Enhancing the capability of digital twins using high fidelity reduced order models

Reduced order models (ROMs) are simplifications of high-fidelity, complex models. They capture the behavior of these source models so that engineers can quickly study a system’s dominant effects using minimal computational resources.

ROMs have become popular in the product development industry because engineers are facing market demands for shorter design cycles that produce higher-quality products. ROMs can be used to simplify various models from full 3D simulations, systems simulations, or embedded software. As a result, engineers can use them to:

Optimize product designs
Create larger systems simulations
Test control software
Create digital twins
Create Internet of Things (IoT) devices
Enable non-experts to use simulations

As a bonus, a ROM’s ability to simplify complex models means that they can often obfuscate proprietary information. Therefore, in many situations, organizations can share a ROM — in the form of a black box — with customers, external partners, or contractors without risking their intellectual property.

How to Create a Reduced Order Model Using Response Surface Models

ROMs can be used to model various static, transient, linear, and nonlinear systems. Specifically, response surface models (RSM) are a great tool to capture the behavior of static linear systems.

To create an RSM, engineers need to map the output parameters of a model based on the inputs to that model. To do this, engineers can create a design of experiments (DOE) using Ansys DesignXplorer. The DOE determines a series of design points that need to be tested in Ansys Fluent. Ansys Twin Builder can then be used to generate ROMs and develop system models to test the real-time behavior of equipment under operational conditions.

To create the RSM and ROM, the software uses advanced algorithms to predict the response between the simulation runs. The ROM can then be exported, in industry-standard formats, to other platforms.

To learn how to create ROMs in detail, watch this webinar:

How Reduced Order Models Improve Product Design

Engineers can use ROMs to reduce the time it takes to optimize and study a complex system. For instance, Fluent can use ROMs to explore design alternatives in the blink of an eye.

Ansys Fluent evaluates complex reduced order models (ROM) to explore design alternatives. For a cross-flow heat exchanger, the ROM capability provides a solution for each design point in as little as a second. A full simulation, however, could take over two hours. ROMs can be post-processed and evaluated in Ansys Fluent to gain insight beyond the numerical output previously available.

A full 3D computational fluid dynamic (CFD) study could take hours to simulate. However, by capturing the fidelity of the 3D model using a ROM, Fluent can assess a design in seconds. As a result, engineers can use ROMs to reduce development cycles, shrink research budgets, get to market faster, and limit computational resources.

Somenath Chatterjee, Ph.D.

Senior Application Engineer

Ozen Engineering, Inc. - Ansys Elite Channel Partner

America’s Channel Partner of the Year: 2015, 2018, 2021

Phone: 408-416-3240 Email: somenath.chatterjee@ozeninc.com

1210 E Arques Ave., #207, Sunnyvale, CA 94085

Tags:

Mechanical, Ansys, Fluent, Electronics, Reduced Order modeling (ROM), Electromagnetics, Workflow, Response Surface

Post by Somenath Chatterjee
December 6, 2023