Study on Data Mining of Jet Field Based on Artificial Neural Network
For the problems of single analysis mean of jet field data and inadequate knowledge discovery, we u-ses standard κ-ε turbulence to found mathematical model of the vertical round buoyant jet in cross flow based on previous test data firstly, and then we obtains a large number of jet field data by numerical experiments, and further we introduces the data mining mode based on neural network theory (EDM-π&-ANN). Through the data mining of jet field, non-linear formulas of velocity trajectory, temperature trajectory and the dilution at the highest position of temperature trajectory have been got finally, which have the features of harmonious dimension and unified form. And some new laws of vertical round buoyant jet in cross flow have also been founded. The investigation results provide an effective method to data analysis for knowledge discovery of jet and other similar scientific issues, and may have some academic reference value for further understanding of movement mechanism of jet.
Cao Xiaomeng Gu Zhenghua Hu Yaan Liu Wang Xu Xiaodong
Institute of Water Resources, Zhejiang University, Hangzhou, 310058 Department of Hydraulic Engineering, Nanjing Hydraulic Research Institute, Nanjing, 210029
国际会议
南京
英文
222-230
2010-09-13（万方平台首次上网日期，不代表论文的发表时间）