WB1440 Engineering Optimization – Concepts and Applications Engineering Optimization Concepts and Applications Fred van Keulen Matthijs Langelaar CLA H21.1.

Slides:



Advertisements
Similar presentations
Engineering Optimization
Advertisements

Kriging.
Cost of surrogates In linear regression, the process of fitting involves solving a set of linear equations once. For moving least squares, we need to form.
Pattern Recognition and Machine Learning: Kernel Methods.
CSCI 347 / CS 4206: Data Mining Module 07: Implementations Topic 03: Linear Models.
Cost of surrogates In linear regression, the process of fitting involves solving a set of linear equations once. For moving least squares, we need to.
Local surrogates To model a complex wavy function we need a lot of data. Modeling a wavy function with high order polynomials is inherently ill-conditioned.
Curve fit metrics When we fit a curve to data we ask: –What is the error metric for the best fit? –What is more accurate, the data or the fit? This lecture.
Data mining and statistical learning - lecture 6
CITS2401 Computer Analysis & Visualisation
The loss function, the normal equation,
Classification and Prediction: Regression Via Gradient Descent Optimization Bamshad Mobasher DePaul University.
Kernel methods - overview
x – independent variable (input)
Radial Basis Functions
Engineering Optimization
Curve-Fitting Regression
Spatial Interpolation
Engineering Optimization
Linear and generalised linear models
Hazırlayan NEURAL NETWORKS Radial Basis Function Networks I PROF. DR. YUSUF OYSAL.
Linear and generalised linear models
Self-organizing Learning Array based Value System — SOLAR-V Yinyin Liu EE690 Ohio University Spring 2005.
Classification and Prediction: Regression Analysis
Radial Basis Function Networks
Chapter 6-2 Radial Basis Function Networks 1. Topics Basis Functions Radial Basis Functions Gaussian Basis Functions Nadaraya Watson Kernel Regression.
Radial Basis Function Networks
Gaussian process modelling
Section 2: Finite Element Analysis Theory
Chapter 9 Neural Network.
Ch4 Describing Relationships Between Variables. Pressure.
Outline 1-D regression Least-squares Regression Non-iterative Least-squares Regression Basis Functions Overfitting Validation 2.
ECE 8443 – Pattern Recognition ECE 8527 – Introduction to Machine Learning and Pattern Recognition LECTURE 16: NEURAL NETWORKS Objectives: Feedforward.
Local surrogates To model a complex wavy function we need a lot of data. Modeling a wavy function with high order polynomials is inherently ill-conditioned.
Jeff Howbert Introduction to Machine Learning Winter Regression Linear Regression.
Geographic Information Science
Model Construction: interpolation techniques 1392.
CSC321: Neural Networks Lecture 2: Learning with linear neurons Geoffrey Hinton.
Curve-Fitting Regression
GEOSTATISICAL ANALYSIS Course: Special Topics in Remote Sensing & GIS Mirza Muhammad Waqar Contact: EXT:2257.
Digital Image Processing Lecture 10: Image Restoration March 28, 2005 Prof. Charlene Tsai.
PROCESS MODELLING AND MODEL ANALYSIS © CAPE Centre, The University of Queensland Hungarian Academy of Sciences Statistical Model Calibration and Validation.
1 Using Multiple Surrogates for Metamodeling Raphael T. Haftka (and Felipe A. C. Viana University of Florida.
5-1 ANSYS, Inc. Proprietary © 2009 ANSYS, Inc. All rights reserved. May 28, 2009 Inventory # Chapter 5 Six Sigma.
Digital Image Processing Lecture 10: Image Restoration
Review of fundamental 1 Data mining in 1D: curve fitting by LLS Approximation-generalization tradeoff First homework assignment.
Lecture 6: Point Interpolation
Neural Networks Presented by M. Abbasi Course lecturer: Dr.Tohidkhah.
Tutorial 2, Part 2: Calibration of a damped oscillator.
Neural Networks The Elements of Statistical Learning, Chapter 12 Presented by Nick Rizzolo.
Curve fit metrics When we fit a curve to data we ask: –What is the error metric for the best fit? –What is more accurate, the data or the fit? This lecture.
Environmental Data Analysis with MatLab 2 nd Edition Lecture 22: Linear Approximations and Non Linear Least Squares.
Introduction to emulators Tony O’Hagan University of Sheffield.
CPH Dr. Charnigo Chap. 11 Notes Figure 11.2 provides a diagram which shows, at a glance, what a neural network does. Inputs X 1, X 2,.., X P are.
Data Mining: Concepts and Techniques1 Prediction Prediction vs. classification Classification predicts categorical class label Prediction predicts continuous-valued.
Kriging - Introduction Method invented in the 1950s by South African geologist Daniel Krige (1919-) for predicting distribution of minerals. Became very.
1 C.A.L. Bailer-Jones. Machine Learning. Data exploration and dimensionality reduction Machine learning, pattern recognition and statistical data modelling.
CSC2535: Computation in Neural Networks Lecture 11 Extracting coherent properties by maximizing mutual information across space or time Geoffrey Hinton.
A Simple Artificial Neuron
Spatial Analysis Longley et al..
CHAPTER 29: Multiple Regression*
Collaborative Filtering Matrix Factorization Approach
Filtering and State Estimation: Basic Concepts
Curve fit metrics When we fit a curve to data we ask:
Chapter 8: Generalization and Function Approximation
Curve fit metrics When we fit a curve to data we ask:
The loss function, the normal equation,
Mathematical Foundations of BME Reza Shadmehr
Machine learning overview
Presentation transcript:

