Control of hybrid electric vehicles

Within the framework of the enhancement of current powertrain concepts aiming at lower fuel consumption and emission levels and a further optimization of the individual driving sensation, hybrid vehicles gain importance. In addition to the combustion engine they are equipped with at least one additional power source which does not rely on fossil fuels. An appropriate supervisory control can optimally exploit the the advantages and compensate drawbacks of the different power sources. The combination of the combustion engine with electric machines as alternative propulsion systems is the most common choice in automobiles.

Full hybrid concepts with ist powerful electric machines enable electric driving during which the combustion engine is decoupled from the drivetrain in order to avoid drag losses. The elctric driving is limited to low speeds, however. At higher power requests the combustion engine must be coupled with the drivetrain by closing a clutch. In order to achieve this coupling without a noticeable jerk and comfort reduction for the passengers during the whole vehicle lifetime, a fast and exact synchronization of the clutch plates is necessary. In cooperation with a German OEM this task was researched by the Institute. The competencies in the fields of robust control and model precitive control were made use of. Moreover an adaptation of the controller based on online system identification was analyzed. For the controller design and validation, a simulation model based on the object-oriented modeling tool Dymola was developed.

In the cooperative research project "Innovative hybrid drivetrain for europe", a hybrid vehicle is constructed on the platform of a series vehicle. The hybrid has a parallel configuration, that means the electrical machine and the internal combustion engine are parallel arranged. The combustion engine can be run in two combustion modi. Beside the normal combustion modi, a gasoline low-temperature combustion (HCCI / CAI) is possible. Therefore a further reduction of the emission is possible.

The vehicle has no GPS or any other facilities to determine the drive state in the future. Thus, the torque split has to be done due to the informations of the past and the presence. A promising approach to solve this optimization problem is the equivalent consumption minimization strategy [1]. The electrical energy is weighted with an equivalent factor and then compared with the fuel consumbtion. The quality of the reduction of the fuel consumption depends heavily on the equivalent factor. Thus at the institute of automatisiation different strategies to determine the equivalent factor are tested.