A Dual-Source Organic Rankine Cycle (DORC) for Improved Efficiency in Conversion of Dual Low- and Mid-Grade Heat Sources

Abstract

Detailed thermodynamic and systems analyses show that a novel hybrid cycle, in which a low-grade (and low-cost) heat source (340 K to 460 K) provides the boiling enthalpy and some of the preheating while a mid-grade source (500 K to 800 K) provides the enthalpy for the final superheating, can achieve dramatic efficiency and cost advantages. Four of the more significant differences from prior bi-level cycles are that (1) only a single expander turbine (the most expensive component) is required, (2) condenser pressures are much higher, (3) the turbine inlet temperature (even with a low-grade geothermal source providing much of the energy) may be over 750 K, and (4) turbine size is reduced. The latent heat of vaporization of the working fluid and the differences in specific heats between the liquid and vapor phases make full optimization (approaching second-law limits) impossible with a single heat source. When two heat sources are utilized, this problem may be effectively solved — by essentially eliminating the pinch point. The final superheater temperature must also be increased, and novel methods have been investigated for increasing the allowable temperature limit of the working fluid by 200 to 350 K. The usable temperature limit of light alkanes may be dramatically increased by (1) accommodating hydrogen evolution from significant dehydrogenation; (2) periodically or continually removing undesired reaction products from the fluid; (3) minimizing the fraction of time the fluid spends at high temperatures. Detailed simulation results are presented for the case where (1) the low-grade heat source (such as geothermal) is 400 K and (2) the mid-grade Concentrated Solar Power (CSP) heat source is assumed to be 720 K. For an assumed condensing temperature of 305 K and working fluid flow rate of 100 kg/s, preliminary simulations give the following: (1) low-grade heat input is 25 MWT; (2) mid-grade heat input is 24 MWT; (3) the electrical output power is 13.5 MWE; and (4) the condenser rejection is only 35 MWT. For comparison, with a typical bi-level ORC generating similar power from this geothermal source alone, the low-grade heat requirement would be ∼100 MWT.