Common Rail Diesel Injectors with Nozzle Wear: Modeling and State Estimation

This study discusses model-based injection rate estimation in common rail diesel injectors exhibiting aging phenomena. Since they result in unexpected injection behavior, aging effects like coking or cavitation may impair combustion performance, which justifies the need for new modeling and estimation approaches. To predict injection characteristics, a simulation model for the bottom section of the injector is introduced, with a main focus on modeling the hydraulic components. Using rail pressure and control piston lift as inputs, a reduced model is then derived in state-space representation, which may be used for the application of an observer in hardware-in-the-loop (HIL) environments. Both models are compared and validated with experimental data, with which they show good agreement. Aging effects and nozzle wear, which result in model uncertainties, are considered using a fault model in combination with an extended Kalman filter (EKF) observer scheme. We analyze the performance of the injection rate estimate, comparing the results to simulation data with modified nozzles. Our evaluation shows that the proposed approach allows accounting in a robust manner for wear and aging effects when estimating the injection rate, demonstrating the significant benefit offered by the proposed observer.