In-cylinder pressure statistical analysis and digital signal processing methods for studying the combustion of a natural gas/diesel heavy-duty engine at low load conditions

Abstract

In-cylinder pressure analysis is one of the most important tools for combustion diagnosis. The dual-fuel compression ignition engines present a different in-cylinder pressure evolution for consecutive cycles due to the high cyclic variability. One source of the cyclic variability is a long ignition delay caused by a low-temperature combustion chamber (at low load conditions) and this effect led to lower efficiency, higher emissions, and driveability problems. In the present study, a six-cylinder turbocharged heavy-duty diesel engine (6.7 L) was used to acquire the in-cylinder pressure signal at low-load operating conditions. Then a statistical methodology is proposed to process the experimental pressure signal for a combustion diagnosis approach obtained at different loads for diesel mode and dual fuel mode. First, a representative sample number of consecutive thermodynamic cycles is determined, and then the cut-off frequencies for in-cylinder pressure signal digital filtering are selected by the analysis of the Fourier Transform spectrum for the test engine using only diesel fuel and natural gas/diesel mode. The results show an effective high-frequency noise diminish (related to the resonance combustion chamber) on the filtered pressure signals. Therefore, a high-quality curve of the heat release rate can be reached, which allows identifying the combustion process at low-loads operating conditions.