Cylinder deactivation (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. The reduction in engine displacement during CDA improves fuel efficiency at low loads primarily through a reduction in pumping work. During deactivation of a given cylinder the drop in pres- sure inside the cylinder could possibly lead to the transport of oil from the crankcase into the cylinder owing to the reduced pressure difference between the crankcase and the cylinder. In addition, cylinder deactivation might inhibit the first fire readiness of a reactivating cylinder as a result of reduced wall, head, and piston temperatures. Both of these potential issues are quantitatively studied in this paper. This paper describes a strategy to estimate in-cylinder oil accumulation during CDA, and first fire readiness following CDA, through comparison of individual heat realease profiles before and after CDA. Cylinder cool-down and oil accumulation dur- ing deactivation could possibly result in misfire or degraded combustion upon an at- tempt to reactivate a given cylinder. Fortunately, experiments described in this paper 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.

Premixed charge compression ignition (PCCI) is a promising combustion strategy for reducing in-cylinder NOx and particulate matter formation in diesel engines without incurring fuel penalty. 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 paper outlines the development, and validation, of a real-time capable estimation strategy for diesel-fueled PCCI CA50 using production-viable measurements that do not include in-cylinder pressure. The CA50 estimation strategy considers both stages of diesel-fueled PCCI combustion—low-temperature heat release and high-temperature heat release, which contributes most to the cumulative heat released during combustion. The strategy is validated using a PCCI CA50 dataset generated with a wide range of positions of a variable geometry turbocharge, exhaust gas recirculation fractions, and intake valve closing timings. The model estimates CA50 within ±2 CAD for 65 out of 80 data points and exhibits an error standard deviation of 2.55 CAD.

Fuel efficient thermal management of diesel engine aftertreatment is a significant challenge, particularly during cold start, extended idle, urban driving, and vehicle operation in cold ambient conditions. Aftertreatment systems incorporating NOx-mitigating selective catalytic reduction and diesel oxidation catalysts must reach ∼250 °C to be effective. The primary engine-out condition that affects the ability to keep the aftertreatment components hot is the turbine outlet temperature; however, it is a combination of exhaust flow rate and turbine outlet temperature that impact the warm-up of the aftertreatment components via convective heat transfer. This article demonstrates that cylinder deactivation improves exhaust thermal management during both loaded and lightly loaded idle conditions. Coupling cylinder deactivation with flexible valve motions results in additional benefits during lightly loaded idle operation. Specifically, this article illustrates that at loaded idle, valve motion and fuel injection deactivation in three of the six cylinders enables the following: (1) a turbine outlet temperature increases from ∼190 °C to 310 °C with only a 2% fuel economy penalty compared to the most efficient six-cylinder operation and (2) a 39% reduction in fuel consumption compared to six-cylinder operation achieving the same ∼310 °C turbine out temperature. Similarly, at lightly loaded idle, the combination of valve motion and fuel injection deactivation in three of the six cylinders, intake/exhaust valve throttling, and intake valve closure modulation enables the following: (1) a turbine outlet temperature increases from ∼120 °C to 200 °C with no fuel consumption penalty compared to the most efficient six-cylinder operation and (2) turbine outlet temperatures in excess of 250 °C when internal exhaust gas recirculation is also implemented. These variable valve actuation-based strategies also outperform six-cylinder operation for aftertreatment warm-up at all catalyst bed temperatures. These benefits are primarily realized by reducing the air flow through the engine, directly resulting in higher exhaust temperatures and lower pumping penalties compared to conventional six-cylinder operation. The elevated exhaust temperatures offset exhaust flow reductions, increasing exhaust gas-to-catalyst heat transfer rates, resulting in superior aftertreatment thermal management performance.

A diode-laser-absorption-spectroscopy-based sensor system was used to perform high-speed (100 Hz to 5 kHz) measurements of gas properties (temperature, pressure, and H(2)O vapor concentration) at the turbocharger inlet and at the exhaust gas recirculation (EGR) cooler exit of a diesel engine. An earlier version of this system was previously used for high-speed measurements of gas temperature and H(2)O vapor concentration in the intake manifold of the diesel engine. A 1387.2 N m tunable distributed feedback diode laser was used to scan across multiple H(2)O absorption transitions, and the direct absorption signal was recorded using a high-speed data acquisition system. Compact optical connectors were designed to conduct simultaneous measurements in the intake manifold, the EGR cooler exit, and the turbocharger inlet of the engine. For measurements at the turbocharger inlet, these custom optical connectors survived gas temperatures as high as 800 K using a simple and passive arrangement in which the temperature-sensitive components were protected from high temperatures using ceramic insulators. This arrangement reduced system cost and complexity by eliminating the need for any active water or oil cooling. Diode-laser measurements performed during steady-state engine operation were within 5% of the thermocouple and pressure sensor measurements, and within 10% of the H(2)O concentration values derived from the CO(2) gas analyzer measurements. Measurements were also performed in the engine during transient events. In one such transient event, where a step change in fueling was introduced, the diode-laser sensor was able to capture the 30 ms change in the gas properties; the thermocouple, on the other hand, required 7.4 s to accurately reflect the change in gas conditions, while the gas analyzer required nearly 600 ms. To the best of our knowledge, this is the first implementation of such a simple and passive arrangement of high-temperature optical connectors as well as the first documented application of diode-laser absorption for high-speed gas dynamics measurements in the turbocharger inlet and EGR cooler exit of a diesel engine.

In-cylinder oxygen fraction serves as a critical control input to advanced combustion strategies, but is extremely difficult to measure on production engines. Fortunately, the in-cylinder oxygen levels can be estimated based on accurate estimates or measurements of the oxygen fraction in the intake and exhaust manifolds, the in-cylinder charge mass, and the residual mass. This paper outlines such a physically based, generalizable strategy to estimate the in-cylinder oxygen fraction from only production viable measurements or estimates of exhaust oxygen fraction, fresh air flow, charge flow, fuel flow, turbine flow and EGR flow. While several of these flows are accurately measured or estimated, significant errors in the turbine and EGR flows are commonly observed and can highly degrade the accuracy of any calculations which utilize these flows. An EGR flow estimator was developed to improve the accuracy of this flow measurement over the stock engine control module (ECM) method and is detailed in this paper. Furthermore, the in-cylinder oxygen estimation algorithm is developed, and proven, to be robust to turbine flow errors. Regulation of in-cylinder oxygen levels is of interest for not only in conventional combustion modes but also in advanced combustion strategies such as premixed charge compression ignition. The proposed oxygen fraction estimator is designed such that its performance and stability is ensured in both conventional and advanced combustion modes. The model-based observer estimates the oxygen fractions to be within 0.5% O2 and is shown to have exponential estimator error convergence with a time constant less than 0.05s, even with turbine flow errors of up to 25%.