Marine Diesel engine response to ventilation loads

Abstract

Ventilation events are the result of masses of air being transported from the water surface along the hull, through the propeller plane. Previous research in literature has shown that ventilation induces large and sudden variations of the load on the propeller. The response of the propulsion system had not been documented before. This is a problem with practical and theoretical relevance as both operators and designers of ships and propulsion systems cannot predict the response of these systems to these significant and sudden variations of the propeller-load. The problem can manifest itself in different ways. A first example lies in a possible loss of propulsion and the damages that could incur. A second example lies in the possible installation of too much engine power to respond to unpredictable load-variations such as those incurred by ventilation events. Increased understanding of the relation between wave properties, -height and -frequency, and the response of a marine Diesel engine, -speed and -torque, subjected to wave-induced ventilation, is the goal of this thesis and leads to the main research question: How does a marine Diesel engine respond to off-design loads, and in particular to frequently varying loads resulting from propeller ventilation? It focuses on the response of the prime-mover to variations of the propeller-load imposed by ventilation events. The scope of this thesis covers a monohull coaster with a medium-speed marine Diesel engine moving forward in head seas. A model is proposed based on the description of immersion by Journée and Massie (2001) [19]. It consists of three sub-models that describe vessel-motion, the propeller and the prime-mover. The vessel-motion is described with a combination of potential-flow based methods and viscous theory. A quasi-static approach is proposed to describe the influence of ventilation on propeller-functioning. The prime-mover is modelled with a closed-cylinder process and an idealised first-order turbo-charger model that applies the exhaust-flow temperature of the closed-cylinder process to describe the charge-pressure. Three limits to this model lie in the application of vessel-motion data in a limited, positive domain, the application of 1st quadrant propeller data and the propeller envelope. This model uses input consisting of waves and the engine speed setpoint. The output consists of the rotational speed and produced torque of the prime-mover subjected to ventilation. Verification showed responses comparable to, and in the range and time frame of experimental results by Koushan (2007) [21]. Validation efforts lie beyond the scope of this thesis. Experiments by means of simulations have been performed for two engine speed setpoints and different wave-types describing head seas: Regular waves characterised by low frequencies, -characterised by high frequencies and -characterised by different wave amplitudes. A final experiment subjected the model to an adverse long-crested wave-spectrum for ocean waves. The research at hand found that the propeller immersion-ratio couples imposed waves and vessel-motions, to the inception of ventilation events. The quasi-static approach to model the influence of ventilation on propeller-functioning provided a reasonable estimate in verification, although further validation efforts are still required. The main research question led to the answer that: Frequently varying loads resulting from propeller ventilation can induce a significant shift in the operational point in the propellers open-water diagram and engines PV diagram and increase of the variations of engine speed and -torque. The final chapter also provides a number of pointers for a further validation effort, possible improvements to the proposed method and advice regarding further research.