In recent years, scientific machine learning (SciML) has emerged as a powerful tool to solve problems in engineering scince . The goal of this special issue is to bring together such research and document the cutting edge of SML in mechanics..The promise of SciML lies in the idea that it is possible and indeed advantageous to learn the behavior of systems, materials and the very nature of solutions to problems in mechanics.
While the idea of machine learning itself is not new, the increasing availability of computing power has made it practicable to use SML in the context of mechanics. As is clear from the number of articles in recent mechanics literature, several researchers are pursuing SML approaches in a wide range of disciplines like solid,structural, thermal and fluid mechanics, to name just a few. We welcome contributions in all areas of mechanics and across the spectrum of machine learning approaches.
This includes (but is not limited to)
- Applications of physics-informed neural networks and physics-informed generative adversarial networks
- SciML for reduced order modeling
- Learning in the presence of uncertainty
- SciML for multiscale and multiphysics problems
- Offline solution of problems and generative design
