首页--工业技术论文--自动化技术、计算机技术论文--自动化基础理论论文--人工智能理论论文

Health Research of Wheel-Set Based on Data Mining

Abstract第5-6页
1. Introduction第9-18页
    1.1. Research purpose and significance第9-11页
    1.2. Research status第11-16页
        1.2.1. Research status of health assessment第11-13页
        1.2.2. Research status of data mining第13-16页
    1.3. Research content and the structure of this thesis第16-18页
2. The basic theory of wheel-set health research第18-32页
    2.1. Wheel-set basic knowledge第18-19页
    2.2. Gray relational analysis第19-22页
    2.3. ARIMA model第22-25页
        2.3.1. The basic concept of ARIMA第22-24页
        2.3.2. Basic steps of ARIMA modeling第24-25页
    2.4. Neural network第25-31页
        2.4.1. Basic concepts of neural network第25-27页
        2.4.2. BP neural network第27-28页
        2.4.3. Basic steps of BP neural modeling第28-31页
    2.5. Summary第31-32页
3. Analysis on the wear law of wheel shape第32-47页
    3.1. Introduction of wheel profile第32-33页
    3.2. Introduction of wheel profile measurement system第33-34页
    3.3. Trend rule of wheel diameter, tread profile and other parameters第34-42页
        3.3.1. Wheel diameter第34-37页
        3.3.2. Tread wear第37-38页
        3.3.3. Flange thickness第38-39页
        3.3.4. Difference of coaxial wheel diameter第39-41页
        3.3.5. QR value第41-42页
    3.4. Numerical analysis among the geometric parameters of wheel-set第42-45页
        3.4.1. Data standardization第42-43页
        3.4.2. Results of correlation analysis第43-44页
        3.4.3. Regression analysis第44-45页
    3.5. Summary第45-47页
4. Prediction model based on historical data of wheel-set geometry parameters第47-62页
    4.1. Data preprocessing第47-48页
    4.2. Modeling and analysis of ARIMA第48-53页
        4.2.1. Modeling of prediction based on ARIMA第48-50页
        4.2.2. Analysis of ARIMA prediction results第50-53页
    4.3. Modeling and analysis of BP neural network第53-59页
        4.3.1. Prediction modeling of wheel-set parameters based on BP neural network第53-55页
        4.3.2. Analysis of prediction results based on BP neural network第55-59页
    4.4. Comparative analysis of prediction results based on ARIMA and BP neural network第59-61页
    4.5. Summary第61-62页
Conclusion and Outlook第62-64页
Acknowledgement第64-65页
Reference第65-70页
Article and result published during the master第70页

论文共70页,点击 下载论文
上一篇:基于柔性阵列压力传感器的坐姿监测与提醒系统的设计与实现
下一篇:基于支持向量机的花生褐斑病图像识别