Altair romAI Contest

About Altair romAI:

• Altair romAI is a game-changing, AI-powered technology within Altair Twin Activate, designed to create Reduced-Order Models (ROMs) from high-fidelity simulation or experimental data.
• romAI enables accelerated system evaluations by leveraging ROMs, reducing the reliance on time-consuming simulations.
• This empowers engineers to explore more design alternatives, optimize system performance efficiently, and streamline development cycles.
• romAI supports a wide range of applications, including rapid performance prediction (e.g., pump curves, motor behavior), virtual sensing, simulation cost reduction, and dynamic system modeling for control strategy development.
• By combining speed, flexibility, and seamless integration, romAI helps reduce time to market, cut development costs, and advance sustainability goals, making it a vital enabler of intelligent, responsive Digital Twins.

Challenge:

• Create Reduced-Order Models (ROMs) from high-fidelity simulation or experimental data using Altair romAI.

Model File:

• It’s an open problem statement contest. Participants can use simulation-generated data, publicly available datasets (e.g., from Kaggle), or any custom/novel dataset relevant to their engineering stream.

Objective:

• Build a reduced-order model using Altair romAI and deploy or optionally integrate it into a 1D system model using Altair Activate to simulate system behavior or predict key performance metrics efficiently.

Themes/Example Use Cases:

• Electrical – Power Grid Stability and Fault Prediction: Leverage electrical grid data to model and predict faults, enhancing grid reliability, operational efficiency, and stability.
• Civil – Structural Response to Seismic Loads: Analyze how buildings and infrastructure respond to seismic events using simulation and real-time data to design earthquake-resilient structures.
• Electronics – Battery State of Charge (SoC) and Health Estimation: Develop predictive models to estimate battery SoC and health, improving performance, safety, and lifecycle management in energy storage systems.
• ECE – RF Circuit Performance Modeling: Utilize data-driven approaches to simulate and optimize RF circuit behavior, supporting faster development cycles and improved signal integrity in communication systems.
• Mechanical – Rapid Performance Prediction of Mechanical Components: Apply machine learning or reduced-order modeling techniques to predict the performance of pumps, motors, or engines, thermal systems enabling faster design cycles and real-time system monitoring.

Evaluation Criteria:

• Build and Deploy a romAI Model in Activate Using Open-Source Datasets:
  Leverage publicly available data to train a romAI model and successfully deploy it within the Altair® Twin Activate® environment. – 30 points
• Use Internal Engineering Data or Self-Constructed Dataset to Build romAI Model:
  Use novel data which is not available in any online platforms or Altair websites to build a romAI model and deploy them in Activate platform. (Meaning do not use tutorial problems) – 20 points
• Create a 1D System Simulation Model and Integrate romAI as a Subset of the System for Output Generation:
  Develop a 1D system simulation and establish a functional connection with a romAI model to evaluate system performance or behavior. – 30 points
• Prepare a Well-Structured Presentation and a Detailed Technical Report:
  Outline the end-to-end process, including problem statement, novelty of the work, data collection, preprocessing, model development, validation, key engineering challenges addressed, Digital Twin workflow and potential applications. The presentation should be visually engaging, and the report should be clear, comprehensive, and structured. – 20 points

Note: Submissions featuring unique or novel problem statements that address complex engineering challenges will be given higher preference during evaluation.

Real-World Applications of romAI:

• Rolling Out a Better Process - Digital Twin Slashes Waste 15% and Runtime Hours to Seconds: Link
• romAI: How to Speed-Up DEM Simulations Leveraging Artificial Intelligence: Link
• Combining AI and Simulation for Efficient Bulk Solids Handling Optimization: Link
• An AI-Augmented Multiphysics Approach to Speed Up the Analysis of a Solenoid Valve: Link
• How to Speed Up Truck Handling Simulation Using Artificial Intelligence: Link
• romAI: Generation of a Dynamic Reduced-Order Model (ROM) Using AI for Linear Actuators: Link
• Altair Community: Link

Resources Provided:

• Free Academic License: Link
• Free Personal Edition: Link
• Free Technical Support Forum: Link
• Altair eLearning: Link
• Help Documents: Welcome to Altair Twin Activate
• Learn from YouTube Videos: Link

What to Submit:

• In this contest, the use of Altair® romAI™ is mandatory, and hence submit related files:
  • Dataset
  • ROM AI model folder
  • Activate model (.scm)
• Presentation and Report
• YouTube link of the video created explaining your work
• Submission Instructions:
  • Zip all contents → Rename as "ManavRachna_YourName.zip" and submit at the given link: Submission Link