ISSUE 011 Autumn 2021 Candela C-7 hydrofoil speedboat dossier l In conversation: Robert Hoevers l Battery recycling focus l Vehicle dynamics insight l ZeroAvia hydrogen-electric aircraft digest l Motor materials

This architecture allows for what Karkare calls “parallel but decoupled business models”, in which top hat innovations and skateboard innovations (in the battery and electric powertrain) can be decoupled, perhaps also shifting innovation costs onto skateboard suppliers. However, the architecture also throws up some dynamic challenges, he cautions, principally in coupling the top hat to the skateboard in a way that achieves the desired overall static and dynamic stiffness targets, both in the low-frequency domain that affects driving dynamics and its high frequency counterpart that impacts noise, vibration and harshness. Another, perhaps less obvious potential limitation is in packaging efficiency and flexibility when it comes to differentiating products that share the platform, AVL says. If one fixed-design battery module is used, it’s likely that there won’t be a wide spread in track width among the vehicles built on a single platform, limiting seat position options in the lateral direction. More broadly, it makes it more difficult to optimise individual models, not only in terms of internal layout but also of driving dynamics. Future directions However, Porsche Engineering anticipates more development of different vehicle types, such as saloons/ sedans, SUVs and crossovers, based on a common platform. “Different heights could be managed with different kinematic hardpoints, while the positions of electric motors and batteries can vary to achieve the best result for each vehicle type,” says Reichenecker. As with all vehicles, the dynamic behaviours of EVs can be shaped and assessed virtually using design software, models and simulations, and Karkare expects virtualisation of the design process to increase. He also anticipates maximal use of data from tests and simulations to feed machine learning algorithms, with the aim of speeding up the development process. He links this with the growing centralisation of control systems, to which electrification lends itself. “Whereas at this stage there are multiple different active systems and controllers, this is expected to move to a ‘whole vehicle control’ concept,” he says. It has been common practice to manage lateral and longitudinal dynamics separately, AVL notes. In contrast, new EV-related technologies need strong coupling of all degrees of freedom all the time and on all roads, particularly as advanced driver assistance systems (ADAS) become more common. The e-chassis as a ‘feature enabler’ has to be robust in all conditions, comfort-oriented in motion control, failsafe and energy saving, the company adds. The large number of ADAS functions, each of which needs several actuators, leads to increased complexity, as well as greater overlap and interdependence between functions, says the company. The pay-off for that is higher levels of safety through redundancy. For example, the steering can be carried out using the normal steering components, with a steer-by-brake (differential braking) system or with a steer-by-torque (torque vectoring) system, the latter two being fallback options for the first. This high level of integration also enables energy efficiency and vehicle dynamics, which have traditionally been engineered separately, to be brought together. “Optimally distributing propulsive and brake torque allows a better trade-off,” says AVL. Systems with multiple interdependencies are complex by definition, so computer simulations that can work through multiple combinations and permutations very rapidly are vital for optimising EV platforms that allow more design freedom. “Simulations have been around for a long time,” Sundstrom says. “What is interesting now though is that we are getting new ways to use the models. “Simulations allow you to explore a very wide design space compared with building physical prototypes, and if you apply these new optimisation methods and machine learning you can make that design space even wider and cover even more ground.” With more design space and faster means of exploring it, EV capabilities can only grow. Insight | EV dynamics Models allow engineers to explore wide design spaces very quickly, and the future will see Nreater use of reÄned oWtiTisation tecOniXues and TacOine learninN (Courtesy of Modelon) 48 Autumn 2021 | E-Mobility Engineering