Non-linear time series analysis of combustion pressure data for neural network training with the concept of mutual information

Abstract

In recent years, after a period of disillusion in the eld of neural processing and adaptive algorithms, neural networks have been reconsidered for solving complex technical tasks. The problem of neural network training is the presentation of input/output data showing an appropriate information content which represent a given problem. The training of a neural structure will de nitely lead to poor results if the relation between input and output signals shows no functional dependence but a pure stochastic behaviour. This paper is concerned with the identi cation of the most relevant input-output data pairs for neural networks, using the concept of mutual information. A general, quantitative method is demonstrated for identifying the most relevant points from the transient measured data of a combustion engine. In this context mutual information is employed for the problem of determining the 50 per cent energy conversion point solely from the combustion chamber pressure during one combustion cycle.

Applying artificial neural networks (ANN) to the estimation of thermal contact conductance in the exhaust valve of internal combustion engine

“…Since heat condition has main contribution of heat transfer in this problem thus the governing equation, boundary conditions and initial condition are given by Equations (1)e (12).…”

“…These researchers used the network to determine optimal combustion time for operating engine [11]. Heister and Froehlich suggested the use of neural network modeling in calculation pressure inside cylinder at different engine speeds and times [12]. Recently O guz et al predicted diesel engine power with biodiesel using neural networks [13].…”

Cylinder Pressure Prediction of An HCCI Engine Using Deep Learning

“…Rezai et al [15] developed two different ANN models, radial basis function and forward feed, in order to predict performance and emission values of a HCCI engine operating with oxygenated fuels. Heister and Froehlich [16] proposed the use of an ANN model to predict the cylinder pressure at different engine speeds, depending on the crankshaft angle. In their study, Benneth et al [17] proposed a Repetitive Nonlinear Autoregressive Neural Network with exogenous input for restoring cylinder pressure in multi-cylinder IC motors using measured crank kinematics.…”