国家自然科学基金资助项目
,
国家"863"计划项目 |
| |
| 摘 要： | 介绍了离心泵汽蚀性能预测的研究现状，分析了离心泵汽蚀性能预测的主要研究方法．根据设计流量下离心泵汽蚀余量的影响因素，确定人工神经网络的拓扑结构．应用MATLAB的神经网络工具箱，建立单级单吸离心泵汽蚀性能预测的BP神经网络（Back Propagation Neural Network）和RBF神经网络（Radial Basis Function Neural Network）两种人工神经网络模型．用工程实践中得到的57台离心泵几何参数和试验数据作为样本来训练建立好的网络，并用6台离心泵的数据来测试网络．预测值与试验值的相关性分析表明，BP和RBF网络的预测结果均较好，其中BP网络预测模型的平均相对偏差为5．69％，RBF网络预测模型的平均相对偏差为6．32％，可满足工程应用的要求．
|
| 关 键 词： | 离心泵 汽蚀性能 预测 神经网络 网络模型 |
Prediction on cavitation performance of centrifugal pumps based on artificial neural network |
| |
| LIU Hou-Lin,WANG Yong,TAN Ming-gao,LU Bin-bin,FENG Yan-hua.Prediction on cavitation performance of centrifugal pumps based on artificial neural network[J].Drainage and Irrigation Machinery,2008,26(3):15-18. |
| |
| Authors: | LIU Hou-Lin WANG Yong TAN Ming-gao LU Bin-bin FENG Yan-hua |
| |
| Institution: | LIU Hou-Lin, WANG Yong , TAN Ming-gao , LU Bin-bin , FENG Yan-hua ( 1. Technical and Research Center of Fluid Machinery Engineering, Jiangsu University, Zhenjiang, Jiangsu 212013, China; 2. China Shipbuilding Industry Corporation No. 704 Research Institute, Shanghai 200031, China) |
| |
| Abstract: | The current status and principal methods for predicting the cavitation performance of centrifugal pumps were presented. Topological structures of artificial neural networks were determined and network models for predicting cavitation performance of centrifugal pumps were established by analyzing the relations between geometric parameters of centrifugal pumps and net positive suction head at rated flow rate, based on the neural network toolbox of MATLAB. The BP and RBF neural networks were trained by 57 example dates, which were obtained from engineering practice, and tested respectively by 6 sets products. The correlation between the predicted and tested values were analyzed by using linear regression method. Results show that the predictions by those two neural networks are satisfied, and the average declination of BP and RBF are 5.69% and 6.32% respectively . |
| |
| Keywords: | centrifugal pump cavitation performance prediction artificial neural network network model |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
