Development of THC Prediction Model by Using FTIR (Second Report)

- Improvement for Low Concentration and Validation on Post-catalyst -

Abstract

A novel total hydrocarbon (THC) emission concentration estimation model is proposed for reducing engine development cost and simplifying measurement systems. In our previous report, we proposed a THC estimation model using the absorbance spectra of the Fourier transform infrared (FTIR) spectroscopy. In this report, the model is improved by employing a piecewise linear method to estimate lower THC concentrations in post-catalyst exhaust gas. The prediction error for THC concentration below 100 ppmC is less than 10 ppmC under both steady state and transient operating conditions. We also validate the model using a different engine and the same accuracy was obtained.