Dissertation > Excellent graduate degree dissertation topics show

Heteroskcedasticity Two-way Linear Regression and Using to Test Concrete Strength of Building Structure

Author: ZhangXiuWei
Tutor: MaQun；ZuoXiaoJun
School: Hebei University of Technology
Course: Structural Engineering
Keywords: Empirical formula Heteroscedasticity bidirectional linear regression Monte Carlo method Field strength curve
CLC: F224
Type: Master's thesis
Year: 2009
Downloads: 15
Quote: 0
Read: Download Dissertation

Abstract


When field data is less, how to build a high precision engineering empirical formula is an urgent problem . In previous studies , people or historical data by borrowing to expand laboratory sample size , and in the field data and to borrow the same variance under the assumption that the data presented bi- linear regression method , to solve these problems for a useful exploration. However, in practical engineering, usually can not meet the same variance assumption of bidirectional linear regression method has some limitations in the application . After extensive research, the problem for heteroscedasticity different situations , proposes a two-way linear regression heteroscedasticity , standard deviation respectively Relations ( field data of the standard deviation of the ratio of standard deviation to borrow data ) known and unknown situations are discussed, to out of the corresponding point estimates and confidence intervals . In the standard deviation between the known cases, heteroscedasticity bidirectional linear regression method gives an estimate of the empirical formula with no bias and variance of the smallest nature ; standard deviation are unknown in the case , through a large number of Monte Carlo (Monte Carlo) test proved heteroscedasticity bidirectional linear regression method is not only superior to the traditional methods, but with two-way linear regression method, the true value is closer to the parent . In the standard deviation between the known cases, the empirical formula is given interval estimation method ; standard deviation in the case of relationship is not perfectly known empirical formula gives the approximate confidence interval estimation methods . The bi- linear regression heteroscedasticity applied to the detection of structural concrete strength by rebound method , in the field or laboratory data and historical data variance under different circumstances , effectively solved because the site can not be established with less data and high precision on-site strength curve issue. Meanwhile, this method is also applicable to other areas of similar problems.

Related Dissertations

  1. Transient Thermal Analysis of Multi-Body System with Complicated Boundary Conditions,TK124
  2. American Option Pricing under Stochastic Market Option Model Based on Dividend and Treatment Fees,O211.6
  3. Computation on Fuzzy Riemann Integral of Type Ⅱ and Order Structure on Ordered Weighted Geometric Operators,O177
  4. Deformationand Strength Characteristics of Glass Fiber-Reinforced Soil Studied by Experiments,TU521.3
  5. MCP face electron reflection on the impact of image intensifier resolution,TN144
  6. Research on Computer Aided Tolerance Optimization in Aircraft Manufacturing,V262
  7. Study on Fast External Dose Estimation Method in Nuclear and Radiological Event with Software Development,R144
  8. Research on the Inversion Method of Aerosol Materials’ Optical Constants,O484.41
  9. The Monte Carlo Study of Polarized Light Propagation in Atmosphere: Simulation and Application,O436.3
  10. Monte Carlo program point energy nuclear data processing method research and Optimization,TL329.2
  11. Study on Comprehensive Treatment Plan of Lightning Disturbance for the Operating Overhead Transmission Line,TM863
  12. Stochastic Heat Transfer Analysis on Buried Pipe of Ground Source Heat Pump with Monte-Carlo Method,TU831
  13. Analysis of Elevator Traffic Flow Based on Five Tipical Periods,TU857
  14. Stochastic Optimal Control of Nonlinear Suspension of Quarter Car Model,U463.33
  15. Research on DOA Estimation and Tracking Approach Based on Monte Carol Method,TN911.7
  16. Option Pricing Under Stochastic Crashes,F830.9
  17. Monte Carlo Simulations on Magnetic Properties of Disordered Nanoparticle Films,O484.43
  18. Monte Carlo Simulation on the Magnetic Properties of a Honeycomb Monolayer Film,O482.5
  19. Monte Carlo Simulations of Magnetic Properties of Spin Frustrated Systems,O469
  20. Generalized impulse response model and its asymmetric effect of monetary policy analysis,F822.0

CLC: > Economic > Economic planning and management > Economic calculation, economic and mathematical methods > Economic and mathematical methods
© 2012 www.DissertationTopic.Net  Mobile