Energy consumption simulation of electric vehicles

Electric vehicles, range, real vehicle characteristics, real conditions of use

The DieMo team of scientists discusses the consumption model developed at Fraunhofer LBF.

Range anxiety is one of the most common terms associated with e-mobility. In addition to the high initial outlay, this is usually a major obstacle to purchase and use. It is based on vehicle users’ fear that their electric vehicle’s battery capacity will not be sufficient for the planned route and they will breakdown. This is doubly difficult because the battery performance and charging infrastructure force inexperienced users out of their comfort zone.

Real consumption estimate prior to travelling

The joint project DieMo RheinMain has developed innovative services for e-mobility in the Rhine/Main area. In this context, Fraunhofer LBF has developed a simulation model which makes it possible to calculate the range of electric vehicles reliably based on a wide variety of driving conditions. By using a model-based consumption calculation in a route planner, it is possible to determine a realistic range before the journey starts.

Although general consumption information is available for each electric vehicle and can be found in the vehicle data sheets, the actual consumption can vary greatly. In winter particularly, the range available under normal driving conditions can be up to 40% lower due to the cold temperatures than the average consumption value specified. An appropriate vehicle model for calculating the energy consumption of electric vehicles was developed at Fraunhofer LBF to map this winter effect and other influencing factors such as load and route gradients.

The model is based on data from the vehicle, the drivetrain and the tires but also takes essential auxiliary equipment, such as the heating and air conditioning system, into account. The model can be flexibly adapted to different makes of vehicle using a parameterization method which has been developed based on available technical data, such as vehicle dimensions, vehicle mass, battery capacity, electric motor type and corresponding motor characteristics, including E-Smart, Nissan Leaf or even Tesla. The vehicle speed or speed profiles with acceleration and braking events, the route with relevant gradients, the outside temperature and the vehicle load serve as input variables for the energy consumption calculation. These variables are available from measured data, synthetic profiles and maps. The final output variable is the overall energy consumption for the simulated journey in kWh/km or the remaining range that is still available.

The simulated results were compared with real journeys of the Fraunhofer LBF’s electric research fleet. In particular, journeys were carried out on the LBF reference route at different outdoor temperatures. The LBF reference route consisted of urban, rural and motorway sections with different proportions of gradient. Individual measurements for the auxiliary equipment were also taken while stationary. During the real test drives, the various vehicle models demonstrated very different temperature dependencies. The studies showed that the air-conditioning and heating system has a considerable influence on the overall consumption. Load, route gradient and speed profile are also far from negligible influencing factors. At an outside temperature around 20 °C, the air-conditioning and heating system generally requires no power to keep the vehicle interior at the desired temperature. In summer at temperatures around 30 °C, the power requirement can be 0.5-1.5 kW and in winter at temperatures around -10 °C it can be 2.5-4.5 kW. These values fluctuated depending on the vehicle and the state of the technology installed. The increased energy requirement for heating the interior is one of the main reasons why the range of an electric vehicle is lower in winter.
The final model showed a good level of agreement with the real test drives for the vehicles examined. While general energy consumption information provided by manufacturers underestimated the vehicles’ consumption on the LBF reference route by on average 25%, the LBF model calculates a slightly increased demand of 5% on average compared to the real consumption figures. However, this overestimation keeps the driver on the safe side and prevents breakdowns.

The LBF reference route in Darmstadt and the surrounding area.

Auf dem Parktplatz vor einem modernen Institutsgebäude stehen verschiedene Pkw. Sie tragen alle ein Logo mit der Aufschrift "LBF-emobil" und sind Forschungsfahrzeuge.

Some of the LBF research fleet vehicles.

The Tesla under extreme road conditions.

Comparison of the simulation model with real test drives shows a good level of agreement

Integration in route planner possible

With regard to the interfaces, the newly developed Fraunhofer vehicle model has been constructed so that it can be used for integration in a route planner. It is possible to simulate the overall model “live” as part of an online application or to provide route-based consumption results for use in a database. In the second case, the energy consumption values are stored in defined stages for each individual map of the route network (e.g. temperature: incrementally from -10 °C to 30 °C, load: empty, medium, full). For this implementation, the vehicle is simulated for all route maps with many combinations of the route planner’s static (e.g. payload or vehicle type) and dynamic (e.g. temperature or traffic situation) input parameters.

Sponsors and partners

Contact