Scientific Machine Learning with Engineering Applications

This book equips readers with a rigorous and practical framework for solving complex engineering problems directly from governing equations using modern machine learning techniques. It bridges established principles from mechanics, numerical analysis, and scientific computing with emerging physics-based learning approaches, enabling reliable modeling, simulation, optimization, and inverse analysis beyond purely data-driven methods. A distinctive feature is its critical comparison of machine learning-based solvers with classical techniques such as the finite element method, isogeometric analysis, and meshfree methods, highlighting strengths, limitations, and domains of applicability. The scope ranges from foundational concepts to advanced engineering applications, supported by worked examples, reproducible code, and extensive references. The book is intended for graduate students, researchers, and practitioners in engineering, applied mathematics, and computational sciences who seek a principled entry point and a state-of-the-art reference for physics-based machine learning in modeling and simulation.

This book equips readers with a rigorous and practical framework for solving complex engineering problems directly from governing equations using modern machine learning techniques. It bridges established principles from mechanics, numerical analysis, and scientific computing with emerging physics-based learning approaches, enabling reliable modeling, simulation, optimization, and inverse analysis beyond purely data-driven methods. A distinctive feature is its critical comparison of machine learning-based solvers with classical techniques such as the finite element method, isogeometric analysis, and meshfree methods, highlighting strengths, limitations, and domains of applicability. The scope ranges from foundational concepts to advanced engineering applications, supported by worked examples, reproducible code, and extensive references. The book is intended for graduate students, researchers, and practitioners in engineering, applied mathematics, and computational sciences who seek a principled entry point and a state-of-the-art reference for physics-based machine learning in modeling and simulation.