An estimation of three-way catalyst performance using artificial neural networks during cold start

Abstract In this paper, an estimation of a three-way catalyst performance with artificial neural networks is presented. It may be an alternative approach far an on-board diagnostic system (OBD) to predict the catalyst performance. This method was tested using data sets from two specific kind of ceramic catalysts, a brand new and an old one on a laboratory bench at idle speed. Further experiments are needed for different catalyst types before the method is proposed generally. The catalyst operation during the “ cold start ” phase (the phase that the catalyst has not reached its operating conditions yet) is examined. It consists of 200 elements of catalyst inlet–outlet temperature difference (DT), hydrocarbons (HC), and carbonmonoxide (CO) and carbon dioxide (CO 2 ) emissions. The simulation: detects the values of HC, CO, CO 2 using the DT as an input to our network forms a neural network. Results showed serious indications that artificial neural networks could estimate the catalyst performance adequately depending their training process. In this paper the “ cold start ” period experimental results are presented.