Organic Rankine Cycles are the preferred choice for low to medium
temperature heat recovery due to its simplicity when compared to other
cycles such as Kalina and Flash cycles. Many waste heat recovery sites such
as in large IC engines, refineries, and process plants have multiple heat
sources existing concurrently at varying heat loads and temperatures. When
applied to multiple heat sources, the proposed ORC systems to date suffer
from system complexity, inability to utilize all the heat sources
effectively and lower thermal efficiencies which ultimately result in lower
power outputs. The primary goal of this thesis is to design, develop and
optimize next generation ORC architectures that are capable of delivering
higher power outputs with improved utilization of all the heat sources
available on site.

Recent studies have reported Series Two-Stage ORC (STORC) architectures as
a promising alternative to the complex ORC architectures proposed in
earlier studies. An advanced STORC architecture, named Transcritical
Regenerative STORC (TR-STORC) is proposed that combines supercritical
evaporation in the high-pressure side and vapor regeneration in the
low-pressure side. Cycle simulations for a 2.9MW natural gas IC engine,
using cyclopentane as the working fluid, shows that the proposed TR-STORC
delivers increased power output by up to 16% and 23% than STORC and
single-stage ORC respectively. TR-STORC maintains exergetic superiority for
all the other working fluids investigated, with maximum net power outputs
exceeding STORC and preheated ORC by 15-34% and 15-52%. Therefore, TR-STORC
holds the promise of superior performance that does not necessitate a
complicated layout. Further studies would focus on the performance
enhancement of TR-STORC by improving the vapor regeneration process using
ejectors. Experimental validation of these advanced ORC layouts would be
carried out using a 50kW Transcritical Two-stage ORC test rig, which is
currently being commissioned at New Rummy Game.