The complexity of modern engine control has increased rapidly. New software tools to support the difficult electronic control units application process are required to fulfill the limited emissions required by legislation. An in-car measurment with an online variation of the ECU maps during the driving cycles can support the application process. The paper describes the development of a digital measurement data processing system for in-car use. With the in-car data acquistion system implemented, it is possible to record data measured during arbitrary driving cycles. The acquired data are used to model the system behavior. In addition to static modelling, vehicle-specific parameters are considered by using system identification with dynamic neural networks. Consequently, the model for the nitrogen oxide emissions and the exhaust gas opacity can be used as a support tool in the application process and as a basis for an automated optimisation.
In-car modelling of emissions with dynamic artificial neural networks
Emissionsmodellierung im Fahrzeug mit Hilfe künstlicher neuronaler Netze
2001
7 Seiten, 12 Bilder, 8 Quellen
Conference paper
English
Abgasemission , adaptives System , Adaptivregelung , Datenerfassung , dynamische Optimierung , dynamisches System , dynamisches Verhalten , elektronische Steuerung und Regelung , Emissionsminderung , Fahrzeugmotor , Kennfeld , Kraftfahrzeugelektronik , neuronales Netzwerk , Pkw (Personenkraftwagen) , Stickoxid
In-Car Modelling of Emissions with Dynamic Artificial Neural Networks
SAE Technical Papers | 2001
|In-car modelling of emissions with dynamic artificial neural networks
Automotive engineering | 2001
|2001-01-3383 In-Car Modelling of Emissions with Dynamic Artificial Neural Networks
British Library Conference Proceedings | 2001
|Diesel Engine Exhaust Emissions Modelling Using Artificial Neural Networks
British Library Conference Proceedings | 1999
|Diesel Engine Exhaust Emissions Modelling Using Artificial Neural Networks
SAE Technical Papers | 1999
|