Multiphysics addresses the reality that physical phenomena do not operate singly in nature. Simply put, design engineers are running more and more multiphysics simulations because they need to add reality to their increasingly complex models.
But being able to simulate the impact of different concurrent physical elements is only half the battle. The other half comes in providing an environment in which engineers from each discipline can collaborate on simulations and analyses with results able to be fed back into the design loop without manual intervention.
Engineers are increasingly looking to design highly complex systems while taking into consideration constraints across multiple physical domains. This multiphysics testing ranges across products and scenarios, from the testing of electronic devices for thermal and airflow issues during use, or off-road equipment for the impact of repeated mechanical motion on structural durability. Multiphysics testing also embraces those systems and components that operate in particularly harsh environments, such as ocean floor, extreme weather or earth orbit and beyond.
Multi-discipline integration
In the laboratory or other simulation environment, the impact of one physical element can be tested in isolation. Typically, the effects from one physics ‘domain’ impacts how a product behaves in another physics ‘domain’. For example, combined thermal-structural effects are crucial to jet engine engineering. Understanding multiphysics behaviour is therefore a major challenge to accurately predicting product performance in the real world.
In order to simulate real-world conditions, analysts must consider the impact of a number of different physics phenomena that occur concurrently, such as structural dynamics, heat transfer, fluid flow, nonlinear material behaviour and motion. This requires a combination of computational power and an integrated platform that can model and simulate interdependent effects.
Physical testing of multiphysics phenomena is very costly and time consuming. Traditional CAE methods are discipline-specific, making it difficult to perform multiphysics simulation. These challenges then result in late changes in the lifecycle which often result in a ripple-effect in the other areas, escalating costs and resulting in longer than anticipated product development schedules.
To meet these challenges, designers require an end-to-end product design platform. That way, the process of conducting multiphysics simulation can be streamlined, by delivering an integrated modelling environment that links coupled physics analyses together and eliminates the need for error-prone external data transfers.
Fully integrated multiphysics simulation workflows allow engineers to spend more time simulating the product under conditions that more accurately represent the real world. But the true value is that it makes it easier to feed back results into the design process. This allows for more product trade-off decisions as designers can evaluate different attributes based on the same design state – which is very difficult when building independent models to for each physics scenario.
The design feedback loop
Many of today's complex product designs run on a tight schedule, with challenging requirements and very little time to physically test design concepts before inclusion in the prototype. As such, a highly integrated analysis-driven design approach is needed to ensure a successful design and proper functionality.
A tightly integrated design platform allows for 3D design, finite element modelling, finite element analysis and data management to be carried out using common data models over multiple design possibilities. This greatly streamlines the design iteration process and quickly eliminates designs that would have had a low probability of success. The ability to easily map results and boundary conditions from one model to another simplifies the analysis and significantly reduces the time required to analyse each aspect of the system design.
When components and designs can be easily simulated within all of their parameters, the results can be more readily compared to operating requirements, fed back into the design process and used to refine the product as a whole.
When this is done within an automated and integrated platform, the development lifecycle involves less manual work and engenders more efficient upstream and downstream modelling and simulation interfacing. Furthermore, not having to re-enter data into multiple applications will avoid introducing errors.
It’s getting more complex
Multiphysics capabilities are evolving as more physics domains are integrated into the common software environment. And with more disciplines and new physics domains, simulation will become that much more granular.
While this improves the accuracy of modelling, it also adds greater levels of complexity and introduces more potential design options. Altogether, this will require an improvement in the algorithms used, as well as a scaling up of the required computational power.
Applications that deal with each element of the simulation will need to be interoperable at every level. More importantly, the results must be easily interpretable by different systems or elements of an integrated design platform.
In recent years we've seen products and designs become increasingly complex, thanks to the demand for more compactness and robustness in the end product. The modelling process now has the benefit of advanced design, testing and simulation tools, which take full advantage of the explosive growth in computational power over the past few decades.
It has also set the stage for addressing even more ambitious goals, many of which can't be achieved through hardware developments alone. Because conventional modelling strategies are typically based on single physical descriptions, they're simply unable to capture the relevant phenomena occurring over many space and time scales.
Use of a unified platform holds the key to ongoing success and development of multiphysics simulation. This allows for a deeper understanding of the impact of a variety of physical phenomena and enables faster and more intelligent design.
Jan Larsson is senior marketing director EMEA – Product Engineering Software, Siemens PLM Software