表征真空玻璃熱性能的最重要參數(shù)-傳熱系數(shù)很難在線測量,因為它會隨著時間的推移而增加,從而降低隔熱性能。確定真空的熱傳遞需要詳細了解它們不同元素的熱特性,這一領域存在一系列標準和指南。總體的熱性能既可以通過詳細的二維數(shù)值方法確定,也可以通過符合歐洲或國際標準的測量來確定。首先,我們研究基于真空玻璃傳熱機理的軟測量智能建模方法,以獲取真空玻璃性能數(shù)據(jù)。該方法保證了智能建模的可行性,為基于軟測量的智能建模預測真空玻璃隔熱性能參數(shù)提供了理論依據(jù)。研究并開發(fā)了一種有效的方法來模擬通過真空玻璃的傳熱�；谙冗M的數(shù)值模擬技術,利用計算流體動力學軟件對傳熱過程進行了分析,并將仿真結果用于指導和分析非穩(wěn)態(tài)測試方法。這種方法保證了加熱板的中心進行一維傳熱,非受熱面中心的溫度測量具有實際意義,對于研究智能化保溫性能建模和預測是必要的。其次,我們應用神經(jīng)網(wǎng)絡方法對真空玻璃的傳熱系數(shù)進行了預測�；贛ATLAB軟件,建立了神經(jīng)網(wǎng)絡智能模型,并對傳統(tǒng)的BP神經(jīng)網(wǎng)絡進行了優(yōu)化。采用遺傳算法對自變量進行降維。然后,利用思維進化計算算法對初始權值和閾值進行優(yōu)化。利用優(yōu)化后的BP神經(jīng)網(wǎng)絡智能模型對真空玻璃隔熱層傳熱系數(shù)進... 

【文章來源】：海南大學海南省 211工程院校

【文章頁數(shù)】：183 頁

【學位級別】：博士

【文章目錄】：
摘要
Abstract
Nomenclature
Chapter 1 Introduction general
    1.1 Introduction
    1.2 Research Background
    1.3 Research heat transfer coefficients of VG and development trends in China and abroad
        1.3.1 Study performance prediction of vacuum glass insulation at Home and abroad
        1.3.2 Main content of research of the heat transfer coefficients of vacuum glass insulation
    1.4 Conclusion
Chapter 2 Mechanism of Vacuum Glass Heat Transfer by Soft Sensor Intelligent Modelling
    2.1 Introduction
    2.2 Data preprocessing
        2.2.1 Implementation of a model of Soft Sensor intelligent modelling
        2.2.2 Online correction of the smart sensor model intelligent modeling
        2.2.3 Modeling methods
    2.3 Selection of historical data
        2.3.1 Pretreatment and transformation of data
        2.3.2 Cleaning and reduction of data
    2.4 Reduction of dimensionality
    2.5 Model selection,training and validation of Prediction coefficient heat transfer VG
        2.5.1 Model training soft sensor intelligent for heat transfer coefficient prediction
        2.5.2 Validation of the model
    2.6 Flexible sensor applications
        2.6.1 Data for flexible sensor modeling
        2.6.2 Modeling approaches
    2.7 White box templates
    2.8 Mathematical modeling equation white boxes on heat transfer coefficient vacuum glass
        2.8.1 Vacuum glass heat transfer principle
        2.8.2 Main factors affecting insulation performance parameters
        2.8.3 Principles of Soft Sensor Intelligent Modeling Technology
        2.8.4 Auxiliary variable selection
    2.9 Gray Box Controller Model
        2.9.1 Data-Driven Modeling Soft Sensor Intelligent Modeling
    2.10 Black box model
    2.11 Feasibility Analysis of the SS Intelligent Modeling of Vacuum Glass
    2.12 Summary of this chapter
Chapter 3 Computational fluid-dynamics-based simulation of heat transfer through vacuum glass
    3.1 Introduction
    3.2 Application of vacuum glass
    3.3 Heat transfer coefficient of vacuum glass
    3.4 Research significance,main content,and innovation points
    3.5 Method of heat transfer in vacuum glass
    3.6 CFD-based vacuum glass heat transfer simulation
    3.7 Discussion
        3.7.1 Steady heat transfer
        3.7.2 Unsteady heat transfer
    3.8.Grid-independent modelling of heat transfer
        3.8.1 Mathematical model
        3.8.2 Numerical methods
    3.9.Vacuum glass thermal property parameters non-steady-state test principle
        3.9.1 Unsteady measuring device
        3.9.2 Analysis of factors affecting the accuracy of non-steady-state measuring devices
        3.9.3 Selection and design of hardware components for temperature measurement system
        3.9.4 System software program
        3.9.5 Physical system
    3.10.Conclusion
Chapter 4 Intelligent modelling to predict heat transfer coefficient of vacuum glass insulation based on thinking evolutionary neural network
    4.1 Introduction
    4.2 Thermal formulation analytical modelling approach
    4.3 Artificial NN(ANN)structure and methodology
    4.4 BPNN model
        4.4.1 BP algorithm
        4.4.2 BPNN algorithm
        4.4.3 Learning process of BPNN algorithms
    4.5 Prediction model of Vacuum Glass insulation performance based on BPNN
        4.5.1 Variable dimension reduction
        4.5.2 Optimisation of initial weights and thresholds
        4.5.3 Simulation results
        4.5.4 Measurements and error analysis
    4.6 Conclusions
Chapter 5 Performance Monitoring of Vacuum Glazing Based on LSSVM
    5.1.Introduction and motivation
    5.2 PCA-RBFNN and Establishing an RBF Neural Network Model
        5.2.1 RBF Neural Network Design
        5.2.2 Analysis of simulation results
        5.2.3 Comparison of PCA and Non-PCA Neural Network Models
    5.3 Vector machine supports least squares
        5.3.1 Prediction modeling Based on LSSVM
        5.3.2 Selection of modeling variables
        5.3.3 Model establishment
        5.3.4 Selecting the parameters of the LSSVM model
        5.3.5 Comparison of modeling methods
    5.4 Conclusion
Chapter 6 Predicting the lifetime of vacuum glass based on fuzzy
    6.1 Introduction
    6.2 Fuzzy Sets,Numbers and Operations
    6.3 Determination of fuzzy regression parameter
        6.3.1 Fuzzy linear regression
        6.3.2 Direct estimation of the parameters
        6.3.3 Fuzzy failure probabilities
        6.3.4 Fuzzy degradation analysis
        6.3.5 Parameter Estimation,Two-Stage Least-Squares
        6.3.6 Maximum Likelihood
        6.3.7 Bayesian Approach
        6.3.8 Estimation of and Prediction from Failure Time Distributions
    6.4 Predicting the lifetime service of vacuum glass based on fuzzy
        6.4.1 Vacuum glass degradation analysis
        6.4.2 Detecting Degraded Data
        6.4.3 Application Fuzzy degradation analysis
    6.5 Fuzzy set theory to predict the probability of lifetime Vacuum Glass
        6.5.1 Failure Time Distribution
Chapter 7 Conclusions general
    7.1 Conclusions
    7.2 Future prospects
Reference
Acknowledgements
Research achievements during the Ph D
致謝


