CAE Tools Used in Automotive Industry

What is Computer Aided Engineering, and how can it help the automotive design industry?

This article demystifies CAE (Computer-Aided Engineering) and explains how its impact as a design tool in the automotive industries, ranging from Tier 1 suppliers to OEMs, is ever-increasing.

Also, how is Artificial Intelligence creating a beautiful revolution? What is the role of innovative companies like Neural Concept in making CAE tools go beyond their fame as research application, making it available to everyone at all product concept and development stages?

Computer-Aided Engineering

CAE tools are traditionally associated with advanced Research & Development services related to a specialist skill in physical modelling.

Computer-Aided Engineering gives analytical predictions to support automotive product designs, showing which method achieves the best performance. Numbers or images produced by CAE can replace laboratory tests on the product.

Imagine, for example, a crash test on a vehicle performed on the computer before the physical prototypes of the vehicles are built. Computer-Aided Engineering is sometimes referred to as an analysis tool. This means it can analyze and enhance the CAD design of components or a whole vehicle by processing its geometry and, with assumptions on the type of problem, can predict results.

However, there are several approaches. How to identify the best ones? And what is the role of AI (artificial intelligence) in the future product development process?

CAE simulation deals with 3D objects, but its function is much more than graphical simulation. When processing a simulation, it is not enough to say that "it looks realistic": CAE should be reliable in predicting engineering variables. Therefore, it is much more than visualization; it is an engineering analysis.

Engineers can develop the vehicle components on CAD before building any prototype. Thus, automotive companies can save the budget previously dedicated to building prototypes and carrying out laboratory work. In this virtualization process, security and safety are enhanced as well as the repeatability and robustness of virtual experiments.

The Role of Computer-Aided Engineering in Concept Design

The decision-making process related to product releases has been dramatically accelerated thanks to engineering simulation. Indeed, it is possible now to envisage product behaviour even upstream of production - i.e. during the conceptual design phase. The concept design can have a computer representation with CAD, even without a physical twin. Hence, automotive engineers can assess their products during the early design stages for aspects regarding external aerodynamics or thermal and mechanical fatigue and driver security or comfort.

The Role of CAE in Product Design

Computational fluid dynamics (CFD) and finite element analysis (FEA) can also contribute, during later stages of automotive product design, to the fine-tuned engineering optimization of products and reduce manufacturing costs.

What are the different main branches of Computer-Aided Engineering analysis? Applications are related to solid mechanical objects (FEA) or include the effects of fluids and thermal exchange (CFD).

The Role of FEA Software

Let's start with FEA Sofware, an acronym meaning Computer-Aided Engineering simulation of solid bodies (Finite Element Analysis is the mathematical approach). Its wide range of engineering use cases makes it the most known of CAE tools.

With FEA simulation, automotive corporations can assess the impact of different materials, their thickness and shape on various key performance indicators such as crashworthiness or durability.

Examples of FEA Automotive applications

The mechanical applications of FEA are numerous. Companies can avoid risks by predicting vibrations in the vehicle and its components, the durability of its parts subject to fatigue, and the safety of passengers during a crash. NVH (i.e., Noise, Vibration and Harshness) and Crashworthiness simulations are now part of the automotive design process of validation and optimization of product performance, with static and dynamic analyses.

The impact of mechanical simulation on automotive industries is evident:

  • Will a vehicle with different materials and components be lighter and guarantee equal or superior safety?
  • Will a new type of engine generate a noise impacting the driver and passenger?
  • Will components crack, subject to thermal fatigue?

The Role of CFD in Fluid Flow Analysis

CFD (Computational Fluid Dynamics) means simulating in 3D and with engineering precision everything related to fluids: air, gas, liquid and combination.

We can broadly divide CFD engineering applications into external and internal flows.

Examples of CFD Automotive applications

As an engineering application of external flows, imagine the streamlines around and behind a Formula 1 car. Computer-Aided Engineering is crucial for automotive aerodynamics teams to assess the aerodynamic forces pushing the car to the race floor, assess the vehicle's drivability during the race and develop new designs or strategies.

Last but not least, simulations can predict the wake behind the cars that can be exploited by the following vehicle to overcome air resistance, leading to the optimization of the vehicles' drivability. We will show how AI will revolutionize this process.

External car aerodynamics prediction with CAE

The CAE Tools in the Automotive Industry

Before introducing the AI revolution, we will summarize the traditional state-of-the-art approaches in simulation services and software for the industry.

CFD

The popularity of CFD dramatically increased when it became a compact, streamlined and highly automated professional software.

To re-cap the story of fluid flow analysis, it was mainly developed in the 80s and 90s by academic startups. Those startups could flourish commercially because they were able to better connect their core physical models (solvers) to the 3D geometry representations (CAD import and meshing) and facilitate the production of 3D maps or plots (postprocessing).

