Smart rule-based diesel engine control strategies by means of predictive driving information

Abstract

The increasing connectivity of future vehicles allows the prediction of the powertrain operational profiles. This technology will improve the transient control of the engine and its exhaust gas aftertreatment systems. This article describes the development of a rule-based algorithm for the air path control, which uses the knowledge of upcoming driving events to reduce especially [Formula: see text] and particulate (soot) emissions. In the first section of this article, the boosting and the lean [Formula: see text] trap systems of a diesel powertrain are investigated as relevant sub-systems for shorter prediction horizons, suitable for Car-to-X communication range. Reference control strategies, based on state-of-the-art engine control unit algorithms and suitable predictive control logics, are compared for the two sub-systems in a model in the loop simulation environment. The simulation driving cycles are based on Worldwide harmonized Light-duty Test Cycle and Real Driving Emissions regulations. Due to the shorter, and consequently more probable, prediction horizon and the demonstrated emission improvements, a dedicated rule-based algorithm for the air path control is developed and benchmarked in the Worldwide harmonized Light-duty Test Cycle as described in the second part of this article. Worldwide harmonized Light-duty Test Cycle test results show an improvement potential for engine-out soot and [Formula: see text] emissions of up to 5.2% and 1.2%, respectively, for the air path case and a reduction of the average fuel consumption in Real Driving Emissions of up to 1% for the lean NOx trap case. In addition, the developed rule-based algorithm allows the adjustment of the desired NOx–soot trade-off, while keeping the fuel consumption constant. The study concludes with brief recommendations for future research directions, as for example, the introduction of a prediction module for the estimation of the vehicle operational profile in the prediction horizon.