Cylinder Pressure Based Closed-Loop ControlCopyright: RWTH Aachen | VKA
In addition to powertrain hybridization and electrification, further optimization of the engine combustion process is required. Particularly promising is the cylinder pressure based closed loop control. This requires a very fast response to the combustion processes. For this purpose, measurement of the cylinder pressure and calculation of parameters in real time are essential. Important parameters are the indicated mean effective pressure and the combustion phasing α50. A promising approach is to perform input signal preprocessing and filtering by using field-programmable gate array (FPGA).
Due to the high performance of FPGAs it is possible to achieve significant acceleration of the control functions with minimal system latency. Future combustion processes require increasing calculation complexity. The expandability of FPGAs offers the new degrees of freedom to meet future demands.
In-Cycle combustion control is of particular interest for auto ignition of gasoline fuels GCAI (Gasoline Controlled-Auto-Ignition). GCAI allows high efficiency and very low NOx-emissions by using high conversion rates. Challenging are high fluctuations of the indicated mean effective pressure and combustion phasing. These lead to unsteady engine operation and low efficiencies. In-cycle control compensates the fluctuations, increases the efficiencies and expands the GCAI operation range.
Current research activities evaluate different in-cycle control concepts on the engine test bench. A one-cylinder research engine is equipped with MicroAutobox from dSPACE and an FPGA expansion module for engine investigations. The use of an electromagnetic valvetrain (EMVT) allows ultra-fast interventions in gas exchange of the current cycle. Fundamental engine process effects can be investigated and transferred to mass-production engines in a next step.
A state-of-the-art approach for closed-loop control of low temperature combustion processes are cycle-based control algorithms. However, these approaches allow only a stable operation in a very limited engine-map. Cycle-based controllers act such that only the system dynamics and disturbances which occur at a cycle-averaged time scale can be controlled.