On the Application of Improved Back Propagation Neural Network in Real-Time Forecast

For the classical algorithm of BP network model, its convergence rate is slow and it may result in locally optimal solution. But on the condition of same arithmetic complicacy, the Fletcher-Reeves algorithm can improve the convergence rate and come to the least point along the conjugate direction so as to improve the forecasting precision of the BP network model. According to the check results of the BP network model in Guanyinge reservoir, it is proved that this model can fulfill the requirement of forecasting precision and is valuable to be used for reference or be generalized in real-time forecast of afflux runoff in other area under the same condition.