Reliability Based Design Optimization. Outline RBDO problem definition Reliability Calculation Transformation from X-space to u-space RBDO Formulations.

Slides:



Advertisements
Similar presentations
Principal Component Analysis Based on L1-Norm Maximization Nojun Kwak IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008.
Advertisements

Lecture 13 L1 , L∞ Norm Problems and Linear Programming
Chapter 9: Vector Differential Calculus Vector Functions of One Variable -- a vector, each component of which is a function of the same variable.
Engineering Optimization
1 OR II GSLM Outline  some terminology  differences between LP and NLP  basic questions in NLP  gradient and Hessian  quadratic form  contour,
1 8. Numerical methods for reliability computations Objectives Learn how to approximate failure probability using Level I, Level II and Level III methods.
Sensitivity Analysis In deterministic analysis, single fixed values (typically, mean values) of representative samples or strength parameters or slope.
Experimental Design, Response Surface Analysis, and Optimization
Optimization of thermal processes2007/2008 Optimization of thermal processes Maciej Marek Czestochowa University of Technology Institute of Thermal Machinery.
Probabilistic Re-Analysis Using Monte Carlo Simulation
May 2007 Hua Fan University, Taipei An Introductory Talk on Reliability Analysis With contribution from Yung Chia HSU Jeen-Shang Lin University of Pittsburgh.
EPIDEMIOLOGY AND BIOSTATISTICS DEPT Esimating Population Value with Hypothesis Testing.
MAE 552 – Heuristic Optimization Lecture 6 February 6, 2002.
Prénom Nom Document Analysis: Data Analysis and Clustering Prof. Rolf Ingold, University of Fribourg Master course, spring semester 2008.
L15:Microarray analysis (Classification). The Biological Problem Two conditions that need to be differentiated, (Have different treatments). EX: ALL (Acute.
1 Reliability and Robustness in Engineering Design Zissimos P. Mourelatos, Associate Prof. Jinghong Liang, Graduate Student Mechanical Engineering Department.
Efficient Methodologies for Reliability Based Design Optimization
Evolutionary Computational Intelligence Lecture 9: Noisy Fitness Ferrante Neri University of Jyväskylä.
Unconstrained Optimization Problem
Planning operation start times for the manufacture of capital products with uncertain processing times and resource constraints D.P. Song, Dr. C.Hicks.
Linear Discriminant Functions Chapter 5 (Duda et al.)
Nonlinear Stochastic Programming by the Monte-Carlo method Lecture 4 Leonidas Sakalauskas Institute of Mathematics and Informatics Vilnius, Lithuania EURO.
1 Efficient Re-Analysis Methodology for Probabilistic Vibration of Large-Scale Structures Efstratios Nikolaidis, Zissimos Mourelatos April 14, 2008.
1 Assessment of Imprecise Reliability Using Efficient Probabilistic Reanalysis Farizal Efstratios Nikolaidis SAE 2007 World Congress.
Stevenson and Ozgur First Edition Introduction to Management Science with Spreadsheets McGraw-Hill/Irwin Copyright © 2007 by The McGraw-Hill Companies,
Component Reliability Analysis
Algebra 2 Chapter 3 Notes Systems of Linear Equalities and Inequalities Algebra 2 Chapter 3 Notes Systems of Linear Equalities and Inequalities.
Algebra 2 Chapter 3 Notes Systems of Linear Equalities and Inequalities Algebra 2 Chapter 3 Notes Systems of Linear Equalities and Inequalities.
Mathematics for Computer Graphics (Appendix A) Won-Ki Jeong.
ENCI 303 Lecture PS-19 Optimization 2
Lecture 12 Statistical Inference (Estimation) Point and Interval estimation By Aziza Munir.
MEGN 537 – Probabilistic Biomechanics Ch.7 – First Order Reliability Methods Anthony J Petrella, PhD.
Some Key Facts About Optimal Solutions (Section 14.1) 14.2–14.16
Engineering Mechanics: Statics
Nonlinear Programming.  A nonlinear program (NLP) is similar to a linear program in that it is composed of an objective function, general constraints,
