./episode_167.sh
📚 The more I read, the more I acquire, the more confident I know nothing.
💥 𝗖𝗿𝗮𝘀𝗵 𝘁𝗲𝘀𝘁 𝗽𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝗼𝗻 𝗮𝘁 𝘁𝗵𝗲 𝘀𝗽𝗲𝗲𝗱 𝗼𝗳 𝗔𝗜
The need for a high number of design iterations comes up for performance-critical applications such as automotive crash testing, where physical tests are very expensive and time-consuming.
In this animation, we can see the SimAI prediction results of a crash impact on a bracket. Imagine a scenario where you’re using a virtual validation approach to assess how to get the best safety and durability performance from a bumper. Using SimAI you can generate an AI model trained on ~50 crash simulations from Ansys LS-DYNA. This would enable your design team to predict the optimal bumper configuration for a new concept in less than a minute, with an error of less than 0.5%.
Ultimately, it’s about enabling designers to use fast and meaningful crash predictions to accelerate a product’s time-to-market. Innovating faster towards a safer vehicle.
💡Introduction to OpenFOAM: Finite Volume Discretization in OpenFOAM
🏗️ Finite Element Analysis in the Browser using Google Collab
🎨 Scientific Visualization: Python + Matplotlib
This book is organized into 4 parts. The first part considers the fundamental principles of the Matplotlib library. This includes reviewing the different parts that constitute a figure, the different coordinate systems, the available scales and projections, and the authors also introduce a few concepts related to typography and colors.
The authors then explore the different types of plot available and see how a figure can be ornamented with different elements. The third part is dedicated to more advanced concepts, namely 3D figures, optimization, animation and toolkits.
📚 Can Physics-Informed Neural Networks beat the Finite Element Method?
TL;DR: “Considering the solution time and accuracy, PINNs are not able to beat the finite element method in our study.”
The recent success of deep neural networks at various approximation tasks has motivated their use in the numerical solution of PDEs. These so-called physics-informed neural networks and their variants have shown to be able to successfully approximate a large range of partial differential equations. So far, physics-informed neural networks and the finite element method have mainly been studied in isolation of each other.
💻 Engineering Tool of the Week - SPARSELIZARD
Sparselizard is a high-performance, multiphysics, hp-adaptive, open-source C++ finite element library running on Linux, Mac and Windows. A fast algorithm for mesh-to-mesh interpolation and a general implementation of the mortar finite element method allows us to easily work with non-matching meshes and provide general periodic conditions. FEM simulations can be weakly or strongly coupled to lumped electric circuits.
📚Book of the Week
“I do like CFD”
"I do like CFD, VOL.1" is a book on Computional Fluid Dynamics: the first one in a series to come. It is written by Dr. Katate Masatsuka, bringing those CFD know-how to the public.
🙃 Meme of the Week
Work With Us
❤️ 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