Fuzzy Logic and Neural Network Based Induction Control in a Diesel Engine
To achieve the superior real time fuel economy, to meet the consistent stringent automotive exhaust emissions or to ensure best energy utilization, lots of new technologies are being adopted by the automotive manufacturers and suppliers. This new technologies comes with more and more complex to the existing system. This leads to the increase in the calibration parameters and indirectly affects the calibration efforts too.
In addition to this, due to the deterioration or failure of the engine components like, exhaust gas treatment devices, intake devices etc., are resulting in high emissions or unexpected uncomfortable driving for the drivers. To resolve this, there is a need for a flexible and intelligent control strategy.
Currently, the available conventional control strategies use the mapping method. The calibration time is long and the work is complex when adopting this mapping method. The model based control strategies also not successful in governing the unexpected behavior in the system.
Hence, a new controller based approach for an air system, based on hybrid of fuzzy logic and neural network is proposed in this research work to control the air mass, EGR (Exhaust Gas Recirculation) ratio, boost pressure and inter cooler. This new approach is designed to be implemented in a standard ECU (Electronic Control Unit) without any change in the current engine hardware design. The fuzzy logic based controller will replace the existing conventional map based PID controller. The neural network will learn the deterioration and failures of engine components and perform online calibration for the fuzzy logic controller.
Thus the combination of fuzzy and neural network approach will help to avoid the high emissions and unexpected uncomfortable driving mode for the drivers. The proposed, new control approach, which uses the hybrid approach of fuzzy logic and neural network, is very easy to tune, simplify the development time, improve the control precision of the air system and reduce cost and time of calibration.
KeywordsFuzzy logic neural network air control intelligent control
Unable to display preview. Download preview PDF.
- 1.Arnold, J.F., Langlosis, N., Chafouk, H., Tremouliere, G.: Control of the air system of a diesel engine: A fuzzy multivariable approach. In: Proceedings of the 2006 IEEE International Conference on Control Applications, Munich, Germany, October 4-6 (2006)Google Scholar
- 2.Wijetunge, R.S., Brace, C.J., Hawley, J.G., Vaughan, N.D.: Fuzzy Logic Control of Diesel Engine Turbo charging and Exhaust Gas Recirculation. University of Bath, UKGoogle Scholar
- 3.Cui, H.: Exhaust Gas Recirculation Control in a Spark-Ignition LPG Engine Using Neural Networks. In: Proceedings of the 6th World Congress on Intelligent Control and Automation, Dalian, China, pp. 21–23 (2006)Google Scholar
- 4.Dotoli, M., Lino, P.: Fuzzy Adaptive control of a variable geometry turbocharged diesel engineGoogle Scholar
- 5.Liu, B., Huang, M., Yang, X., Xia, X.: The State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body. In: An Optimization of EGR Control System for Gasoline Using Fuzzy PID Strategy. Hunan University, ChangshaGoogle Scholar
- 6.Obodeh, O., Ajuwa, C.I.: Mechanical engineering department, Ambrose Alli University, Ekpoma, Aigeria, Calibration of Aging Diesel Engine with Artificial Neural Networks. European Journal of Scientific Research 24(4), 520–531 (2008) ISSN 1450-216XGoogle Scholar
- 7.Atkinson, C.M., Long, T.W., Hanzevack, E.L.: Virtual sensing: A neural network based intelligent performance and emissions prediction system for on-board diagnostics and engine control. In: International Congress and Exposition, Detroit, Michigan, February 23-26. SAE Technical paper series (1998)Google Scholar
- 8.Tasdemir, S., Saritas, I., Ciniviz, M., Allahverdi, N.: Artificial neural network and fuzzy expert system comparision for prediction of performance and emission parameters on a gasoline engine. Expert Systems with Applications 38, 13912–13923 (2011)Google Scholar