To Excel at Engineering Design, Generative AI Must Innovate!
Spend less time running expensive, repetitive tests and more time learning from your engineering data to predict the exact tests to run.

To Excel at Engineering Design, Generative AI Must Innovate!

In this newsletter, I share the learnings from Monolith from over 300+ AI projects with engineering leaders and provide:

  • Helpful?insights?from our learnings working at the forefront of product development
  • Expected benefits from adopting a more data-driven workflow
  • Guidelines?to evaluate a suitable use case for AI

Follow me for more insights on AI in Engineering Product Development.


Prediction of Lithium-Ion Battery Thermal Runaway via Multiphysics-Informed Neural Network (#MPINN) ??

Multiphysics-informed neural network proposed for thermal runaway prediction in LIBs.

In this study, a multiphysics-informed neural network (MPINN) is proposed for the estimation and prediction of thermal runaway (TR) in lithium-ion batteries (LIBs). MPINNs are encoded with the governing laws of physics, including the energy balance equation and Arrhenius law, ensuring accurate estimation of time and space-dependent temperature and dimensionless concentration in comparison to a purely data-driven approach.

Read the full paper here??


?? AI Models that prioritize similarity falter when asked to design something completely new.

AI models that prioritize similarity falter when asked to design something completely new.

Deep generative models, like ChatGPT, are highly skilled at mimicking existing content, but their application in engineering tasks faces limitations, as they prioritise statistical similarity over innovation.

A study from MIT highlights these challenges and suggests that AI models need to be reoriented to focus on design requirements rather than pure mimicry. In a bicycle frame design case study, models trained with engineering-focused objectives produced more innovative and high-performing designs compared to similarity-focused models.

By adapting AI models to prioritise factors like performance, design constraints, and novelty, engineers can leverage their potential to create better, more innovative products in various fields beyond multimedia.

Read the full article from MIT News here: https://news.mit.edu/2023/generative-ai-must-innovate-engineering-design-1019


4 Applications for AI in Validation Test ??

Based on hundreds of AI projects in engineering product development, we’ve identified the four use cases in validation test specifically where AI delivers the most significant impact.

Moreover, we’ll show how you can reduce your validation testing by up to 73% based on battery test research done by teams at Stanford, MIT, and Toyota Research Institute.

By embracing AI and machine learning principles, engineering teams can navigate the intricate challenges of understanding - and validating - the intractable physics of complex products more efficiently, leading to streamlined development, optimised designs, and faster time to market.


?? Find a Test Plan You Can Trust | Battery Testing with Monolith

The complexity of EV batteries, which involves electrical, chemical, and thermal mechanisms, leads to a costly and time-consuming process for their testing and validation.

In the first part of the EV webinar series, we reviewed the latest research on using AI models to significantly reduce the testing needed for EV batteries. In this follow-up webinar, we’ll show how to implement these concepts using Monolith software.

Monolith Lead Principal Engineer Jo?l Henry will demonstrate how to train a model using the latest active learning techniques to characterise battery performance in much fewer test steps than traditional approaches.


Reducing EV Battery Testing by 70% ?? | Trailer

Real-world data from a study conducted by Toyota Motor Corporation, Massachusetts Institute of Technology, and Stanford Universitydemonstrates that AI can reduce testing efforts by a remarkable 98%.

?? In our Monolith webinar, my colleague Richard explains how you can achieve these same results: https://www.monolithai.com/webinars/ev-battery-testing

Additionally, we will cover:

?? Early Stopping of Tests: Predicting battery failure after a certain number of cycles for efficient testing.

?? Lifetime Prediction of Batteries: Ensuring optimal performance and reliability by accurately estimating battery lifetimes.

?? AI-recommended Test Planning: AI-based models act as a "next test recommender" to identify the tests that provide maximum information


Feel free to contact the experts at Monolith to discuss your AI Use Case and run a feasibility analysis.


If you’d like to see more tips on AI in Engineering and Machine Learning for Battery Design, Lifetime and Charging Strategies, consider subscribing to this newsletter!

Test Less. Learn More.

Richard

要查看或添加评论,请登录

Richard Ahlfeld, Ph.D.的更多文章