A comparison of sliding mode and FRF-based observers for cylinder pressure estimation of spark ignition engine

Abstract

A sliding mode observer (SMO) is proposed for the estimation of cylinder pressure using crankshaft speed fluctuations. SMO parameters are updated using the difference between measured and observed crankshaft speed. The governing equations of cylinder pressure and crankshaft speed are described by a one-zone combustion model and crankshaft model dynamics, respectively. The observer is found to be unstable at top dead centre (TDC) due to zero combustion torque at this point. To prevent instability, observer gains are switched off near TDC and observer performance depends only on the modelling accuracies of the combustion model. This model is further improved by modelling flame development angle θ d and rapid burning angle θ b as a function of spark advance, equivalence ratio, mean speed and manifold pressure. Experimental investigations are carried out to test and validate the model on a spark ignition engine coupled with a dynamometer. The results of the SMO are compared with cylinder pressure profile estimated using a frequency response function (FRF). It is seen that the averaged relative error, using SMO, is reduced for all the test conditions in comparison with cylinder pressure estimation using FRF. This reduction in averaged relative error is more significant at the low-load, high-speed conditions where cycle-by-cycle variations are quite high.

Prediction of CI Engine Emissions Fueled with Multiwalled Carbon Nanotube-Doped Waste Cooking Oil Biodiesel using Multilayer Neural Network

“…Even so, testing of engine for all operating conditions needs more money and time. As an alternative, the numerical/ mathematical approach and neural network (NN) approach can be put to use for the investigation of engine variables associated with thermos-physical properties, performance, and emissions [20][21][22][23][24][25]. However, computational complexity limits the use of mathematical models.…”

“…Alternatively, NN can be exploited for the same as it is advantageous over other techniques, viz., generalization capability, huge data-handling ability, mapping ability, etc. [20][21][22][23][24][25]. A NN model based on a backpropagation algorithm is utilized to determine the parameters/variables of the engine operated with blended biodiesel that has appeared in a study by Yusaf et al [22].…”