Analysis Of Influence Of Meteorological Factors An Ganjiang River Basin In The Long-Term Hydrological Forecast

The Ganjiang River is a river in Poyang Lake river runoff,the largest river length of the longest and most important.All along,the Ganjiang River Basin suffers from frequent flood threat.However,there are still some problems in the long-term hydrological forecasting,such as the low precision and the low generality of the forecasting model,which makes it difficult to effectively guide the production practice.Long term hydrological forecasting is relatively backward in theoretical research and practical application,and it is in a stage of continuous development.As the capital of Jiangxi Province,Nanchang in the middle and lower reaches of Ganjiang River,therefore,the research of long-term hydrological forecast in the middle and lower reaches of Ganjiang River and some more important significance.The choice of export in the valley of the lower reaches of Ganjiang River control station,station outside the state as a representative in the middle and lower reaches of Ganjiang River Station,recognition and statistical analysis of genetic analysis,combining the method of influence of previous forecast factor of hydrological elements.Then using the association rules analysis,artificial neural networks were qualitative prediction of the Ganjiang River Basin(foreign station abundant water change forecast and quantitative forecast(out of state)annual maximum peak flow,to establish a quantitative prediction)of the Ganjiang River Basin and simple and practical model for forecasting accuracy.The main conclusions are: from the qualitative forecast results,relative sunspot number and the eastern Pacific sea surface temperature and the West Pacific subtropical high abnormal state can cause abnormal changes of runoff in the station,but this change is not absolute.Moreover,the support and confidence of low flow forecast are far greater than that of abundant water forecast from the existing dataFrom the quantitative forecast results,based on the prediction results of BP neural network LM algorithm to improve the annual peak flow in foreign station is quite good,and with the passage of time,hydrological data increasing,prediction accuracy will improve the model.