Title :
Dynamic programming for hybrid pneumatic vehicles
Author :
Dönitz, Christian ; Vasile, Iulian ; Onder, Christopher ; Guzzella, Lino
Author_Institution :
Dept. of Mech. & Process Eng., ETH Zurich, Zurich, Switzerland
Abstract :
Pneumatic hybridization of internal combustion engines may prove to be a viable and cost-efficient alternative to electric hybridization. This paper investigates the fuel consumption reduction that is possible with this rather new concept in combination with the well-known downsizing and supercharging method. Depending on the available hardware, pneumatic modes can be based on either two-stroke or four-stroke operation. Both configurations are investigated. Similarly to electric hybrids, hybrid pneumatic engines have an energy buffer, namely the internal energy of the air pressure tank, that provides an additional degree of freedom for the propulsion system. This entails the necessity of an optimal supervisory control algorithm that chooses the mode of engine operation at every time instant of the drive cycle while guaranteeing charge sustenance. In this study, a deterministic dynamic programming algorithm is used to ensure the optimal use of the energy stored to minimize fuel consumption. Obtained results show that the combination of engine downsizing and pneumatic hybridization yields a fuel consumption reduction of up to 34% for the MVEG-95 drive cycle. Additionally, the "turbo-lag" normally associated with heavy downsizing can be overcome with this concept by using pressurized air from the tank to supercharge the engine during the speed-up of the turbocharger. A standard gasoline engine has been modified, and first measurements presented in this paper confirm the validity of the concept.
Keywords :
automobiles; dynamic programming; internal combustion engines; optimal control; pneumatic control equipment; MVEG-95 drive cycle; air pressure tank; downsizing method; dynamic programming; energy buffer; fuel consumption reduction; gasoline engine; hybrid pneumatic vehicles; internal combustion engines; optimal supervisory control algorithm; pneumatic hybridization; propulsion system; supercharging method; Dynamic programming; Fuels; Hardware; Heuristic algorithms; Internal combustion engines; Measurement standards; Petroleum; Propulsion; Supervisory control; Vehicles;
Conference_Titel :
American Control Conference, 2009. ACC '09.
Conference_Location :
St. Louis, MO
Print_ISBN :
978-1-4244-4523-3
Electronic_ISBN :
0743-1619
DOI :
10.1109/ACC.2009.5160387