Ken YoussefiMechanical Engineering Dept. 1 Design Optimization Optimization is a component of design process The design of systems can be formulated as.
Stochastic Linear Programming by Series of Monte-Carlo Estimators Leonidas SAKALAUSKAS Institute of Mathematics&Informatics Vilnius, Lithuania
Module 1: Statistical Issues in Micro simulation Paul Sousa.
ECE 8443 – Pattern Recognition LECTURE 07: MAXIMUM LIKELIHOOD AND BAYESIAN ESTIMATION Objectives: Class-Conditional Density The Multivariate Case General.
7. Reliability based design Objectives Learn formulation of reliability design problem. Understand difference between reliability-based design and deterministic.
Vector Norms and the related Matrix Norms. Properties of a Vector Norm: Euclidean Vector Norm: Riemannian metric:
EASTERN MEDITERRANEAN UNIVERSITY Department of Industrial Engineering Non linear Optimization Spring Instructor: Prof.Dr.Sahand Daneshvar Submited.
Application of the two-step method for the solution of the inverse gravity problem for the Kolárovo anomaly.
Silesian University of Technology in Gliwice Inverse approach for identification of the shrinkage gap thermal resistance in continuous casting of metals.
Reliability-Based Design Methods of Structures
Monte-Carlo method for Two-Stage SLP Lecture 5 Leonidas Sakalauskas Institute of Mathematics and Informatics Vilnius, Lithuania EURO Working Group on Continuous.
A Flexible New Technique for Camera Calibration Zhengyou Zhang Sung Huh CSPS 643 Individual Presentation 1 February 25,
McGraw-Hill/Irwin © The McGraw-Hill Companies, Inc., Table of Contents CD Chapter 14 (Solution Concepts for Linear Programming) Some Key Facts.
Machine Design Under Uncertainty. Outline Uncertainty in mechanical components Why consider uncertainty Basics of uncertainty Uncertainty analysis for.
FORS 8450 Advanced Forest Planning Lecture 6 Threshold Accepting.
Chapter 10 Minimization or Maximization of Functions.
Multi-objective Optimization
3 Components for a Spreadsheet Optimization Problem  There is one cell which can be identified as the Target or Set Cell, the single objective of the.
METHOD OF STEEPEST DESCENT ELE Adaptive Signal Processing1 Week 5.
Diversity Loss in General Estimation of Distribution Algorithms J. L. Shapiro PPSN (Parallel Problem Solving From Nature) ’06 BISCuit 2 nd EDA Seminar.
Optimization in Engineering Design 1 Introduction to Non-Linear Optimization.
Optimization formulation Optimization methods help us find solutions to problems where we seek to find the best of something. This lecture is about how.
The Unscented Particle Filter 2000/09/29 이 시은. Introduction Filtering –estimate the states(parameters or hidden variable) as a set of observations becomes.
INTRO TO OPTIMIZATION MATH-415 Numerical Analysis 1.
CHAPTER 2.3 PROBABILITY DISTRIBUTIONS. 2.3 GAUSSIAN OR NORMAL ERROR DISTRIBUTION  The Gaussian distribution is an approximation to the binomial distribution.
Chapter 4 Statistical Inference  Estimation -Confidence interval estimation for mean and proportion -Determining sample size  Hypothesis Testing -Test.
1 Introduction Optimization: Produce best quality of life with the available resources Engineering design optimization: Find the best system that satisfies.
Fundamentals of Data Analysis Lecture 11 Methods of parametric estimation.
Boundary Element Analysis of Systems Using Interval Methods
Chapter 3 Component Reliability Analysis of Structures.
MEGN 537 – Probabilistic Biomechanics Ch
Statistical Learning Dong Liu Dept. EEIS, USTC.
Dr. Arslan Ornek IMPROVING SEARCH
Where did we stop? The Bayes decision rule guarantees an optimal classification… … But it requires the knowledge of P(ci|x) (or p(x|ci) and P(ci)) We.
Dr. Arslan Ornek MATHEMATICAL MODELS
Presentation transcript:

