Determining Fuel Type from Estimates of Bulk Modulus Using Rail Pressure Measurements
Abstract
The intention of this project is to measure fuel properties that would lead to determination whether the fuel is diesel or biodiesel. This research is focused on developing strategies to estimate the isentropic bulk modulus, which is a critical parameter for fuel characterization. Bulk modulus of fuel affects pressure rise due to pumping events differently for different fuels. With the help of sensors available on Cummins XPI fuel system, obtaining the data for pressure rise is accomplished. The pressure rise information is extracted from raw transient rail pressure data after filtering. The effects of filtering, engine speed and sampling rate have been taken into account while extracting the pressure rise estimates. The results obtained from test data have been applied on data from simulation models and from a test rig, which include high frequency oscillations due to sensor noise and rail dynamics. The bulk modulus estimation technique in this research is able to estimate the bulk modulus value for diesel fuel, with errors in the range of 0.71% to 2.89%, depending on pressure. The difference in values for bulk modulus of diesel and biodiesel is 7%, which is not high enough to accommodate the uncertainties produced from bulk modulus estimation calculations. The value of Δ V used for bulk modulus changes with changing pressure and depends on the compressibility of the fuel, which in turn is dependent on the bulk modulus of the fuel. A possible approach to tackle this problem is to develop equations that use tangent and secant bulk modulus and their inter-relationship to determine ΔV and bulk modulus simultaneously from data collected at several rail pressures. Apart from this, several recommendations have been made in order to reduce the uncertainty in the results for pressure rise as well as bulk modulus estimation.
Degree
M.S.M.E.
Advisors
Meckl, Purdue University.
Subject Area
Mechanical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.