Waste Heat Recovery from Combustion Engines based on the Rankine Cycle

Doctoral thesis, 2016

Most of the energy in the fuel burned in modern automotive internal combustion engines is lost as waste heat without contributing to the vehicle’s propulsion. In principle some of this lost energy could be captured and used to increase the vehicle’s fuel efficiency by fitting a waste heat recovery system to the engine. This thesis presents investigations into the design and functioning of waste heat recovery systems based on Rankine cycle technology for vehicular applications. To facilitate the design of such systems, the performance of different working fluids and expansion devices was investigated using a zero-dimensional model of the Rankine cycle. Simulations using this model indicated that water-based fluids should perform well when recovering waste heat from a high temperature source such as a combustion engine’s exhaust gas. In addition, evaluations based on similarity parameters indicated that displacement expanders are optimal in systems having low flow rates and high expansion pressure ratios, both of which are to be expected in vehicular systems using water as the working fluid. Organic working fluids allow higher flow rates in the cycle, making the efficient use of turbines possible. Data from the simulations using the zero-dimensional model were used to guide the design and construction of a demonstrator test bench featuring a Rankine cycle-based recovery system that recovers waste heat from the exhaust gas recirculation system of a heavy duty Diesel engine. The test bench uses water as the working fluid and a piston expander as the expansion device. The Rankine cycle’s thermal efficiency was 10%, corresponding to 1-2% of the engine’s power output. To find ways of improving the system’s performance, one-dimensional models of the expander and the system as a whole were created and then validated by comparing their output to experimental data obtained with the test bench. The expander model suggested that reducing the compression ratio would make it possible to reduce the steam inlet pressure by 30% without affecting the expander’s power output. This hypothesis was then confirmed experimentally. The expander model was used to rank the relative influence of selected steam boundary conditions and expander geometry parameters on the performance of a piston expander. The inlet pressure, steam inlet cut-off timing, expander speed and outlet pressure were found to be the most significant main effects on expander performance. It was also shown that interaction effects between steam conditions and expander geometry had considerable influence on both power output and efficiency.