by Nam-Ho Kim
Optimization using surrogate models has been popular in many engineering applications due to its versatility and efficiency. Surrogate models approximate complex relationship between quantity of interest and input variables in the form of simple analytical/statistical functions using a set of samples. The goal of surrogate modeling is to obtain acceptable prediction capability at unsampled locations. However, in reality, it requires significant technical expertise to use surrogate models properly. This is partly because building surrogate models is one thing and understanding them is another. An important goal of this book is to explain the meaning of built surrogates and to provide in-depth understandings of such surrogates. This book emphasizes when and how a surrogate model can fail in approximating complex functions. One of the best ways to understand the theory of surrogate models is to implement it using a computer program. Throughout this book, many Matlab codes are provided as practice tools.
Book in amazon.com or Barns and Noble
Contents: | |
| Preface | xiii |
| PART I: Basics of Surrogate Modelings | 1 |
| 1. Introduction to Surrogate Models | 3 |
| 2. Polynomial Response Surfaces | 17 |
| 3. Design of Experiments | 71 |
| PART II: Design Optimization | 117 |
| 4. Optimization Definition and Formulation | 119 |
| 5. Numerical Optimization Algorithms | 161 |
| 6. Global Search Optimization Algorithms | 197 |
| PART III: Advanced Topics in Surrogate Modeling | 227 |
| 7. Kriging Surrogate-Gaussian Process Model | 229 |
| 8. Neural Network Model | 295 |
| 9. Multi-Fidelity Surrogate Models | 353 |
| 10. Efficient Global Optimization | 399 |
| References | 433 |
| Index | 439 |