For those who rely on simulation, AI can add speed to the power already on offer and, as a result, support a more iterative engineering approach that comes with fewer limitations, writes Mazen El Hout of Ansys
The advantages of AI-enhanced simulations are substantial, particularly when it comes to making rapid predictions that optimise the design processes. Training AI models accelerates this process, by enabling faster, datadriven insights that can be used to refine product development.
AI’s transformative potential is wellknown, especially for engineers, designers and other professionals who rely on simulation technologies to create innovative solutions. Integrating AI into simulation workflows can dramatically speed up design and optimisation, which is especially beneficial in industries where precision and efficiency are paramount, such as high tech, automotive and aerospace.
AI-enhanced simulations offer four key advantages. First, they bring speed: AI can analyse past simulations to quickly identify complex patterns while incorporating new data to uncover meaningful relationships. Second, AI can democratise simulation, making advanced tools accessible to non-technical users through easy-tonavigate, web-based platforms. Third, AI simulations also allow for the integration of multiple models to create comprehensive representations of complex systems, providing greater insight into product designs. Finally, AI-powered simulations enable a more iterative engineering approach, allowing designers to refine their work with fewer limitations.
Exploring AI
With these benefits in mind, let’s explore some real-world examples where AI and simulation are enhancing product design and driving innovation forward.
Designing optical systems that perform well under diverse environmental conditions – such as fog, haze, or varying weather – requires precise and efficient methodologies.
By integrating simulation datasets with AI-driven technologies, it becomes possible to predict illuminance fields with exceptional accuracy. For example, using just 10 optical simulation results as input, Ansys’ AI model demonstrated only a 2% error when calculating the maximum illuminance for unseen designs.
This approach drastically reduces the need for extensive simulation runs, empowering engineers to evaluate designs under numerous operational conditions quickly. By optimising optical system performance with AI, designers can accelerate development timelines while maintaining accuracy, ensuring products meet real-world demands efficiently.
Integrating AI into simulation goes beyond thermal analysis; in aerospace, structural analysis and design validation are essential to ensure an aircraft’s safety for take-off. This applies not only to large components, but also to smaller parts like jet engine brackets, which bear the engine’s weight and must be exceptionally strong.
As aircraft designs evolve, developing new jet engine brackets that meet both weight and structural strength requirements can be challenging. Engineers often start with prior design iterations that have been thoroughly simulated and validated. Building on this, they can incorporate an AI model trained on a diverse set of bracket designs. By retaining data from previous simulations, the model can quickly predict the behaviour of new shapes and can be updated with additional sample designs over time, allowing it to continually refine its predictions.
Using AI-driven simulation in structural analysis not only minimises material waste, but also turns failed designs into learning opportunities
Using AI-driven simulation in structural analysis not only minimises material waste, but also turns failed designs into learning opportunities. Each iteration provides valuable insights, helping both the model and engineers to identify best practices in jet engine bracket design. This ultimately enables them to develop an optimised, reliable product.
Back to earth
AI simulations are just as valuable for ground applications. In the automotive industry, environmental regulations on emissions – such as Euro 7 – are a top priority for manufacturers, requiring them to assess and demonstrate vehicles’ aerodynamic performance.
While engineers can rely on physical wind tunnel testing, this can be costly and time-consuming. Instead, they can turn to simulation for quick and costeffective assessments. AI enhances this by analysing variations in vehicle shape and topological changes, like rear mirrors, ski racks or spoilers, to predict aerodynamic performance and optimise designs.
Incorporating AI simulations early in the design process allows engineers to generate rapid, meaningful aerodynamic predictions at every stage and produce a better, compliant product in less time.
AI and simulation each offer significant benefits. Combining them to amplify their impact is a logical progression. This integrated approach enables engineers to tackle complex physics and engineering challenges, while organisations benefit from increased workforce efficiency, reduced costs and faster development timelines, ultimately bridging the gap between design and reality.
About the author:
Mazen El Hout, senior product marketing manager at Ansys, has a Master’s degree in Systems Engineering, a decade of experience in the software industry and extensive knowledge of AI, model-based systems engineering (MBSE) and embedded software and systems optimisation.
This article first appeared in DEVELOP3D Magazine
DEVELOP3D is a publication dedicated to product design + development, from concept to manufacture and the technologies behind it all.
To receive the physical publication or digital issue free, as well as exclusive news and offers, subscribe to DEVELOP3D Magazine here