Breakdowns history: Does the digital twin of the wiring system make predictive auto maintenance possible?
15. SEP 2019
Modern cars are more reliable than ever. According to ADAC breakdown statistics, in the year 1978 23 of 1,000 cars broke down during the year, but this value has fallen to less than three in 1,000, even though the average age of cars registered in Germany has risen to 9.5 years. The enhancement of simulation techniques used in development contributed significantly to this. Using sophisticated mathematical models, the physical behavior of components can be predicted in advance, before the first prototype is even tested. That is exactly what we at LEONI do for all the important components and sub systems in on-board wiring systems. Future vehicle generations will have even more complex technology on board with electric drive systems and automated driving. Is that why the pendulum is swinging back?
Automobile breakdowns might soon be a thing of the past with the digital twin.
I am convinced that this is not the case. On the contrary: Future cars will be even more reliable, and that must be the case because when we temporarily entrust our lives to automated driving, we want to be certain that the technology on board will function under all conditions. The key to this, in my view, is to continuously monitor how technical systems age and to request maintenance proactively even before a malfunction occurs. Predictive maintenance is a possibility with the development of a digital twin. Per definition, this is a virtual representation of the real product which is linked to the real physical product. Unlike previous simulation models, the digital twin “ages” with the real vehicle, to the same degree of environmental stress to which the vehicle is actually exposed. All automobile manufacturers are working towards the development and implementation of such digital twins in the coming decade. As a system partner in terms of wiring harness we would also like to develop digital twins for our wiring systems and provide the same upon request, in the coming decade. We are already in discussions on the topic with our customers – for example at the Bordnetz Kongress in Landshut, Germany.
Several challenges must be overcome to further enhance the existing simulation methods at the component and system level in order to develop a digital twin for the entire on-board electrical system. The first and most important is to measure the actual stresses that occur due to mechanical, electrical, and thermal effects with optimal installation of sensors throughout the wiring system. Secondly, sufficient connectivity is a must in order to record and process the measured data at corresponding intervals. Furthermore, a considerable amount of work goes into minimizing the impreciseness that every model has compared to its physical counterpart. Last but not least, the issues of data security and protection pose additional challenges.
On the way to digital twin
LEONI proposes a three step approach towards the digital twin transition. First, to categorize the complexity of the digital twin while also defining its system limits. Then, to determine the relevance and consequence of deviations from the nominal state. And, finally, to decide on the technical parameters that need to be monitored.
Today’s vehicle wiring systems are extremely complex; they are made up of several thousand components. In addition, the electric and electronic components connected to the wiring system, such as actuators, control devices, and sensors, have a significant influence on how this system behaves. At LEONI, we therefore categorize the modeling approaches into four system levels. Level 1 comprises components for which there are already very precise 3D Simulation models. Level 2, the subsystem level, consists of a topologically-limited section of the wiring harness including components specific to the wiring system. Such simulation models are already partially in use at LEONI in the design and dimensioning of high voltage cable harnesses for electric vehicles. The network level, which is the level 3, includes the entire electric network responsible for power and data distribution between the connected components and control devices. As a system partner, LEONI is currently targeting on building a complete model for the network level. On the other hand, the entire vehicle system level could be categorized as the fourth level, that includes all the power and information systems connected to the wiring system, from the battery to the control device. The responsibility for this level logically lies with the OEM.
Simulation techniques like Design of Experiments aid in the optimization process for the wiring system during the pre-development phase
Simulation of optimal cable placement using various parameter influences
Prior to building a digital twin, the question of what should be achieved with monitoring during operation, must be resolved – for this is one of the ways to analytically construct the model and make related decisions on the hardware equipment required. This facilitates the indication of safety-related deviations from the nominal state in future partially, highly, or fully automated vehicles. This information is the focus of all predictive maintenance approaches. Furthermore, the data collected during operation could also be of use for development, production, and even sales. Finally, information on both the functioning of the wiring system as well as its current status could be of interest.
If these questions have been answered, the next step is to make decisions regarding: Which technical parameters are relevant to acquiring the desired information? And which sensors are used to acquire it? Our hypothesis in this regard is that the digital twin should be based on the monitoring of one parameter or a combination of multiple technical parameters. The sensors equipped must follow this hypothesis, for example recording thermal and electric parameters individually or as a combination of both.
In a few years, it will be a matter of course that cars bring themselves to the workshop before things go wrong. “Dead” vehicles will be even more rare despite the more complex technology. I am happy that my work allows me to help achieve this target.