17:00
20 mins
|
ORGANIC RANKINE CYCLE POWER SYSTEMS: THE PATH FROM THE CONCEPT TO CURRENT APPLICATIONS AND AN OUTLOOK TO THE FUTURE
Piero Colonna, Jaakko Larjola, Antti Uusitalo, Teemu Turunen-Saaresti, Jaakko Honkatukia, Emiliano Casati, Tiemo Mathijssen, Carsten Trapp
Abstract: The concept of an engine based on the Rankine thermodynamic cycle, whereby the fluid is an organic compound instead of water, originates from two main observations: \cite{Angelino1984_AreviewofItalianactivity,Adam1995EncyclopediaEnergyTechnology}
1) If the selection of the working fluid is an additional degree of freedom for the design of the thermodynamic cycle, the fluid can be chosen such that it is optimal from a thermodynamic and technical point of view. The properties of the fluid directly affect how well the temperature profile of the thermal energy source and sink can be matched by the corresponding cycle heating and cooling processes. The conversion efficiency of the power system in fact strongly depends on the exergy loss in both the primary heat exchanger and the condenser. Furthermore, cycle configurations that are not possible if water is the working fluid, can be contemplated. As for the advantages with respect to technical aspects, it is notable that: (i) for the majority of the ORC working fluids, the expansion process is completely dry, thus avoiding blade corrosion issues in turbines, and inherent expansion inefficiency due to condensation, (ii) the working fluid can often serve as a lubricant for rotating machinery, and, (iii) the fluid pressure
%and density
levels within the system can be selected, to a certain extent, independently of the cycle temperatures (low fluid temperature can correspond to high pressure, and vice versa).
2) For low power output, from few kW up to 1-2 MW, the realization of an an efficient steam expander is challenging: the volumetric flow can be too small, the expansion ratio comparatively large, and the specific work over the expansion is also very large, thus the design of a simple axial or radial turbine is problematic and the efficiency bound to be low. Steam volumetric expanders in turn must be complex as lubrication issues must be dealt with, and the net expansion efficiency is heavily affected by blow-by and friction losses. In displacement-type expanders, often the working fluid must be mixed with a lubricant, which decreases thermodynamic efficiency and can thermally decompose, if it flows through the evaporator. In addition, for some applications, the freezing temperature of water is too high, and the very low pressure in the condenser can lead to unfeasibly large dimensions of this component. If the working fluids is organic, the higher volumetric flow and much smaller enthalpy drop of
its expanding vapor make it possible to obtain an efficient and simpler expansion device, be it a turbine, a screw, scroll, vane, or even a piston expander.
The selection of the working fluid affects at the same time the thermodynamic performance of the system, and the design of all its components. The working fluid is also subjected to a number of other constraints, which can be more or less stringent depending on the application, namely the fluid should be: non-toxic, non-flammable, non-corrosive, cost-effective, characterized by a low or zero Global Warming Potential (GWP) and Ozone Depletion Potential (ODP), thermally stable and compatible with all the containing and sealing materials up to the cycle maximum temperature, and possibly a good lubricant, featuring also good heat transfer properties.
%Fields of applications (starting from ''anything as a external source'')
One of the main and unique advantages of ORC power conversion systems is that the technology is applicable to virtually any \emph{external} thermal energy source,\footnote{External with respect to the power system, as opposed to the internal combustion of recip engines or gas turbines.} with temperature differences between the thermal source and sink ranging from approximately 30 (Ocean Thermal Energy Converter -- OTEC) to 500~\celsius.
If large-capacity high-temperature energy conversion systems are excluded from the comparison (therefore steam power plants), competing technologies for the conversion of the mentioned energy sources are the Stirling engine, the Closed Brayton Cycle (CBC) power plant, and the externally-fired gas turbine (EFGT). For low-temperature energy sources, e.g., geothermal reservoirs or heat recovery, the Kalina cycle concept is also a potential competitor, though power plants based on this concept are at a lower development stage vs. ORC power systems \cite{Bombarda2010ApplThEng}.
% Status of research and development, general perspective
%Motivation/relevance (market/adoption) and outline of the paper
Research and development of ORC technology has been receiving an ever increasing impulse starting from the beginning of this century, together with a rapid increase of the installed power capacity, and the number and diversity of applications. This work is born out of the need for a reasoned synthesis about the evolution of this technology, its state-of-the-art, and an outlook toward the future, thus providing information on both commercial applications and active research topics. This extended abstract is a synthesis of an ample article which will soon be submitted for journal publication.
