Predictive model-based control offers benefits of reduced engine calibration time as well as enabling faster system response by anticipating future engine states and acting accordingly. Transient air/fuel ratio control, or more correctly, oxygen/fuel ratio control, is seen as a major challenge to achieving EURO 4 and U.S. Federal Tier 2 NOx and particulate emissions compliance. A key component to maintaining minimal air/fuel ratio excursion from a target value during engine transients is the ability to deliver the prescribed air to the cylinder in a rapid manner. When an EGR system is used, the response time of the EGR valve(s) must be fast enough to not starve the cylinder of oxygen (which can lead to significant particulate increases).
Several different EGR valve actuators have been evaluated through bench testing to determine their response times (opening and closing) to various driving voltages. Results of some of these tests are given. The measured valve response characteristics can be used in the model-based control algorithm development to more accurately represent the total system behavior.
The model-based predictive control approach is explained in general terms as well as the specific EGR control application used here. An adaptive predictive control strategy is discussed which continually updates a linear model of the system by employing an on-line parameter estimator. Scenarios to control both boost pressure and EGR flow using a model-based predictive control are discussed. Results of transient air/fuel ratio control with open loop, PI and MBC systems are presented for the 3.0 L HSDI engine under examination. Finally, plans to implement EGR hardware on a mobile emissions chassis and to continue control model development will be discussed.
Back to Control Seminars Page.