感潮河段水位过程预报探讨
Preliminary Study on the Water Level Forecast in the River with Tidal Effect
-
摘要: 感潮河段水位过程同时受径流、潮流影响,其预报较为困难.将水位过程视为时间序列,用不同的模型进行了预报比较,认为以潮汐作用为主的水位预报可采用神经网络模型,对高水位预报可采用ARMA(2,1)模型,而对洪水作用为主的河道,各模型的预报能力均有待提高Abstract: Forecasting the water level with tidal effect is difficulty,because it is influenced by flood and tidal.This paper treats the water level as time series and compares thewater level forecasted by some model with observed,the result shows that the neural network model is suitable for tidal river and the ARMA(2,1) is suitable for high level forecasting,but for floodriver,the forecasting ability of all models should be improved.