Computationally inexpensive methods of ion current signal manipulation for predicting the characteristics of engine in-cylinder pressure

Recent research on the use of ion current has focused on matching the characteristics of the in-cylinder pressure, thus avoiding the use of a pressure transducer. This paper explores techniques of calculating these pressure characteristics through the use of simple and computationally inexpensive artificial neural networks. Two neural networks are presented to deduce the in-cylinder pressure from ion current measurements, where one is used to predict the characteristics directly and the other is used to calculate the in-cylinder pressure curve. Experimental results show that both networks give satisfactory results for different purposes. Some engineering implementation issues and the further improvement of the developed techniques are discussed.