氣象資料是關(guān)系到國(guó)計(jì)民生的重要基礎(chǔ)資源,社會(huì)經(jīng)濟(jì)發(fā)展、可持續(xù)發(fā)展等各個(gè)領(lǐng)域都需要?dú)庀蠊ぷ魈峁┛煽勘Ｕ�。隨著氣象現(xiàn)代化的進(jìn)步和發(fā)展,氣象行業(yè)積累了大量的數(shù)據(jù),從海量的氣象數(shù)據(jù)中挖掘出有用的信息,對(duì)于氣象預(yù)測(cè)、預(yù)報(bào)和災(zāi)害預(yù)警都扮演著至關(guān)重要的作用。隨著人們對(duì)氣象預(yù)報(bào)精度要求的逐漸提高,提高預(yù)報(bào)準(zhǔn)確率成為氣象部門的工作重點(diǎn),要提高數(shù)值天氣預(yù)報(bào)準(zhǔn)確率就必須解決計(jì)算量大和計(jì)算實(shí)時(shí)性問題,這就對(duì)氣象部門的硬件設(shè)施和專業(yè)人才提出了很高的要求,同時(shí)帶來(lái)了硬件成本的急劇增加。近年來(lái),國(guó)內(nèi)外大氣科學(xué)領(lǐng)域?qū)W者基于數(shù)據(jù)挖掘方法的預(yù)報(bào)技術(shù),即人工神經(jīng)網(wǎng)絡(luò)、遺傳算法、支持向量機(jī)、貝葉斯、決策樹和關(guān)聯(lián)規(guī)則挖掘等方法開展了大量研究,并取得了顯著成果,天氣預(yù)報(bào)的準(zhǔn)確率在不斷提高,但是與人們的期望還有距離。因此,如何充分有效的發(fā)揮數(shù)據(jù)挖掘在天氣預(yù)報(bào)中的重要作用,滿足人們的需求至關(guān)重要。隨著氣象數(shù)據(jù)規(guī)模飛速增長(zhǎng),BP神經(jīng)網(wǎng)絡(luò)由于其強(qiáng)大的非線性系統(tǒng)擬合能力,在氣象數(shù)據(jù)的分析和預(yù)測(cè)中得到廣泛應(yīng)用。因此,本研采用BP神經(jīng)網(wǎng)絡(luò)方法,以降水量和氣溫作為輸入因子,建立天氣預(yù)報(bào)模型。對(duì)降水天氣的研究,本研究對(duì)所選的樣本進(jìn)行4種方式的組合...

【文章頁(yè)數(shù)】：53 頁(yè)

【學(xué)位級(jí)別】：碩士

【文章目錄】：
摘要
Abstract
1 Introduction
    1.1 Research Background
    1.2 Research Objective and Significance
        1.2.1 Research Objective
        1.2.2 Research Significance
    1.3 Domestic and O verseas Progress
        1.3.1 The Present Situation of C loud Computing
        1.3.2 The C urrent Situation of Data Mining
        1.3.3 Application of Data Mining Method in Weather Forecast
    1.4 Main Content and Research Methods
        1.4.1 Main Content
        1.4.2 Research Methods
2 Theoretical and Model Analysis
    2.1 The Basic Introduction of Data Mining
        2.1.1 Data Mining Definition
        2.1.2 The Data Mining Process
        2.1.3 Methods of Data Mining
    2.2 Artificial Neural Network
        2.2.1 Overview of the Artificial Neural Network
        2.2.2 BP Algorithm Description
        2.2.3 BP Neural Network Process
    2.3 Summary
3 Set up Weather Forecasts model base on ANN
    3.1 Weather Forecast Analysis and Comparison
        3.1.1 Weather Forecast
        3.1.2 Main Methods of Weather Forecast
    3.2 Application of BP Neural Networks in Weather Forecasting
        3.2.1 Data Sources
        3.2.2 Basic Condition of Sample
    3.3 Set up Forecasting Model
        3.3.1 Perceptron Model
        3.3.2 Construction of the Theoretical Model
4 BP Structure Build
    4.1 Application of BP Neural Networks in Rainfall
        4.1.1 Data Preprocessing
        4.1.2 Training Sample Selection
    4.2 Application of BP Neural Networks in Temperature
        4.2.1 BP Neural Network Design
        4.2.2 Pretreatment of Input and O utput Data
        4.2.3 Simulation of Temperature Sequence
    4.3 BP Building
    4.4 Conclusion
5 Experimenting the Performance
    5.1 Simulation Environment
    5.2 Simulation Result
        5.2.1 Forecasting Model
        5.2.2 Input Data
        5.2.3 Simulation Result
Conclusion
References
Research Projects and Publications in Master Study
Acknowledgement



本文編號(hào)：3820253