摘要
河川径流预测是一个十分复杂的问题,生命旋回模型在进行径流趋势预测时具有对资料要求少、计算简单等优点,但由于模型方程的限制,进行预测时得到的序列很难反映径流序列的随机波动变化,且存在预测结果精度不高的缺点。根据黄河龙门水文站提供的49a的径流水文资料。应用神经网络进行预测时和生命旋回模型建模使用,得到预测值。结果表明预测结果明显优于单一的生命旋回模型和神经网络模型,可以用于径流预测。
The life cycle model is a new long-term prediction model for river runoff, which requires less data, simple calculation. However, due to the physical mechanism, the prediction results of the life cycle model can＇t reflect the fluctuation and random characteristics of the river annual runoff, and it cannot meet required precision for hydraulics and hydropower engineering management. Aiming at these problems, the life cycle neural network prediction model was put forward. The life cycle neural network model combines lifecycle with neural network, and it includes two kinds of prediction model structure： parallel life cycle neural network（PLC- NN）, series life cycle neural network（SLCNN）. PLCNN uses life cycle model and neural network to predict, then combines the results. SLCNN uses life cycle model to predict the tendency item of the river annual runoff, then neural network is involved in modifying the results, The above two life cycle neural networks are used to predict annual runoff at Longmen Station of Yellow River with satisfying precision, the results show that life cycle neural network model overmatches the single life cycle model or neural network, therefore annual runoff foreasting based on life cycle neural network model is feasible. At the same time, the new combination models have some value to other hydrological and meteorological series prediction.
出处
《中国农村水利水电》
北大核心
2011年第12期15-18,共4页
China Rural Water and Hydropower
关键词
径流预测
生命旋回模型
神经网络
混合模型
预测
runoff predietion
life cycle model
neural network
combination model
forecast