WB1440 Engineering Optimization – Concepts and Applications Engineering Optimization Concepts and Applications Fred van Keulen Matthijs Langelaar CLA H21.1

WB1440 Engineering Optimization – Concepts and Applications Contents ● Optimization problem checking and simplification ● Model simplification

WB1440 Engineering Optimization – Concepts and Applications Model simplification ● Basic idea: Expensive model Optimizer Cheap model Optimizer ● Motivation: – Replacement of expensive function, evaluated many times – Interaction between different disciplines – Estimation of derivatives – Noise

WB1440 Engineering Optimization – Concepts and Applications Model simplification (2) ● Drawback: loss of accuracy ● Different ranges: local, mid-range, global ● Synonyms: – Approximation models – Metamodels – Surrogate models – Compact models – Reduced order models Extract information Construct approximation Procedure:

WB1440 Engineering Optimization – Concepts and Applications Model simplification (3) ● Information extraction: linked to techniques from physical experiments: “plan of experiments” / DoE ● Many approaches! Covered here: – Taylor series expansions – Exact fitting – Least squares fitting (response surface techniques) – Kriging – Reduced basis methods – Briefly: neural nets, genetic programming, simplified physical models ● Crucial: purpose, range and level of detail

WB1440 Engineering Optimization – Concepts and Applications Taylor series expansions ● Approximation based on local information: Truncation error! ● Use of derivative information! ● Valid in neighbourhood of x

WB1440 Engineering Optimization – Concepts and Applications Taylor approximation example 1 st order 2 nd order 3 rd order 4 th order5 th order 20 th order Function Approximation (x = 20) x

WB1440 Engineering Optimization – Concepts and Applications Exact fitting (interpolation) ● # datapoints = # fitting parameters ● Every datapoint reproduced exactly ● Example: x1x1 x2x2 f2f2 f1f1

WB1440 Engineering Optimization – Concepts and Applications Exact fitting (2) ● Easy for intrinsically linear functions: ● No smoothing / filtering / noise reduction ● Danger of oscillations with high-order polynomials ● Often used: polynomials, generalized polynomials:

WB1440 Engineering Optimization – Concepts and Applications 9 th order polynomial Oscillations ● Referred to as “Runge phenomenon” ● In practice: use order 6 or less 5 th order 9 th order

WB1440 Engineering Optimization – Concepts and Applications Link between FD and exact fitting ● Example: 2 nd order polynomial fit: xx+hx-h

WB1440 Engineering Optimization – Concepts and Applications Least squares fitting ● Less fitting parameters than datapoints ● Smoothing / filtering behaviour ● “Best fit”? Minimize sum of deviations: ● “Best fit”? Minimize sum of squared deviations: x f

WB1440 Engineering Optimization – Concepts and Applications Least squares fitting (2) ● Choose fitting function linear in parameters a i : ● Short notation:

WB1440 Engineering Optimization – Concepts and Applications LS fitting (3) ● Minimize sum of squared errors: (Optimization problem!)

WB1440 Engineering Optimization – Concepts and Applications Polynomial LS fitting ● Polynomial of degree m :

WB1440 Engineering Optimization – Concepts and Applications Polynomial LS example samples quadratic 6th degree

WB1440 Engineering Optimization – Concepts and Applications Multidimensional LS fitting ● Polynomial in multiple dimensions: ● Number of coefficients a i for quadratic polynomial in R n : Curse of dimensionality!

