In-Car Modelling of Emissions with Dynamic Artificial Neural Networks

Since the introduction of ECU's (Electronic Control Units) in the 1980's, the complexity of modern engine control has increased rapidly. The adaptation to newly developed diesel engines and newly composed powertrains is a long and difficult procedure. Every possible application is a balance between economical fuel consumption and the conflicting performance of emitted nitrogen oxides and particles. New software tools to support the difficult ECU application process are required to fulfil the limited emissions required by legislation. The paper describes the development of a digital measurement data processing system for in-car use. The ECU maps can be changed during the driving cycles. With the information obtained, a Dynamic Artificial Neural Network is trained and modelling of the vehicle emissions becomes possible.