Analysis and algorithm development for diesel engine systems utilizing variable valve actuation to enable premixed charge compression ignition and cylinder deactivation
This thesis presents algorithm development and analysis strategies for diesel engine systems incorporating variable valve actuation (VVA) as a means to enable premixed charge compression ignition (PCCI) and cylinder deactivation (CDA). The VVA system enables cylinder-independent, cycle-by-cycle control of the opening timing, closing timing, and lift for both the intake and exhaust valves. PCCI is a low-temperature combustion strategy for reducing in-cylinder NOx and particulate matter (PM) formation without additional fuel penalty. This is achieved by allowing the injected diesel fuel and charge gas to premix before the mixture auto-ignites as a result of compression. This type of combustion tends to eliminate high temperature reaction zones and the fuel rich pockets which typically lead to formation of NOx and PM, respectively. However, one of the challenges in PCCI implementation is that the process does not allow direct control of the combustion timing. The crank angle of 50% heat release, known as the CA50, is generally a reasonable proxy for the quality of combustion in terms of maximum pressure rise rate, combustion noise, and fuel conversion efficiency. This thesis outlines the development, and validation, of a real-time capable estimation strategy for PCCI CA50 using production-viable measurements that do not include in-cylinder pressure. The strategy is demonstrated to estimate CA50 within 2 CAD for 65 out of 80 data points, and has an error standard deviation of 2.55 CAD. CDA is a technology that can improve the fuel economy and exhaust thermal management of compression ignition engines (diesel and natural gas), especially at low loads and engine idling conditions. At low loads, the reduction in engine displacement during CDA improves fuel efficiency primarily through a reduction in pumping work. However, with no combustion occuring in the deactivated cylinders, in-cylinder pressure drops, enabling the transport of engine lubricating oil from the crankcase into the deactivated cylinder, an undesirable result. As an example, during experiments described in this thesis, in-cylinder oil mass levels reach 72 mg to 488 mg following 20 minutes of deactivation. Cylinder cool-down and oil accumulation during deactivation could possibly result in misfire or degraded combustion upon an attempt to reactivate a given cylinder. Fortunately, experiments described in this thesis demonstrate no cases of misfire at any speed/load conditions for the CDA durations tested, specifically, 100 ft-lb load at 800 rpm and 1200 rpm with deactivation intervals of 0.5, 5, 10 and 20 minutes. Although pilot heat release in the reactivated cylinders was delayed by approximately 1 CAD after 5 minutes of CDA, the main heat release was very similar to the heat release of a continuously activated cylinder. As such, results show no first fire readiness issues at the conditions tested. The duration of time the engine could be operated in CDA mode without significant oil accumulation, and other methods to minimize oil accumulation during CDA have also been proposed.
Shaver, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our