FEATURE

Racing ahead

Gemma Church finds out how electric vehicles can make performance and user experience improvements

As the internal combustion engine raced into the mainstream some 200 years ago, electric vehicles were left at the starting line due to the complexity of the design process.

The storage of electrical energy and its conversion to mechanical energy is an overriding challenge when designing such vehicles. You must also take into consideration the multiphysics nature of simulating an electric vehicle. 

Mark Walker, application engineering manager at MathWorks, said: ‘Electric vehicle simulations depend on the interaction between multiple domains much more than internal combustion powertrain development. For example, effective battery management depends on a well-validated understanding of the electrical and thermal domains.’ 

Walker added: ‘Models are especially important because one of the most important variables in controller design – battery state of charge – cannot be directly measured. By contrast, most variables in traditional powertrain design are observable.’

Quality and safety standards add another layer of complexity as Frédéric Merceron, transportation and mobility solution experience director at Dassault Systèmes, explained: ‘Software is of very high importance [in any vehicle design process, but] even more so with electric vehicles. This is why the embedded systems now have a quality and safety standard (ISO26262) on top of all the other standards that car manufacturers and suppliers have to conform to.’

The electric motor has come of age 200 years

But the electric motor is gaining ground on its conventional combustion engine counterpart. Simulation and modelling are helping to unlock the complexities of electric vehicles and demonstrate the benefits to consumers and the wider transport industry.

David Moseley, manager and CAE analyst at electric car company Lucid Motors, said: ‘If we tried to design a vehicle without the aid of integrating simulation and modelling, then we would have a poor vehicle in terms of the design and acceptance by the public. We need to use sophisticated design space exploration to be competitive.

‘Simulation and modelling provide such exploration intelligence. We can get a solution that is better in performance than a human could achieve with a purely holistic approach [but] this requires high investment in the right engineers and the right analysis capability,’ Moseley added.

Lucid Motors is developing a new class of premium electric vehicle, and its first Lucid Air all-electric sedan is due to be launched in 2019. In preparation for production, Lucid Air prototypes are currently undergoing a rigorous development programme, which incorporates Esteco’s modeFrontier software. 

Such simulation, modelling and optimisation tools are, generally, used for two main design areas for electric vehicles: the system design and the simulation of the vehicle at a range of fidelities.

For the system design process, trade-offs are necessary to meet the range of requirements demanded to produce a viable, and competitive, electric vehicle. To address this issue, Altair has developed a three stage concept development process, called C123, where optimisation technology is deployed at each stage of the design process. 

C123 provides the designer with structural data relating to the mass and attribute performance consequences of various design decisions and helps them to manage target trade-offs. Throughout the process, the sophistication of simulation technology is matched against the maturity of the available design data. ‘The designer must make some tough choices on what technology to use and this really helps them make the best call for the overall design,’ Richard Yen, senior VP of auto engineering at Altair, added.

Dealing with multiphysics

If we drill down to the component level, then the multiphysics nature of such components and their impact on the wider vehicle performance must be addressed.

For example, let’s look at a motor rotor, a moving component of the electromagnetic system in an electric motor. Its rotation is due to the interaction between the windings and magnetic fields, which produces a torque around the rotor’s axis. As such, the design of the rotor is important with regards to the electromagnetics of the system. But, as that rotor is spinning, there are also mechanical demands to take into account. Energy is also dissipated, so you have to ensure you can remove heat or counteract its effects in the wider system. So, we have a thermal challenge too. Plus, the material properties of the rotor will change as the temperature changes.

Moseley said: ‘All these different problems are traditionally dealt with by different analysts but, in reality, they are a simultaneous problem and so that requires multiphysics analysis. That’s a big shift in thinking and it’s also a challenge as to how to get the right expertise to solve that complex problem.’

Lucid Motors recently optimised the design of an inverter, an electronic device that converts direct current (DC) stored in the battery into alternating current (AC) before sending that electricity to the three phases of the AC induction motors. 

Overheating presented the most critical issue, plus the effects of vibration, humidity and dust, when designing it. A low chip-to-coolant thermal conductivity together with a temperature balance and low pumping pressure had to be achieved to improve the component’s efficiency. So, the Lucid Motors team focused on designing an inverter cooling system to keep the temperature under control.

The team started with a conceptual design of a cooling channel with different configurations. Then, Lucid’s engineers carried out a series of design evaluations and sensitivity analysis using a fully-parametric CFD model with modeFrontier.

This optimised the design in terms of temperature reduction, lowering pressure and minimising the channel size. After deciding on an optimum channel solution, the manifold design was further optimised by including a mesh-morphing step in the modeFrontier process integration workflow. ‘The objectives were to keep pressure variations low and reduce velocity variation,’ Moseley added. 

The use of modeFrontier enabled Lucid engineers to make more power available to the inverter and increase AC from 1,200 to 1,500A. Such increased efficiencies have a knock-on effect for another core design consideration: the miniaturisation of the powertrain.

The Lucid Air has an exterior volume similar to the Audi A6, but the interior is more like the space you would find in an Audi A8 L. Moseley explained: ‘For Lucid, one of the guiding design principles is to optimise the space the occupant sits in, where we want to shrink everything back within the powertrain to give space back to the occupant.’ 

Moseley said: ‘We are using these techniques to drive the miniaturisation of the powertrain to unlock space to improve the user experience of the vehicle. It’s like a TARDIS experience. On the outside it looks like a normal vehicle but on the inside it is bigger than you would expect.’

