| Summary: | Stringent emission regulations imposed by governments and depleting fossil fuel reserves have promoted the development of the automotive industry towards novel technologies. Various types of hybrid power plants for transport and stationary applications have emerged. The methodology of design and development of such power plants varies according to power producing components used in the systems. The practical feasibility of such power plants is a pre-requisite to any further development. This work presents thermodynamic analysis and modelling of such a novel power plant, assesses its feasibility and further discusses the development of a suitable control system. The proposed system consists of a hybrid configuration of a solid oxide fuel cell and IC engine as the main power producing components. A reformer supplies fuel gas to the fuel cell while the IC engine is supplied with a liquid fuel. The excess fuel from the fuel cell anode and the oxygen-depleted air from cathode of the fuel cell are also supplied to the engine. This gas mixture is aspirated into the engine with the balance of energy provided by the liquid fuel. The fuel cell exhaust streams are used to condition the fuel in the engine to ensure minimum pollutants and improved engine performance. Both, fuel cell and engine share the load on the system. The fuel cell operates on a base load while the engine handles majority of the transient load. This system is particularly suitable for a delivery truck or a bus cycle. Models of the system components reformer, solid oxide fuel cell, IC engine and turbocharger were developed to understand their steady state and dynamic behaviour. These models were validated against sources of literature and used to predict the effect of different operating conditions for each component. The main control parameters for each component were derived from these models. A first law analysis of the system at steady state was conducted to identify optimum operating region, verify feasibility and efficiency improvement of the system. The results suggested reduced engine fuel consumption and a 10 % improvement in system efficiency over the conventional diesel engines. Further, a second law analysis was conducted to determine the key areas of exergy losses and the rational efficiency of the system at full load operating conditions. The results indicate a rational efficiency of 25.4 % for the system. Sensitivity to changes in internal exergy losses on the system work potential was also determined. The exergy analysis indicates a potential for process optimisation as well as design improvements. This analysis provides a basis for the development of a novel control strategy based on exergy analysis and finite-time thermodynamics. A dynamic simulation of the control oriented system model identified the transient response and control parameters for the system. Based on these results, control systems were developed based on feedback control and model predictive control theories. These controllers mainly focus on air and fuel path management within the system and show an improved transient response for the system. In a hierarchical control structure for the system, the feedback controllers or the model predictive controller can perform local optimisation for the system, while a supervisory controller can perform global optimisation. The objective of the supervisory controller is to determining the load distribution between the fuel cell and the engine. A development strategy for such a top-level supervisory controller for the system is proposed. The hybrid power plant proposed in this thesis shows potential for application for transport and stationary power production with reduced emissions and fuel consumption. The first and second law of thermodynamics can both contribute to the development of a comprehensive control system. This work integrates research areas of powertrain design, thermodynamic analysis and control design. The development and design strategy followed for such a novel hybrid power plant can be useful to assess the potential of other hybrid systems as well.
|
|---|
