Waste Heat Recovery in Heavy Duty Diesel Engines

Abstract: Over 50% of the energy released by burning fuel in a truck engine is lost as heat rather than being used to propel the vehicle. A promising method for capturing and reusing this heat, and thereby improving engine efficiency, is to exploit thermodynamic cycles for waste heat recovery (WHR). The goal of this thesis is to evaluate the thermodynamic performance of multiple thermodynamic cycles using many different working fluids, considering all relevant low- and high-temperature heat sources available in a heavy duty Diesel engine to be able to identify the best possible combination of heat source, working fluid and thermodynamic cycle. To evaluate the potential of each heat source, the operating conditions of a real heavy duty Diesel engine were used to define boundary conditions. A GT-Power model of such an engine was previously developed and experimentally validated for the stationary points of the European stationary cycle (ESC). Using the results from this model, an energy and exergy analysis was performed, which revealed four heat sources with the potential for waste heat recovery: the charge air cooler (CAC), the coolant flow, the exhaust gas recirculation cooler (EGRC), and the exhaust flow. Modelica models were developed for four different thermodynamic cycles: the organic Rankine cycle (ORC), the transcritical Rankine cycle (TRC), the trilateral flash cycle (TFC), and the organic flash cycle (OFC). Simulations with different boundary conditions, constraints, and engine operating conditions showed that variation in these conditions significantly affected the results obtained. In general, the best WHR performance was achieved when the thermal profiles of heat source and the chosen thermodynamic cycle were closely matched. Using realistic constraints and boundary conditions, the ORC gave the best performance with acetone, cyclopentane, or methanol as the working fluid. However, taking flammability and toxicity into account, the best-performing fluids were R1233zd(E), MM, and Novec649.

  CLICK HERE TO DOWNLOAD THE WHOLE DISSERTATION. (in PDF format)