Dissertation > Excellent graduate degree dissertation topics show

Response Surface Modeling by Local Kernel Partial Least Squares

Author: LiuYu
Tutor: LuoGuiMing
School: Tsinghua University
Course: Software Engineering
Keywords: response surface modeling partial least squares local modeling kernel function prediction
CLC: TP301.6
Type: Master's thesis
Year: 2013
Downloads: 7
Quote: 0
Read: Download Dissertation

Abstract


Response surface modeling (RSM) is often used to analyze the relationshipbetween multiple variables. RSM samples the data through desigh of experiment, andestimates parameters by least squares. However, small sized samples and multiplecorrelations could lead to multi-collinearity problem. Because of this problem, theaccuracy and the reliability of model would not be guaranteed. Recenly, Partial LeastSquares (PLS) has been introduced to build the response surface for this problem.However, PLS is a linear method, which is not capable to deal with the non-linearmodel. Therefore, the purpose of this paper is to establish the nonlinear model. Insummary, this paper has the following main contributions:To solve this problem, we propose local modeling. In local modeling, we alsodefine a new performance criterion, which is used to estimate the parameters of eachsub-model. To adjust the influence of experimental point on each sub-space, weintroduce a weight function matrix in the new performance criterion.Based on local modeling, we propose local partial least squares (LPLS). Thisapproach aim to achieve the global nonlinear by local linear, and select the model byusing the cross-validation. LPLS can successfully solve multi-collinearity problems.However, the local space presents linear characteristics in this method. In addition,LPLS, KPLS may lead to "extreme value missing phenomenon".To avoid the "extreme value missing phenomenon", we propose local kernel partialleast squares (LKPLS). This approach maps the data in the original space into a featurekernel space by using the kernel function in the reproducing kernel Hilbert space.LKPLS builds a non-linear model in the local area. Except building a nonlinear PLSmodel directly, this approach also builds a nonlinear transformation for original data.At last, we examine the new approaches in several experiments to verify theproposed method. Besides, the latter experiments also verify the stability of the newalgorithm on small data sets. Moreover, the results show the proposed method workswell when occurring to"extreme value missing phenomenon." Finally, the new methodshave been applied in the research project of Tsinghua&Mitsubishi Heavy Industries.

Related Dissertations

  1. Research on Graph-Based Algorithm for Tagsnps Selection,Q78
  2. The Brake Performance of Hydraulic Retarder and Simualtion Research on Its Application for Aircraft Arrestment,TH137.331
  3. Modulation Pattern Recognition and Parameter Estimation for Ground Wave Radiation Sources,TN957.51
  4. Process Support Vector Machine and Its Application to Satellite Thermal Equilibrium Temperature Prediction,TP183
  5. The Study on Metallogenic Regularities and Metallogenic Prediction of Nanpo Copper Deposit in Mengla County, Yunnan,P618.41
  6. Predicting Wheat Grain Yield and Quality Based on Population Indexes and Nitrogen Nutrient Status,S512.1
  7. Study on Growth Predicting Technique Based on Integration of Remotely Sensed Information and Crop Model in Rice,S511
  8. Genetic Dissection and Elite Allele Identification of Seed Traits in Soybean Cultivars Released from Huanghuai Valleys and Southern China,S565.1
  9. A Photo-Thermal Model for Predicting Growth and External Quality of Dendrobium Nobile in Greenhouse,S682.31
  10. Studies on Changes of Quality Characters and Prediction Model of Postharvest Tomato Fruit,S641.2
  11. Studies on Prediction Models for Fruit Decay and Shelf-Life of Postharvest Chinese Bayberry,S667.6
  12. The Research of the Relationship between Psychological Qualities of Students of Teachers’college and Their Adaptation to College,B844.2
  13. Preliminary Research for Application and Improvement of Essen Stroke Risk Score,R743.3
  14. Adaptive Adjustment of fire emergency plan,X928.7
  15. Zhaoguan Lower Coal Group water inrush prediction and control techniques,TD745
  16. Research of the Influencing Factors on Constrained Concrete Flexural Member Filled with Steel Tube,TU398.9
  17. Prognosis and Regularity of Metallogeny in the Hedong Gold Ore Deposits of the Northwest Jiaodong,P618.51
  18. Research and Application of Gas Emission Prediction Based on LS-SVM,TD712.5
  19. Research on Vibration and Noise of Electric Tool,TB535
  20. Research of Forecast Method in Fire Danger Rating and Weather Situation of HeiLongJiang Province,S762
  21. Research on Two-Dimensional Non-contact Measurement System of Yong Men’s Bodies Based on Digital Images,TP391.41

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > General issues > Theories, methods > Algorithm Theory
© 2012 www.DissertationTopic.Net  Mobile