¾ÅÓÎÌåÓý

In the last (3rd) industrial revolution, ¡°virtual twins¡± (emulating a physical system by one, or more mathematical models to describe its complex behavior) were major protagonists. Nowadays, numerical simulation is present in most scientific fields and engineering domains, making accurate designs and virtual evaluation of systems responses possible, drastically cutting the number of experimental tests. Well-established physical-based models have been partially or totally replaced by these data-based models, mainly due to their computational complexity; this is especially true for applications requiring real-time feedback. Vast amounts of collected and carefully curated data have provided the key to interpretation and are thus able to advise users on imminent events.

This makes improved data-based predictive maintenance and efficient inspection & control possible by allowing for real-time decision making. However, arriving at an optimal learning stage takes considerable time and effort, just as the establishment of validated models took in the previous engineering revolution. The successes reported, and the many possibilities imagined, have led to an exponential increase in popularity of these ¡°digital twins¡±. There has been a rapid development of data-driven models that allow for the representation of a system, with all its richness, while ensuring real-time access to its governing model. However, replacing the rich history of engineering sciences, which proved their potential with spectacular success over the course of more than a century, led to feelings of bitterness and of a waste of acquired knowledge.

It was again at the end of the 20th century and the beginning of the 21st century that major scientific accomplishments in theoretical and applied mathematics, applied mechanics, and computer sciences contributed to new modeling and simulation procedures. Model Order Reduction (MOR) techniques were one of these major achievements. These techniques do not proceed by simplifying the model; models continue to be well-established and validated descriptions of the physics at hand. Instead, they rely on an adequate approximation of the solution that allows simplifying the solution procedure without any sacrifice of the model solution accuracy, in view of accommodating real-time constraints. In this way, the next generation of ¡®twins¡¯ was born. The ¡°Hybrid Twin¡± combines physics-based models within a MOR framework (to accommodate real-time feedback) and data science.

The three main criteria of the Hybrid Twin are:

• a simulation core able to solve complex mathematical problems representing physical models under real-time constraints
• advanced strategies able to proceed with data assimilation, data curation, and data-driven modeling
• a mechanism to adapt the model online to evolving environments (control). The Hybrid Twin embraces these three functionalities into a new paradigm within simulation-based engineering sciences (SBES)

You might be wondering if the Hybrid Twin can be useful to the automotive industry? The answer is, absolutely! Introducing this innovative application improves the reliability of Electric Vehicle (EV) safety and comfort. Once integrated into a connected EV via a smart human-machine interface, the solution could enable the driver to receive real-time alerts in the event that the car foresees an issue with the completion of the planned journey. The interface could then advise the driver on how to adjust the controls related to in-car comfort, for maximum driving range. Going further, the interface could advise an alternative route depending on traffic and weather forecasts, all the while ensuring the driver and his passengers are aware of degraded driving conditions that could impact their safety.

One crucial problem for all EVs is the management of energy to increase vehicle range. One of the main challenges is the passengers¡¯ consumption of thermal comfort in different weather conditions, including extreme cold or hot temperatures. Balancing battery capacity and comfort requirements are more becoming extremely demanding. Let's summarize the recent achievements made in this field and take a closer look at the results from the OPTEMUS project.

ESI had the pleasure of actively contributing to the OPTEMUS project by lending our expertise in material physics and combining established 1D and 3D simulation methodologies with our latest Hybrid Twin technologies. The results are tremendous: Together we achieved up to 40% range increase for electrical vehicles, while at the same time offering the highest cabin comfort and safety.

In a recent presentation at in September 2021, I shared when to deploy a Hybrid Twin for EVs and how to leverage optimal energy consumption and energy harvesting.

Ultimately, we focused on the virtual prototyping methodology where we develop a fully detailed virtual cabin for EVs. This 3D model consists of the 3D detailed models of the cabin, the smart seat, and the human models for thermal comfort assessment, with specific thermal comfort index, and the HVAC simplified models. It interacts with the full vehicle in a holistic vehicle-occupant-centered approach. New methods in design space exploration based on the Parametric Model Order Reduction enable the real time response of complex interactions interconnected requirements such as comfort, safety and range in a wide array of weather conditions.

Find out about ESI¡¯s Hybrid Twin solution for optimizing your electric vehicle cabin

Don¡¯t miss the next ESI Live! Register for our upcoming ESI Live 2022 where we will dive deeper into the expectations on the automotive industry related to digital transformation ¨C including the digital twin ¨C vision zero and sustainability.