Name
Title | ||
|---|---|---|
The application of artificial neural network on forecast of meiyu rainfall at lower Yangtze River Valley |
Abstract | ||
|---|---|---|
ANN (the artificial neural network) is applied on the products from the numerical model of WRF (the Meso-Scale Weather Research Forecast Model) for forecasting categorical Rainfall over lower Yangtse river Valley during Meiyu season. The products are as the basic forecast factors, then they are united into combined factors with atmospheric dynamic, thermodynamic and humidity features. The combined factors are further selected into two groups by statistic testing methods. The factors are put into ANN in groups, the categorical rainfall forecasts are obtained through training procedures. The rainfall forecasts from ANN and the prediction of WRF model, and the forecasts by regression equations of the two group factors are compared each other. The real rainfalls as verified values in the comparison are recorded at the three typical stations of Nanjing , Yangzhou and Nantong that all located along the lower Yangtze River Valley. The results showed that the ANN can give better forecast of categorical rainfall during Meiyu season than the other two methods (WRF and regression), the missing and wrong forecast rates are reduced substantially, the accuracy rate of the categorical rain forecast increase over 10% , the accuracy rate of the heavy rain forecast is even higher, near 60% for the chosen cases. During the categorical forecast, the optimization of forecast factors is very important to a better ANN application. |
Year | DOI | Venue |
|---|---|---|
2010 | 10.1109/ICNC.2010.5584627 | ICNC |
Keywords | Field | DocType |
ann(the artificial neural network),numerical model,geographic information systems,regression analysis,numerical analysis,nanjing,nantong,combined factors,yangzhou,ann,artificial neural network,geographic information system,regression equations,weather research forecast model,meiyu rainfall forecasting,wrf model,weather forecasting,the forecast of categorical rainfall,yangtze river valley,neural nets,accuracy rate of forecast,seasonality,accuracy,artificial neural networks,statistical test,forecasting,regression equation,thermodynamics | Meteorology,Quantitative precipitation forecast,Regression,Statistic,Regression analysis,Computer science,Categorical variable,Weather Research and Forecasting Model,Artificial intelligence,Weather forecasting,Machine learning,Precipitation | Conference |
Volume | ISBN | Citations |
4 | 978-1-4244-5958-2 | 0 |
PageRank | References | Authors |
0.34 | 0 | 6 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jianhong Wang | 1 | 6 | 7.92 |
| Zuocheng Yin | 2 | 0 | 0.34 |
| Chunshengng Miao | 3 | 0 | 0.34 |
| Wenxiu Wei | 4 | 0 | 0.34 |
| Yu Zhao | 5 | 34 | 14.73 |
| Yunrong Wu | 6 | 0 | 0.34 |