数据挖掘技术在水库调度中的研究与应用

水文预报对于水库调度、洪水控制、发电、灌溉等工作至关重要。从目前的研究情况看，各种传统的径流预报方法用于中长期径流预报时，预报精度都比较低。本文首先介绍了数据挖掘的基本知识，并将基于神经网络的数据挖掘方法应用于水文预报中，使用boosting回归算法建立数据挖掘模型，进而发现和挖掘水库调度的内在规律和潜在效能。

Study and Application of Data Mining on Reservoir Scheduling

Ability to properly forecast the natural inflow is important to reservoir scheduling,flood controlling,generating and so on.In recent research,all kinds of traditional runoff forecast methods have low precision when applying to mid and long term runoff forecasting.This paper presents the principle of data mining and the application of it based on neural network to hydrological forecasting.We use the boosting regression algorithm to build the data mining model,and discover the inherent discipline and the latent efficiency between data mining and reservoir scheduling.