Derivation of Real Driving Emission Cycles based on Real-world Driving Data Using Markov Models and Threshold Accepting

The European Union has decided to bring the Real Driving Emissions (RDE) law into force in 2016. From this point onward, the air pollutants a vehicle emits under real driving conditions will be measured by means of a so-called Portable EmissionsMeasurement System(PEMS) and then used as the basis for licensing. Compared to the emission values presently determined in the New European Driving Cycle (NEDC), a significant rise can be expected. This change is on the one hand caused by a substantially more dynamic driving style prescribed by RDE regulations, and on the other hand by considerably larger variations of ambient conditions. A trend of development resulting from this conversion is the creation of test cycles conforming to RDE regulations, which enable vehicle development to adhere to the new licensing regulations. The validity of a RDE drive is gradually verified based on multiple criteria before respective emission values are determined at the end of the process. The contribution at hand presents a new approach for generating RDE substitute cycles. At first, the criterion of driving dynamics will be focussed upon. To realize this, combinatorics of a large set of real driving data will be used to generate substitute cycles, which will exhibit driving dynamics as high as possible. This specification achieves universal, vehicle independent limitation cycles featuring high emission levels. By using the described limitation cycles, a first vehicle examination concerning the fulfilment of RDE regulations is made possible.