【參考文獻】：
期刊論文
[1]Adoption of wide-bandgap microcrystalline silicon oxide and dual buffers for semitransparent solar cells in building-integrated photovoltaic window system[J]. Johwa Yang,Hyunjin Jo,Soo-Won Choi,Dong-Won Kang,Jung-Dae Kwon.  Journal of Materials Science & Technology. 2019(08)
[2]基于模糊信息粒化和LSSVM真空玻璃保溫性能預測研究[J]. 張亮,王磊,王元麒,李益紅,譚毓銀,宋浩.  廣西大學學報(自然科學版). 2017(06)
[3]基于思維進化優(yōu)化灰色神經(jīng)網(wǎng)絡在真空玻璃需求中的研究[J]. 杜萍,王磊,王元麒.  真空. 2016(05)
[4]基于多退化量的動量輪剩余壽命預測方法[J]. 劉勝南,陸寧云,程月華,姜斌,邢琰.  南京航空航天大學學報. 2015(03)
[5]基于LSSVM的真空玻璃傳熱過程建模[J]. 王元麒,王磊,李桂香.  真空. 2015(02)
[6]基于局部PLS的多輸出過程自適應軟測量建模方法（英文）[J]. 邵偉明,田學民,王平.  Chinese Journal of Chemical Engineering. 2014(07)
[7]真空玻璃技術與應用分析[J]. 王元麒,王磊,李繼定.  真空科學與技術學報. 2012(12)
[8]真空玻璃真空度在線檢測技術與應用[J]. 劉小根,包亦望,宋一樂,龐世紅,邱巖.  鄭州大學學報(工學版). 2009(01)
[9]Matlab遺傳算法優(yōu)化工具箱（GAOT）的研究與應用[J]. 周正武,丁同梅,田毅紅,王曉峰.  機械研究與應用. 2006(06)
[10]統(tǒng)計模式識別中的維數(shù)削減與低損降維[J]. 宋楓溪,高秀梅,劉樹海,楊靜宇.  計算機學報. 2005(11)



本文編號：3545113