Reliability Based Design Optimization

Outline RBDO problem definition Reliability Calculation Transformation from X-space to u-space RBDO Formulations Methods for solving inner loop (RIA, PMA) Methods of MPP estimation

Terminologies X : vector of uncertain variables η : vector of certain variables Θ : vector of distribution parameters of uncertain variable X( means, s.d.) d : consists of θ and η whose values can be changed p : consists of θ and η whose values can not be changed

Terminologies(contd..) Soft constraint: depends upon η only. Hard constraint: depends upon both X(θ) and η [θ,η] = [d,p] Reliability = 1 – probability of failure

RBDO problem Optimization problem min F (X,η) objective f i (η) > 0 g j (X, η ) > 0 RBDO formulation min F (d,p) objective f i (d,p) > 0 soft constraints P (g j (d,p ) > 0) > P t hard constraints

Comparison b/w RBDO and Deterministic Optimization Deterministic Optimum Reliability Based Optimum Feasible Region

Basic reliability problem

Probability of failure Reliabilty Calculation

Reliability index Reliability Index

Formulation of structural reliability problem Vector of basic random variables represents basic uncertain quantities that define the state of the structure, e.g., loads, material property constants, member sizes. Limit state function Safe domain Failure domain Limit state surface

Geometrical interpretation uR uR failure domain   f   S safe domain uS uS 0 limit state surface Transformation to the standard normal space Distance from the origin [ u R, u S ] to the linear limit state surface Cornell reliability index

Hasofer-Lind reliability index Lack of invariance, characteristic for the Cornell reliability index, can be resolved by expanding the Taylor series around a point on the limit state surface. Since alternative formulation of the limit state function correspond to the same surface, the linearization remains invariant of the formulation. The point chosen for the linearization is one which has the minimum distance from the origin in the space of transformed standard random variables. The point is known as the design point or most probable point since it has the highest likelihood among all points in the failure domain.

For the linear limit state function, the absolute value of the reliability index, defined as, is equal to the distance from the origin of the space (standard normal space) to the limit state surface. Geometrical interpretation

Hasofer-Lind reliability index

RBDO formulations RBDO Methods Double Loop Decoupled Single Loop

Double loop Method Objective function Reliability Evaluation For 1 st constraint Reliability Evaluation For m th constraint

Decoupled method (SORA ) Deterministic optimization loop Objective function : min F(d,µ x ) Subject to : f(d,µ x ) < 0 g(d,p,µ x -s i, ) < 0 Inverse reliability analysis for Each limit state dk,µxkdk,µxk k = k+1 s i = µ x k – x k mpp x k mpp,p mpp

Single Loop Method Lower level loop does not exist. min { F(µ x ) } f i (µ x ) ≤ 0 deterministic constraints g i (x) ≥ 0 where x - µ x = -β t *α*σ α=grad(g u (d,x))/||grad(g u (d,x))|| µ xl ≤ µ x ≤ µ xu

Inner Level Optimization (Checking Reliability Constraints) Reliability Index Approach(RIA) min ||u|| subject to g i (u,µ x )=0 if min ||u|| >β t (feasible) Performance Measure Approach(PMA ) min g i ( u,µ x ) subject to ||u|| = β t If g(u*, µ x )>0(feasible)

Most Probable Point(MPP) The probability of failure is maximum corresponding to the mpp. For the PMA approach, -grad(g) at mpp is parallel to the vector from the origin to that point. MPP lies on the β-circle for PMA approach and on the curve boundary in RIA approach. Exact MPP calculation is an optimization problem. MPP esimation methods have been developed.

MPP estimation inactive constraint active constraint RIA MPP PMA MPP RIA MPP U Space

Methods for reliability computation First Order Reliability Method (FORM) Second Order Reliability Method (SORM) Simulation methods: Monte Carlo, Importance Sampling Numerical computation of the integral in definition for large number of random variables ( n > 5) is extremely difficult or even impossible. In practice, for the probability of failure assessment the following methods are employed:

FORM – First Order Reliability Method

SORM – Second Order Reliability Method

Gradient Based Method for finding MPP find α = -grad(u k )/||grad(u k )|| u k+1 =β t * α If |u k+1 -u k |<ε, stop u k+1 is the mpp point else goto start If g(u k+1 )>g(u k ), then perform an arc search which is a uni-directional optimization

Abdo-Rackwitz-Fiessler algorithm Rackwitz-Fiessler iteration formula find subject to Gradient vector in the standard space:

where is a constant < 1, is the other indication of the point in the RF formula. for every and Convergence criterion Very often to improve the effectiveness of the RF algorithm the line search procedure is employed Merit function proposed by Abdo Abdo-Rackwitz-Fiessler algorithm

Alternate Problem Model solution to : min ‘f’ s.t atleast 1 of the reliability constraint is exactly tangent to the beta circle and all others are satisfied. Assumptions: minimum of f occurs at the aforesaid point

Alternate Problem Model Reliability based optimum β1β1 β2β2 x1 x2

Scope for Future Research Developing computationally inexpensive models to solve RBDO problem The methods developed thus far are not sufficiently accurate Including robustness along with reliability Developing exact methods to calculate probability of failure