Edited by Rick DeMeis
Part 1 of this series covered in-vehicle networking simulation and analysis.
Part 3 goes into development of robust wiring harnesses.
Due to the increase in vehicle comfort and safety features, there is a significant demand for more reliable electrical systems. An increasing number of vehicle electrical components require a greater amount of electrical power, which has to be ensured at all times. In order to effectively meet this challenge and enable early validation of the power network quality, vehicle manufacturers are now relying on simulation-based development methodologies.
In a modern automobile it is increasingly critical to balance sometimes competing power usage requirements, as well as regulate overall energy consumption. After turning off the combustion engine for instance, there must remain a sufficient amount of energy in the vehicle battery for the starter motor to re-start the engine.
Even when the engine is not running, some electricity is still being consumed, and a sufficient power supply must be ensured for this "sleep" mode, as well as for the active operation of the electrical network. The power supply must also be structured to avoid critical voltage drop-downs and compensate for voltage-drops as quickly as possible in active mode.
Today modern electrical networks leverage power management systems, which control the network and are responsible for electrical energy distribution. The intention is to make sure that power bottlenecks are addressed in an appropriate manner, which includes monitoring the on-board battery.
Simulation of power management strategies provides the best possible development tool for power network validation. The figure below illustrates a typical vehicle power network simulation.
Simulation enables virtual test drives, "drive cycles," as well as the accurate prediction of the electrical network's behavior. Simulation considers the following input data:
Driving cycle (e. g. US06, NEDC (NEFZ in German), etc.)
Environment temperature (e. g. impact on generator)
Vehicle features (e. g. energy "consumers")
Energy storage and transformer (e. g. battery, DC/DC transformer)
Utilizing the input data, simulation provides information on the characteristics of all voltages and currents within the network for evaluation. The behavior of the battery state of charge (SOC) for example is available right after simulation has been completed. This information allows seeing whether there is a sufficient amount of energy available in the vehicle battery to restart the combustion engine.
Information on the generator's utilization is also available as part of the simulation results. As a result one can immediately determine if the generator size is sufficient, or whether the power management system has to apply additional strategies such as increasing the idle-running speed. By considering the algorithm, the power management's quality and impact can be evaluated in a very clear and straightforward manner.
The overall model and its architecture
In terms of constructing a simulation model a logical question is how should the model of a power network be designed? Conceptually the power network consists of the following components, as shown in the figure below:
Generator/alternator
Battery
Consumers (loads)
Power management
Electrical system architecture for energy budgeting simulation
The complete system is separated into two parts—software algorithms and physical components—which are being validated together during simulation of the whole power network. Two signal paths have to be distinguished: The power line, which provides the consumers with electrical power, and the consumer control by the power management. Because the networking details, such as CAN networking, can be ignored for those examinations, the figure above accounts for the load control via directly connected control signals from the power management. This is sufficient for early verification of the power management algorithm.
In the past, these tests had to be performed through conventional drive cycles with real vehicles, which are very effort- and cost-intensive.
With the new simulation-based approach, a virtual prototype of the power network can instead be used for evaluation. This approach requires both models of the physical components and the software used for power management. Both parts are consolidated within an overall model, as shown below.
Virtual prototype of vehicle power network and energy management
A highly flexible example of this comprehensive simulation approach is the Synopsys Saber development environment. The physical power network is modeled completely within Saber. The simulator includes the vehicle battery, generator, consumers, drive cycle, etc. The energy management software is consolidated with the physical network existing as an additional module. The software code, which is usually represented as a state machine, can be hand-written as well as model-based (e. g.
in Matlab and Simulink).
