Research On Model-based Fault Diagnosis For Urea-SCR Injection System

In order to cope with the increasingly stringent emission regulations，SCR technologyin diesel engine has been widely applied. Because the entire SCR post processing systemcomponents are numerous with complicated structure, the reliability of their work is veryimportant. Therefore, HJ437--2008vehicle diagnostics (OBD) systems technicalrequirements of automotive compression ignition and gas spark ignition engine was enactedin our country in2008, along with the introduction of it, research of OBD for heavy-dutydiesel engine SCR systems is becoming a hotspot in scientific research institutes andenterprises. Clearly in the OBD regulations, reprocessing system failures have to bemonitored in real time. In the SCR aftertreatment system of fault, fault of urea injectionsystem occupies a large proportion.Urea injection system mainly includes the supply module, piping and nozzles. Any faultof these can cause SCR system can’t work, which not only cause unnecessary damage to thesystem, can also lead to diesel engine emission levels, causing damage to the environment.With the research of Urea injection system based on SCR fault diagnosis technology, mainwork is as follows.The structure and working principle of Urea injection system are studied. Based onmathematical model established with analytic method, the dynamic characteristics of ureainjection system have carried on the qualitative analysis. As a theoretical basis for usingAEMSim software, the physical model of the entire injection system is set up, includingsignal input module, urea supply nozzle and pressure pipe, and so on. Model validation isdone in view of the process and building pressure injection process, using the measured dataof experiment platform, which shows that the model simulation of the dynamiccharacteristics agrees well with experimental results.Fault-injection technology is studied and various faults of SCR urea injection system are injected using the physical model established with the software of AMESim, providing atheoretical basis for fault simulation. The characteristics of faults are analyzed, andinformation is extracted, which lay a theoretical foundation for the establishment of faultdiagnosis strategy.Fault diagnosis technology is studied, and different fault diagnosis strategy areestablished according to different operating process based on the dynamic characteristics ofurea injection system. In the process of pressure buildup, method of observer with unknowninput is used aiming at the pump failure and overflow valve jam fault.First, systemidentification based on least square method is used to establish system modeling,thenaccording to establish the system, model of fault detection observer is designed.And methodof authenticity test based on Pump speed and nozzle flow is used in the process of injecting.The results have shown that it has good identification of the faults of the injection systemusing the above two methods.This paper provides an effective method to detect the faults of urea injection systemaccurately.