WB1440 Engineering Optimization – Concepts and Applications Fractional factorial design Response surface ● Generate datapoints through sampling: – Generate design points through Design of Experiments – Evaluate responses ● Fit analytical model ● Check accuracy 2 n full factorial design x1x1 x2x2 x3x3

WB1440 Engineering Optimization – Concepts and Applications Latin Hypercube Sampling (LHS) ● Popular method: LHS ● Based on idea of Latin square: ● Properties: – Space-filling – Any number of design points – Intended for box-like domains – Matlab: lhsdesign

WB1440 Engineering Optimization – Concepts and Applications (LS) Fit quality indicators ● Accuracy? More / fewer terms? ● Examine the residuals – Small – Random! xixi   Okay: >0.6 Okay: >>1 ● Statistical quality indicators: – R 2 correlation measure: – F-ratio (signal to noise):

WB1440 Engineering Optimization – Concepts and Applications Nonlinear LS ● Linear LS: intrinsically linear functions (linear in a i ): ● Nonlinear LS: more complicated functions of a i : ● More difficult to fit! (Nonlinear optimization problem) ● Matlab: lsqnonlin

WB1440 Engineering Optimization – Concepts and Applications LS pitfalls ● Scattered data: ● Wrong choice of basis functions: x f x f

WB1440 Engineering Optimization – Concepts and Applications Kriging ● Named after D.C. Krige, mining engineer, 1951 ● Statistical approach: correlation between neighbouring points – Interpolation by weighted sum: – Weights depend on distance – Certain spatial correlation function is assumed (usually Gaussian)

WB1440 Engineering Optimization – Concepts and Applications Kriging properties ● Kriging interpolation is “most likely” in some sense (based on assumptions of the method) ● Interpolation: no smoothing / filtering ● Many variations exist! ● Advantage: no need to assume form of interpolation function ● Fitting process more elaborate than LS procedure

WB1440 Engineering Optimization – Concepts and Applications Kriging example ● Results depend strongly on statistical assumptions and method used: Dataset z(x,y) Kriging interpolation

WB1440 Engineering Optimization – Concepts and Applications Reduced order model ● Idea: describing system in reduced basis: – Example: structural dynamics ● Select small number of “modes” to build basis – Example: eigenmodes

WB1440 Engineering Optimization – Concepts and Applications Reduced order model (2) ● Reduced basis: ● Reduced system equations: N1N1NkNkk1k1kNkNNkNkNNNNN1N1kNkN

WB1440 Engineering Optimization – Concepts and Applications Reduced order models ● Many approaches! – Selection of type and number of basis vectors – Dealing with nonlinearity / multiple disciplines ● Active research topic ● No interpolation / fitting, but approximate modeling

WB1440 Engineering Optimization – Concepts and Applications Aerodynamic model Example:Aircraft model Structural model Mass model

WB1440 Engineering Optimization – Concepts and Applications Neural nets

WB1440 Engineering Optimization – Concepts and Applications  (input) output Neural nets To determine internal neuron parameters, neural nets must be trained on data. x f(x)f(x)

WB1440 Engineering Optimization – Concepts and Applications Neural net features ● Versatile, can capture complex behavior ● Filtering, smoothing ● Many variations possible – Network – Number of neurons, layers – Transfer functions ● Many training steps might be required (nonlinear optimization) ● Matlab: see e.g. nndtoc

WB1440 Engineering Optimization – Concepts and Applications Radial basis functions ● Often used as transfer function in neural networks ● Can either be used for interpolation or smoothing ● Very versatile ● Check out

WB1440 Engineering Optimization – Concepts and Applications Genetic programming ● Building mathematical functions using evolution-like approach ● Approach good fit by crossover and mutation of expressions ^2 + / x2x2 x3x3 x1x1

WB1440 Engineering Optimization – Concepts and Applications Genetic programming ● LS fitting with population of analytic expressions ● Selection / evolution rules ● Features: – Can capture very complex behavior – Danger of artifacts / overfitting – Quite expensive procedure

WB1440 Engineering Optimization – Concepts and Applications Simplified physical models ● Goal: capture trends from underlying physics through simpler model: – Lumped / Analytic / Coarse ● Parameters fitted to “high-fidelity” data Simplified model Correction function x f(x)f(x) ● Refinement: correction function, parameter functions...

WB1440 Engineering Optimization – Concepts and Applications Model simplification summary Many different approaches: – Local: Taylor series (needs derivatives) – Interpolation (exact fit):  (Polynomial) fitting  Kriging – Fitting: LS – Approximate modeling: reduced order / simplified models – Other: genetic programming, neural nets, etc

WB1440 Engineering Optimization – Concepts and Applications Response surfaces in optimization ● Popular approach for computationally expen- sive problems: 1. DoE, generate samples (expensive) in part of domain 2. Build response surface (cheap) 3. Perform optimization on response surface (cheap) 4. Update domain of interest, and repeat Expensive model Optimizer Cheap model Optimizer ● Additional advantage: smoothens noisy responses ● Easy to combine with parallel computing

WB1440 Engineering Optimization – Concepts and Applications Example: Multi-point Approximation Method Trust region Design domain Response surface Sub-optimal point Optimum (Expensive) simulation