Fault Diagnosis of a Turbocharged Diesel Engine with Dynamic Neural Networks and Parity Methods

R. Isermann and A. Schwarte (Germany)


Nonlinear System Identification, Neural Networks, Intelligent Data Systems, Fault Diagnosis, Application Automotive Systems, Diesel Engine, Intake-Air-System.


Modern Diesel engines with direct fuel injection and turbo charging have shown significant progress in fuel consumption, emissions and driveability. Together with exhaust gas recirculation and variable geometry turbochargers they became complicated and complex processes. Therefore, fault detection and diagnosis is of high importance. This contribution shows the development of fault detection and diagnosis methods for the intake system of a Diesel engine. By applying semi-physical dynamic neural networks, signal models and parity equations residuals are generated. Detectable deflections of these residuals lead to faults. Experiments with a 2.01 Diesel engine on a dynamic test bench have demonstrated the detection and diagnosis of several implemented faults in real time with reasonable calculation effort.

Important Links:

Go Back