數(shù)據(jù)挖掘技術(shù)及其在徑流預(yù)報中的研究與應(yīng)用.pdf

1、華中科技大學(xué)碩士學(xué)位論文數(shù)據(jù)挖掘技術(shù)及其在徑流預(yù)報中的研究與應(yīng)用姓名：許惠君申請學(xué)位級別：碩士專業(yè)：水利水電工程指導(dǎo)教師：張勇傳20060512IIAbatract Data Mining is a newly developing technology based on machine study in artificial intelligence and database. it's an advanced process

2、 which is obtained from abundant of incomplete, noised, fuzzy and stochastic real data, and it's concealed, unknown, believable, novel, efficiency and comprehensible pattern to us. Applications of using data mining t

3、echnology in hydrological forecasting will be more widely applied and makes more practical value. After describing the current research situation of data mining and hydrological forecasting, this paper discuss the archit

4、ecture model of hydrological forecasting system, and then the thesis puts emphasis on studying of runoff forecasting which based on the technology of data mining. This article has six chapters: Chapter 1 Mainly summarize

5、s the content and significance of this article, the theory of the data mining technology as well as its development. Chapter 2 introduces the current research situation of hydrological forecasting and emphasizes on the

6、 forecasting method of neural network. Then this article analyzes and compares all kinds of methods for the hydrological forecasting The design process of long term runoff forecasting is presented in chapter 3, particula

7、rly the choosing strategy of the neural network structure, the method of dealing with original data and the studying and forecasting process of runoff forecasting model which adopts the BP algorithm. Chapter 4 introduce

8、s the design and realization of database of the forecasting model. Chapter 5 introduces the forecasting effect of the long term runoff forecasting model. Comparing with common forecasting model, it is proved that this mo

9、del has better precision, and makes more practical value. Chapter 6 sums up the paper and puts forward the working plan in the future. Key words: Data Mining Hydrological Forecasting Artificial Neural Network BP Al