Thus, the most CFD successful tools combine three operations: CAD import & meshing, solver and postprocessing. As we will see later, AI can provide a surrogate of those operations, working in a reduced time frame: the reduction is from hours or days to fractions of a second.

Finite Element Analysis FEA Software

The original FEA applications were generated by pioneering research applications focused on analyzing mechanical structures to predict their stress and deformations. One of the first sponsors was NASA for NASTRAN, as of today, a widely used analysis tool in the general field of stress analysis.

On top of static stress analysis, dynamic analyses were also introduced in crashworthiness analysis.

As we will see later, even apparently "impossible" applications for AI, such as crashworthiness, can be successfully reproduced by AI. But AI goes beyond reproduction - we will show how AI will revolutionize the design process.

What Is the Best Way to Adopt CAE Tools?

How to adopt CAE tools in the automotive industry, especially to the needs of CAD users? This is the big question, and the traditional answers before AI were the following:

  1. Go for external services, i.e. rely on external help from automotive engineering teams, often located offshore. The offshore services approach bypasses the decision on internal software investments. However, it relocates a key capability out of the company. Thus, Engineering Services can be a solution to kick off CAE software usage with turnkey solutions and acquire simulation competencies later to increase data security.
  2. Invest in High-End CAE. This choice relies on hired skilled specialists and big hardware such as purchased compute clusters or rented cloud computing. While yielding quality results, this approach calls for significant investments and dedicated support services by software vendors.
  3. Go for "CAE+CAD", i.e. integrate CAE software into CAD tools. Simulation and CAD tools are installed on the same platform, enhancing the ease-of-use side of CAE and speeding up the initial adoption phase. The associated cost is less precise simulations. This can be an issue when accurate, mission-critical engineering answers are needed.

CAD and CAE Tools in the Automotive Industries

CAE tools can read the products (their shape, their materials etc.) as data stored in CAD. A critical link between CAD and CAE is the response time between a CAD shape modification and the CAE software feedback on the change. The shorter the response time, the larger the number of engineering modifications in a given time and the impact on product innovation and competitiveness.

As seen, Computed Aided Engineering is a type of analysis software yielding predictions on critical functional requirements of products, like durability and energy efficiency, giving feedback on needed shape modifications.

Given the critical feedback of simulation tools on products, Computer Aided Engineering software should be deployed in the earliest possible stages of the vehicle's development. A bottleneck needs to be removed: product engineers with CAD and product skills are not also specialists in engineering analysis.

A solution devised in the last few years has been to embed Computer Aided Engineering inside CAD platforms.

Advantages and Drawbacks of Using CAE and CAD Tools

What are the drawbacks of CAD+CAE tools, and what could be a different way to empower automotive CAD users with predictive tools? The underlying idea of CAD+CAE tools is to empower CAD operators with simulation software embedded into their favourite CAD platform.

An evident advantage of embedding CAE into CAD is the immediate deployment of such a solution for any operator without specific CAE skills.

However, when making a decision based on simulation, it is not always possible to be sure that the CAD+CAE simulation is 100% reliable - further analyses with high-end solutions may be required.

Fortunately, there is a way to empower automotive CAD users with reliable simulations. Engineering Intelligence with Neural Concept Shape is an approach based on Deep Learning, a subfield of Artificial intelligence (AI). AI processes the CAE data produced by Research & Development specialists. Those data are coupled to CAD and elaborated into AI predictors; i.e. bespoke software services such as accessible Apps for CAD users.

We will now see how the revolutionary Engineering Intelligence approach is taking place.

How Can AI Support the Automotive Design Process?

Artificial Intelligence (AI) will profoundly impact the automotive industry in the future. Automotive management expects real-time tools to help them make quick decisions. AI can revolutionize CAE simulation, making it accessible to everyone in Engineering departments.

How Can AI Revolutionize CAE?

AI can take engineering to the next level of real-time computer-aided engineering coupled with CAD.

With traditional CAE simulation tools, there is always some time gap, or delay, between the time of shape creation (CAD) and the time when CAE answers. Can the delay be reduced to a few seconds instead of hours? Can an F1 car be simulated in fractions of seconds instead of hours? This White Paper on F1 applications shows it's already possible.

The same concepts extend to the real-time aerodynamic simulation of passenger vehicles.

How Can CAE Become Real-Time and Be Used by Anyone in the Industry?

The two essential functions of AI for CAE are:

  1. Make it available to anyone involved in the engineering process in a company and
  2. Make it possible to process a simulation in real-time (both for static and dynamic analyses).

AI is part of a new set of tools. But, differently from previous tools, process simulation is designed to minimize learning effort and hardware investment budget.

As automotive use cases show, AI can efficiently recycle automotive data and produce data-driven tools that mimic resource-intensive computer-aided engineering tools.

Even real-time virtual crash tests of vehicles have an AI counterpart.

The AI processes will dramatically improve job quality for product design staff, who will be delivered intuitive predictive tools usable on their desks.

