Multidisciplinary Optimization: Enhancing Efficiency and Performance with Artificial Intelligence (AI)
Altair EmployeeOrganizations commonly incorporate virtual modeling into product development cycles to reduce costly reliance on physical prototyping and expedite time-to-market. With mainstream accessibility to simulation technology, virtual prototyping has become standard practice along with high-performance computing (HPC). Broad accessibility to HPC, which enables faster simulation of computationally demanding models, has increased demand for multidisciplinary simulation.
Engineers are particular interested in adopting simulation practices because of the unparalleled gains simulation achieves for optimizing modeled systems. Consequently, there is a growing necessity for multidisciplinary optimizations (MDO). Throughout the optimization process, numerous simulations are essential for models to reach the best technical and performance outcomes. Yet, with these simulations, virtual models are taxed on what they can deliver. To address this, a common standard is emerging – the use of reduced order models (ROM).
ROMs are derived from high-fidelity models but operate at significantly faster speeds while maintaining a commendable level of accuracy. The efficiency of these models lies in their focus on calculating specified critical information, in contrast to the comprehensive calculations performed by high-fidelity models. For example, ROMs might calculate the temperature at specific points within a vehicle cabin, as opposed to estimating the entire temperature distribution in the cabin volume.
Because of the need to identify relationships among selected variables, adopting equation-based approaches for model-order reduction is challenging. This is where AI plays a crucial role to simplify the identification task. MDO can be executed on hybrid models that are comprised from both high-fidelity and reduced-order models that are generated using different tools. This emphasizes the need for open solutions like Altair® romAI™ that can integrate seamlessly into workflows without adding complexity.
Altair’s groundbreaking romAI technology generates accurate and efficient ROMs, both static and dynamic, and it easily integrates into any software, hardware, or cloud-based platform. Federico D’Amico, digital engineer specialist, recorded a recent testimonial describing the impact of romAI on design challenges at Leonardo:
“The romAI technology seamlessly integrated into our current workflow and gave us highly accurate results while reducing the HPC time needed to perform our simulations.”
To learn more about Leonardo’s romAI application, visit https://altair.com/romai-applications.