ETRF surrogate gasoline fuels test bench data

Zenodo (CERN European Organization for Nuclear Research)

Abstract

The dataset contains test bench data from a single-cylinder research engine run with 10 ETRF gasoline surrogate fuels. The fuels were characterised with the same baseline calibration, related to standard RON95E10, and then the optimal fuel-wise calibration was identified. Combustion-related quantities, such as the crank angle of 50% mass fraction burned, and sensor-related quantities, such as the exhaust gas temperature and fuel mass flow rate, were used to develop two machine learning models that relate differences in these signals relative to the RON95E10 baseline to the RON and OS of the fuels. The RON and OS neural networks were embedded in a Simulink model, coupled with the test bench, and experimentally validated.

Authors 4

1. Frithjof Schomaker Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University

2. Bastian Lehrheuer Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University

3. Nicolò Cavina

   University of Bologna

   Affiliation as printed

   University of Bologna

4. Stefan Pischinger Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University