Low temperature difference power systems and implications of multi-phase screw expanders in Organic Rankine Cycles

Abstract: New and old data on screw expanders operating with 2-phase mixtures in the admission line has been combined to enable the first public correlation of adiabatic expansion efficiency as a function of entry vapour fraction. Although not yet perfected, these findings have enabled an entirely new approach to the design and optimisation of Organic Rankine Cycles, ORCs. By allowing a continuous variation of vapour fraction at expander entry optima for thermal efficiency, second law efficiency and cost efficiency can be found. Consequently one can also find maxima for power output in the same dimension.This research describes a means of adapting cycle characteristics to various heat sources by varying expander inlet conditions from pure liquid expansion, through mixed fluid and saturated gas expansion, to superheated gas. Thermodynamic analysis and comparison of the above optimisations were a challenge. As most terms of merit for power cycles have been developed for high temperature applications they are often simplified by assuming infinite heat sinks. In many cases they also require specific assumptions on e.g. pinch temperatures, saturation conditions, critical temperatures etc, making accurate systematic comparison between cycles difficult. As low temperature power cycles are more sensitive to the ‘finiteness’ of source and sink than those operating with high temperatures, a substantial need arises for an investigation on which term of merit to use.Along with an investigation on terms of merit, the definition of high level reversible reference also needed revision. Second law efficiency, in the form of exergy efficiency, turned out to be impractical and of little use. A numerical approach, based on a combination of first and second law, was developed. A theory and method for the above is described. Eventually low temperature power cycle test data was compiled systematically. Despite differences in fluid, cycle, temperature levels and power levels the data correlated well enough to allow for a generalised, rough correlation on which thermal efficiency to expect as a function of utilization of source and sink availability. The correlation on thermal efficiency was used to create a graphical method to pre-estimate key economic factors for low temperature site potential in a very simple manner. A major consequence from the findings of this thesis is the reduced dependency on unique choices of process fluid to match heat source characteristics. This development significantly simplifies industrial standardisation, and thereby potentially improves cost efficiency of commercial ORC power generators.