Metamodel-Based Multidisciplinary Design Optimization of Automotive Structures.

• Superior document: Linköping Studies in Science and Technology. Dissertations Series ; v.1870
• Author: Ryberg, Ann-Britt.
• Place / Publishing House: Linköping : : Linkopings Universitet,, 2017.; {copy}2017.
• Year of Publication: 2017
• Edition: 1st ed.
• Language: English
• Series: Linköping Studies in Science and Technology. Dissertations Series
• Physical Description: 1 online resource (61 pages)

Table of Contents

Intro -- Preface -- Abstract -- Populärvetenskaplig sammanfattning -- List of Papers -- Contents -- Part I -- 1. Introduction -- 1.1 Scope of Work -- 1.2 Outline -- 2. Structural Optimization -- 2.1 Multi-Objective Optimization -- 2.2 Optimization Under Uncertainties -- 2.3 Multidisciplinary Design Optimization -- 2.4 Metamodel-Based Design Optimization -- 2.5 Optimization Algorithms -- 2.6 Topology Optimization -- 3.1 Design of Experiments -- 3.1.1 Latin Hypercube Designs -- 3.1.2 Distance-Based Designs -- 3.1.3 Sequential Sampling -- 3.2 Metamodels -- 3.2.1 Feedforward Neural Networks -- 3.2.2 Radial Basis Function Neural Networks -- 3.3 Metamodel Validation -- 3.3.1 Error Measures -- 3.3.2 Cross Validation -- 3.3.3 Generalized Cross Validation -- 4. Multidisciplinary Design Optimization -- 4.1 Single-Level Methods -- 4.2 Multi-Level Methods -- 4.3 An MDO Process for Automotive Structures -- 5. Support Vector Machines -- 5.1 SVM Decision Boundary -- 5.2 SVMs to Improve Metamodel Accuracy -- 6. Summary of Appended Papers -- Paper I -- Paper II -- Paper III -- Paper IV -- Paper V -- 7. Conclusions and Outlook -- References.