Turbocharger map reduction and estimation of effective compression ratio in a modern diesel engine utilizing flexible intake valve actuation

Karla C Stricker, Purdue University

Abstract

The gas exchange and combustion processes in a modern diesel engine are dynamically linked — outputs of the combustion process (exhaust temperature, exhaust pressure and exhaust enthalpy) directly affect inputs to the air handling system. Likewise, the gas exchange process directly influences the inputs to the combustion process, including the in-cylinder trapped mass or charge mass, charge temperature and pressure. In modern diesel engines, this complex dynamic interaction is influenced by conventional actuators like variable-geometry turbochargers and exhaust gas recirculation valves. In newer architectures, this relationship is further influenced by flexibility in the valve train, and modulation of valve opening and closing times greatly impacts mass flows through the engine. Two major drivers of the gas exchange process are the turbocharger and the engine's effective compression ratio. The turbocharger, driven by exhaust gas energy from the combustion process, drives compressed fresh air into the intake manifold, as well as determines how much exhaust gas is recirculated through the engine. The effective compression ratio is a measure of the effective in-cylinder piston motion- and momentum-induced compression of the trapped gases, and is directly impacted by modulation of intake valve closing time. Modeling and control of the gas exchange process, as well as the impact effective compression ratio has on it, is essential for the promotion and control of advanced mode combustion techniques aimed at reducing emissions while maintaining efficiency. To date, gas exchange and turbocharger modeling efforts largely make use of complex stand-alone packages or depend upon interpolation from empirically-derived look-up tables. Accurate estimation and control techniques often require use of expensive sensors that may not be commonplace in production engines. The most common method of determining effective compression ratio relies heavily on the availability of accurate in-cylinder pressure measurements, which is rare in production engines. One of the main contributions in the work presented here outlines a strategy for modeling the complex turbocharger system using analytical equations, rather than performance maps, and implements the resulting equations in a control-oriented gas exchange model. The gas exchange model is tested against experimental engine data, first using the traditional turbocharger performance maps, then using the developed analytical equations. The model results when using the analytical functions show good agreement with the experimental engine data. The gas exchange model is then leveraged to develop an estimation scheme for effective compression ratio. Rather than rely on lab-grade in-cylinder pressure measurements, this work details an estimation scheme based only on knowledge available from typical production-viable on-engine sensors. This estimation scheme is based on a high-gain observer design for a first order system, and leverages a previously validated physically-based volumetric efficiency model to determine effective compression ratio. The effective compression ratio estimation scheme is validated against experimental engine data, with steady-state errors less than 3%. Additionally, the estimator is shown to be robust to 10% uncertainty in exhaust gas recirculation flow measurements, while still converging to within one half a compression ratio in well under 4 engine cycles. This estimation scheme is applicable to engines with flexible intake valve actuation, and is an essential control input for advanced-mode combustion techniques aimed at reducing emissions while maintaining or increasing engine efficiency.

Degree

Ph.D.

Advisors

Shaver, Purdue University.

Subject Area

Mechanical engineering

COinS