Variable-Fidelity Surrogate

Experiment Design, Modeling, and Applications on Design Optimization

Jin Yi, Jun Zheng, Xinyu Li, Liang Gao (Autoren)

Buch | Hardcover

174 Seiten

2026
Springer Verlag, Singapore
978-981-95-5526-0 (ISBN)

Titel nicht im Sortiment

Artikel merken

This book delves deeply into the field of variable-fidelity surrogate modeling, examining its application in the optimization of complex multidisciplinary design optimization problems. The text presents a detailed exploration of surrogate modeling techniques, with a focus on variable-fidelity approaches that integrate models of varying accuracy to enhance the efficiency of optimization processes. Covering foundational concepts, the book progresses through diverse modeling strategies, including scaling function-based approaches, sequential techniques, physics-informed neural networks-based and deep transfer learning-based methods. It also addresses critical aspects such as the development of surrogate-assisted optimization algorithms.

By adopting a holistic perspective, this book emphasizes the importance of integrating surrogate models with optimization algorithms to tackle real-world multidisciplinary design challenges. The book is designed for graduate students, researchers, and engineers working in areas such as engineering design, optimization, and computational modeling. It is an essential resource for those interested in advancing the field of surrogate modeling and its applications to complex design optimization tasks, providing both theoretical insights and practical guidance.

Jin Yi received the B.S. and Ph.D. degrees in Industrial Engineering from Huazhong University of Science and Technology (HUST), Wuhan, China, in 2012 and 2017, respectively. He was a Research Fellow with the Department of Industrial Systems Engineering at the National University of Singapore from 2017 to 2020. Since 2020, he has been an Associate Professor in the Department of Mechanical Engineering at Chongqing University, Chongqing, China. His research interests include intelligent design optimization of advanced equipment, machine learning, and intelligent optimization. Dr. Jin has led multiple research projects, including a Youth Fund project from the National Natural Science Foundation of China, as well as independent research topics funded by the National Key Laboratory of Mechanical Transmission for High-End Equipment. Dr. Jin has published over 50 peer-reviewed papers in leading international journals such as IEEE Transactions on Industrial Informatics, Knowledge-Based Systems, Applied Soft Computing, and Structural and Multidisciplinary Optimization. His research achievements have earned him several prestigious awards, including the First Prize of Science and Technology Progress in Chongqing, the First Prize of Science and Technology Progress from the Chinese Association of Automation, and he holds 22 patents for invention applications/authorizations. Additionally, he serves as a Youth Editorial Board Member for the journal Complex Systems Modeling and Simulation (English edition).

Preface.- Chapter 1 Introduction.- Chapter 2 Key Technologies for Surrogate Modeling.- Chapter 3 Fast Nested Latin Hypercube Design via Translation Propagation.- Chapter 4 Nested Maximin Designs Based on Successive Local Enumeration and Discrete Optimization.- Chapter 5 Variable-Fidelity Surrogate Modeling via Scale Functions.- Chapter 6 Variable-Fidelity Physics-Informed Neural Networks.- Chapter 7 Multi-Fidelity Transfer Learning Model Based on Dynamic Task-Weighted Loss.- Chapter 8 Online Variable-Fidelity Surrogate-Assisted Harmony Search Algorithm with Multi-Level Screening Strategy.- Chapter 9 Expensive Design Optimization With Transfer-Learning Based Sequential Variable-Fidelity Surrogate.- Chapter 10 Conclusion Remarks.

Erscheint lt. Verlag	28.4.2026
Reihe/Serie	Engineering Applications of Computational Methods
Zusatzinfo	69 Illustrations, color; 7 Illustrations, black and white
Verlagsort	Singapore
Sprache	englisch
Maße	155 x 235 mm
Themenwelt	Mathematik / Informatik ► Mathematik ► Angewandte Mathematik
	Mathematik / Informatik ► Mathematik ► Finanz- / Wirtschaftsmathematik
	Technik ► Maschinenbau
Schlagworte	Computationally expensive optimization • Gaussian Process Model • machine learning • Multidisciplinary Design Optimization • Radial Basis Function • transfer learning • Variable-fidelity surrogate
ISBN-10	981-95-5526-4 / 9819555264
ISBN-13	978-981-95-5526-0 / 9789819555260
Zustand	Neuware