An example of a series of bespoke AI tools for design teams is Neural Concept Shape Production.

Providing AI with data is not an issue for automotive companies that already store thousands of files in their PLM systems, just waiting to be recycled with AI methods such as Deep Learning.

CAE or data science experts collect and process CAD and CAE with a platform that gives total control to experts and helps them to produce bespoke AI apps for designers.

The Impact of AI on Automotive Design in the Future

Vehicles in the future will be designed with a more collaborative and streamlined process. Management and staff will sit and review design solutions and product performance, with a reduction of iterations in meetings. Design modifications will be discussed interactively. AI will shorten the timing of external services by Computer Aided Engineering partners and make services more nested in the loop of the product development process.

Interaction between Tier 1 suppliers and OEMs is accelerating.

With Engineering Intelligence, Automotive Tier 1 suppliers will be able to check in advance all designs of their components by processing them during the RFQ phase. Thus, competitive prices will not compromise component quality, robustness or later issues in the manufacturing stage of vehicles.

Based on proprietary CAD and CAE data, AI can support better knowledge security for automotive industries and preserve their research investments even when dated back years or decades!

Conclusion: The AI Revolution in Automotive Engineering - from Conceptual Design to Manufacturing

In conclusion, AI tools support companies in a revolutionary transition in simulation. CAD and CAE will evolve from open-loop analysis to closed-loop product performance optimization with continuous feedback between the shape and functionality of products.

Simulations will become increasingly more coupled to Engineering and Manufacturing, empowering any automotive engineer with decision-making tools across the product development process, from concept design to final product release.

About the author
Anthony Massobrio
Anthony is a CFD expert since 1990, working initially as a senior researcher, then moved to Engineering, acting also as technical director in a challenging Automotive Tier 1 supplier environment. Since 2001, Anthony has worked in Software & Engineering Consultancy as a Sales Engineer and manager. In 2020, Anthony fell in love with AI and has worked since then in the field of “AI for CAE” at Neural Concept and as an independent contributor.
About the author
Anthony Massobrio
Anthony is a CFD expert since 1990, working initially as a senior researcher, then moved to Engineering, acting also as technical director in a challenging Automotive Tier 1 supplier environment. Since 2001, Anthony has worked in Software & Engineering Consultancy as a Sales Engineer and manager. In 2020, Anthony fell in love with AI and has worked since then in the field of “AI for CAE” at Neural Concept and as an independent contributor.
About the author
Anthony Massobrio
Anthony is a CFD expert since 1990, working initially as a senior researcher, then moved to Engineering, acting also as technical director in a challenging Automotive Tier 1 supplier environment. Since 2001, Anthony has worked in Software & Engineering Consultancy as a Sales Engineer and manager. In 2020, Anthony fell in love with AI and has worked since then in the field of “AI for CAE” at Neural Concept and as an independent contributor.
About the author
Anthony Massobrio
Anthony is a CFD expert since 1990, working initially as a senior researcher, then moved to Engineering, acting also as technical director in a challenging Automotive Tier 1 supplier environment. Since 2001, Anthony has worked in Software & Engineering Consultancy as a Sales Engineer and manager. In 2020, Anthony fell in love with AI and has worked since then in the field of “AI for CAE” at Neural Concept and as an independent contributor.
About the author
Anthony Massobrio
Anthony is a CFD expert since 1990, working initially as a senior researcher, then moved to Engineering, acting also as technical director in a challenging Automotive Tier 1 supplier environment. Since 2001, Anthony has worked in Software & Engineering Consultancy as a Sales Engineer and manager. In 2020, Anthony fell in love with AI and has worked since then in the field of “AI for CAE” at Neural Concept and as an independent contributor.
About the author
Anthony Massobrio
Anthony is a CFD expert since 1990, working initially as a senior researcher, then moved to Engineering, acting also as technical director in a challenging Automotive Tier 1 supplier environment. Since 2001, Anthony has worked in Software & Engineering Consultancy as a Sales Engineer and manager. In 2020, Anthony fell in love with AI and has worked since then in the field of “AI for CAE” at Neural Concept and as an independent contributor.
About the author
Anthony Massobrio
Anthony is a CFD expert since 1990, working initially as a senior researcher, then moved to Engineering, acting also as technical director in a challenging Automotive Tier 1 supplier environment. Since 2001, Anthony has worked in Software & Engineering Consultancy as a Sales Engineer and manager. In 2020, Anthony fell in love with AI and has worked since then in the field of “AI for CAE” at Neural Concept and as an independent contributor.
About the author
Luca Zampieri
Luca joined the Application team in 2018, aiming to build the next generation Deep-Learning tool dedicated to CAD and CAE.
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
About the author
Thomas von Tschammer
Thomas joined the team in 2018 as Director of Operations, aiming to empower engineers with next a next generation Deep-Learning tool dedicated to CAD and CAE.
About the author
About the author
About the author
About the author
About the author
About the author
About the author