‘We are becoming a more urbanised world and to try and base our transport system on larger vehicles is impossible. The role of electric powertrains is very important to make the urban space more palatable and the drive experience better,’ Moseley added.

While simulation is useful at many stages of electric vehicle development, it is particularly beneficial in arriving at an overall vehicle design, where such miniaturisation can be achieved. 

Walker said: ‘Designers will be thinking about how to size the components on the vehicle to achieve overall performance and range targets. It is not practical to build prototype vehicles to answer questions such as “what size battery do I need”, since the battery is so tightly integrated into the overall vehicle design. The choice has to be right from the outset and simulation is the only way to do this.’

Electric vehicles have certainly changed the complexity of the design process compared to traditional vehicles, as Merceron explained: ‘Although they no longer have complex internal combustion engines, the architecture of electric drivetrains is much more integrated into the car.

Electric vehicles are developed using Catia

'Different disciplines must work hand in hand now. For example, modifying the battery pack under the floor could change the vehicle’s crash resistance. So, electrical engineers now have to work much closer to body in white and safety engineers specifications.’

Modelling the flow of the electrical power from the battery and across the vehicle is also an issue, as Vincent Marche, marketing director at Altair, explained: ‘You need to choose the best path to ensure you reduce the weight and size of the vehicle and this requires electric and thermal physics to be taken into account [and] the designer must make some technical choices on what technology to use during the design process.’

For example, when incorporating the battery into the vehicle, such technical decisions must be made by the designer, as Yen added: ‘The weight of the battery is hard to bring down, so we need to make savings on the structural side by using materials such as aluminium and composites instead. The designer can select different materials and optimise the battery design using C123 to achieve this.’

Lightweight designs

Lightweight materials are often used to help both traditional and electrical vehicles make the necessary weight reductions. For example, engineers at ARRK took a conventional electric transmission housing, made of aluminium, and substituted this with a housing made from fibre reinforced thermoplastics in an electric vehicle. The design consisted of an optimised organo-sheet layer and UD-Tapes placement, with ribs overmolded by a standard injection process. 

The ESI PAM-FORM simulation module gave a first indication of the manufacturability of the organo sheet and highlighted the problematical areas in the stamping process. Stamping and injection molding simulation techniques were further used to assess and optimise the manufacturability and a weight reduction of 30 per cent was realised in the design.

Gearbox for an electric vehicle

Hervé Motte, R&D innovation manager at ARRK, said: ‘Function complexity of electrical gearboxes is mostly lower compared to conventional gears because it is not necessary to have a high number of gears (i.e. conventional vehicles have up to seven gears and electricals have one or two gears).’ 

Motte added: ‘Most challenging for electrical gearboxes is the acoustics because there is no combustion engine, only a silent electrical engine, which makes the gearbox acoustics dominant, especially at low velocity. The sound of a transmission (usually a whining sound) itself is annoying for the customer, so it has to be optimised. Also, weight is really essential for electrical cars to improve their range as much as possible by appropriate costs.’

Flying high

Electric cars are not the only form of vehicle under development. Electric aircraft also rely on simulation and modelling tools to optimise such designs. Electric aircraft company Ampaire, is developing a high-performance, zero-emission aircraft with the assistance of such techniques. The company’s director of Powertrain Engineering, Omar Laldin, said: ‘The most significant challenge on an aircraft is mass, which constrains the power-to-weight ratio of the vehicle. While also important in the car, aircraft systems may require five to ten times more power per kg from the overall system for take-off, including the battery and motors.

‘As a result, increasing range is significantly more challenging as the battery is limited primarily by mass, instead of available package volume. Our simulation tools allow us to investigate every early-stage innovation, even if it is not yet implemented in an industrial setting,’ he added.

The reduced development times facilitated by simulation and modelling are key, according to Laldin: ‘Models and their simulations allow for rapid development of the vehicles, significantly reducing design iterations down to one or two. As such, the overall cost and time of the projects become a fraction of what they had been. Detailed simulations allow engineers to develop more targeted solutions, often triggering the creative genius of ‘out-of-the-box’ thinkers.’

Simulations will also aid market acceptance of electric vehicles, as Laldin explained: ‘Modelling and simulations tend to facilitate wide-scale acceptance by providing visualisation to the mathematics, making it easier to convey the message to non-technical personnel and over a wide range of media.’

Sometimes you need to be in the driving seat to truly understand their integration into the automotive industry, as Laldin said: ‘Acceptance for an individual is most significantly tied to the user experience during, for instance, a test drive.’

This is why the optimisation of the user experience using simulation and modelling tools is a vital component of the design process – but proving that an electric vehicle provides a comfortable and enjoyable drive is one area where such tools cannot compete with the real world. 

So, while simulation and modelling tools optimise electric vehicles to reduce the complexity of the design process and ensure real cost and energy saving capabilities, they can only take this work so far. The automotive industry needs to get drivers in such vehicles to ensure they never get stuck on the starting line again. As Moseley said: ‘I don’t think people will ever believe it until they drive it.’

Analysis and opinion
Feature

Robert Roe looks at research from the University of Alaska that is using HPC to change the way we look at the movement of ice sheets

Feature

Robert Roe talks to cooling experts to find out what innovation lies ahead for HPC users

Analysis and opinion