Meshless FEA, CFD + AI, Neural Operator - ML for SciComputing
🧠 The learned man knows that he is ignorant.
💻 Meshless FEA: Simplify, Simulate, Succeed! | Deep Dive
🌎 Neural Operator - Machine learning for Scientific Computing
Problems in science and engineering involve solving partial differential equations (PDE) systems. Sometimes, these PDEs are very hard. It could take traditional PDE solvers days and months to simulate some 3D fluid dynamics. Data-driven, learning-based methods promise to solve these problems faster and more accurately.
The classical development of neural networks has primarily focused on learning mappings between finite-dimensional Euclidean spaces or finite sets. To better approximate the solution operators raised in PDEs, we propose a generalization of neural networks to learn operators mapping between infinite-dimensional function spaces.
🚀 How optiSLang Leverages AI/ML
Finding an appropriate metamodel can take a lot of time and its number of parameters influences the data needed to train the AI/ML algorithm. To remedy this, Ansys offers the metamodel of optimal prognosis (MOP) approach. MOP is an automatic ML (AutoML) algorithm in optiSLang that finds the best metamodeling approach and prepares its settings. It also filters important parameters and works for 0D scalar-values. OptiSLang also includes the signal metamodel of optimal prognosis (signal MOP) algorithm — which is designed for curves — and the field metamodel of optimal prognosis (field MOP) algorithm — which is best for 2D or 3D models.
💦 CFD + AI - Real-Time Prediction
‘We are shortening development times by 25%,’ says Thomas Ulbrich, Member of the Board of Management of the Volkswagen brand in a recent interview. ‘In the future, vehicle projects will be created in 40 months instead of the previous 54 months.’
📊 Deep Learning Turns Professor’s Whiteboard Sketches into Working Code
The application is a deep learning system that takes a picture of the whiteboard and automatically recreates the FCA in MATLAB. This idea was possible within the MathWorks framework, where control tools like Simulink live in the same environment as discipline-based toolboxes like Deep Learning Toolbox™ and Computer Vision Toolbox™.
🎬 Video of the Week - Coding Adventure on (Meshless) Fluid Simulations
💻 Engineering Tool of the Week - FEM Operations Toolbox
The Finite Element Method (FEM) is the state-of-the-art method in the automotive and aerospace industries for structural design and lightweight construction. FEM results postprocessing analysis traditionally consists of the execution of fixed, predefined scripts for evaluation of standard results and the manual execution of in-depth analysis by specialized high-level experts.
The FEM Operations Toolbox enables the broad, systematic, and comprehensive application of methods from the fields of artificial intelligence, machine learning, data mining, and pattern recognition for a highly flexible, results-adaptive, and automated high-level analysis of FEM simulation results.
The proven mathematics of MATLAB® provides a trustworthy, reliable, and flexible foundation for this analysis.
📚Book of the Week
My First FEA Book Ever! A First Course in Finite Elements
The text material evolved from over 50 years of combined teaching experience it deals with a formulation and application of the finite element method. A meaningful course can be constructed from a subset of the chapters in this book for a quarter course; instructions for such use are given in the preface.
The course material is organized in three chronological units of one month each: 1) the finite element formulation for one-dimensional problems, 2) the finite element formulation for scalar field problems in two dimensions and 3) finite element programming and application to scalar field problems; and finite element formulation for vector field problems in two dimensions and beams.
The access and use of ABAQUS software and MATLAB exercises will be in conjunction with the book.
❤️ Support the Blog & Newsletter
Let’s connect on Twitter, Instagram or LinkedIn!
For any business-related issues or collaborations, email me at support@jousefmurad.com!
Keep engineering your mind! 🧠
Jousef