Renewable energy sources, such as solar-thermal
or
geothermal heat, and low-/medium-grade industrial waste-heat
can be converted into useful power and/or heating with a variety
of technologies, including organic Rankine cycle (ORC
) and
vapour-compression heat-pump systems. The thermodynamic
performance and cost of these technologies depends crucially on
the efficiency of key components, including the compressor or
expander used. Reciprocating-piston machines can be
advantageous over turbomachines and other positive-
displacement machines at intermediate scales (~1
0s-100s of kW)
thanks to
their ability to operate with relatively high isentropic
efficiencies at large expansion ratios. However, modelling the
thermodynamic losses in reciprocating-piston expanders, with a
view towards designing high-performance machines,
is a
complex undertaking. The aim of this paper is to develop a
spatially-lumped, yet dynamic model of a piston expander
suitable for early-stage engineering design, that can provide
simplification without sacrificing accuracy. The unsteady heat
transfer between the gas and the cylinder walls, and the mass
leakage are predicted independently with correlations available
in the literature and simplified one-dimensional models,
respectively. However, the turbulence induced by the mass
intake through the piston rings can affect the gas-
to-wall heat
transfer. In order to address this dependency two complementary
approaches are used. Compression and expansion processes are
simulated in a gas spring configuration (i.e. without valve
systems) using a computational fluid dynamics (CFD) model
developed using the open-source code OpenFOAM, where the
loss mechanisms are solved directly. The results are then
compared with predictions from the heuristic lumped model
based on heat transfer correlations. Finally, the lumped model is
used to derive performance maps for a reciprocating-piston
expander over a range of pressure ratios and mass flow rates.