{\bf EVOLUTION}\\
The idea of using an organic fluid in a Rankine cycle for power conversation is rather old. Probably the first organic working fluid used in Rankine cycle engines is naphtha. A patent of Franck W. Ofeldt \cite{Ofeldt1886} is at the basis of several ORC engines adopting a reciprocating expander fed by a naphtha vaporizer and powering launches. Naphtha was used as fuel, working fluid and lubricant, and its use allowed avoiding the cost of the specialized operator needed for steam engines because of the much lower evaporation pressure in the boiler.
Professor Luigi D'Amelio (1893-1967), chair of fluid machinery at the University of Naples, is possibly the father of modern ORC technology. He performed extensive studies and realized several prototypes of solar and geothermal power systems of low capacity \cite{DAmelio1935,DAmelio1936Laturbinavapore_a,DAmelio1936Laturbinavapore_b,DAmelio1939Icombustibilinazionali,
DAmelio1955steamengineusing,DAmelio1963Thermalmachinesconversion}.
Dr. Bronicki met D'Amelio during his PhD studies in the late 50's\footnote{Personal communication.} and started to study the application of the ORC principle to small solar power plants,\cite{Tabor1963Smallturbinesolar} and made an important contribution by outlining for the first time the relation between the working fluid and the design of the expander in an article of an international journal \cite{Tabor1964EstablishingCriteriaFluid}. In the same period, perfluorocarbons were studied as working fluids for mini-ORC turbines \cite{Ray1966FluorochemicalsAsWorking}.
Dr. Bronicki succeeded in deploying the results of his studies in the first commercial application of mini-ORC turbogenerators, that is the powering of remote telecommunication stations and of the auxiliaries of gas pumping stations \cite{Anon.1972TurbogeneratorProvides}. The most important requirement was reliability in order to allow for a very long operation without service, while conversion efficiency was not so relevant ($\approx 5~\%$). The first units of this type, using monochlorobenzene as the working fluid, were operational in 1961 (3 kW). In the period between 1961 and 1988, thousands of these small ORC turbogenerators were installed. Successively, photovoltaic panels substituted mini-ORC turbogenerators for these applications.
The first commercial ORC power plant for the exploitation of geothermal reservoirs was commissioned in 1984 in Nevada, having a capacity of $700~\kWe$ \cite{Bronicki2007OrganicRankineCycles}. As a consequence of the oil crisis of the late 70's, many other units for geothermal power plants manufactured by several companies followed, while also the capacity of these plants gradually increased toward the multi-Megawatt range. The working fluids were mainly light hydrocarbons, chlorobenzenes, and chloro-fluoro-carbons (CFC). In this period few ORC power plants were used also for other renewable energy sources, like industrial waste heat, waste heat of engine flue gases, and solar radiation, the largest plant being built in Japan at Mitsui Engineering \& Shipbuilding, featuring a power output of $15~\MWe$ \cite{Larjola1988}.
In the 80's, intense research and development activity occurred in East Germany, Finland, France, Japan, Israel, Italy, USSR. In the US, notable developments were related to five $600~\kWe$ units for industrial heat recovery, \cite{Prasad1980} and to a concept for electricity generation for the international space station \cite{Boretz1986_SuperCriticalOrganicRankine,Angelino1993_CyclicMethylsiloxanesas}. The oil crisis also stimulated considerable investigations into the application of the ORC concept for the heat recovery from truck engines, and prototypes of mobile ORC turbogenerators of approximately $10~\kW$ were successfully tested \cite{Koplow1984,Lang2013AssessmentWasteHeat}.
Particularly relevant in Italy were the studies of Prof. G. Angelino, Prof. M. Gaia and Prof. Macchi at Politecnico di Milano, which later led to successful commercial applications \cite{Angelino1984_AreviewofItalianactivity}. In these first prototypes, axial turbines were directly coupled to an ansynchronous generator rotating at 3000 rpm. In Finland, Prof. Larjola led the development of high-speed hermetic turbogenerators in the hundreds $\kWe$ range, in which the turbine, generator and pump share the same shaft. One of the first applications was as the electrical generator for the batteries of a deep-see submersible \cite{Jokinen1998}. The shared-shaft configuration was the same as in the early mini-ORC turbogenerators for remote power applications \cite{Bronicki1988,Larjola1984}. The first commercial application of ORC power plants for the conversion of biomass, mainly residuals from wood manufacturing started occurred in the late 90's, and were born out of the studies of the Italian school on siloxanes as working fluids. The main features of these high-temperature power plants are: axial turbines (two or three stages), operating with superheated vapor at the inlet, directly coupled with the electric generator, a large degree of regeneration, indirect heating of the working fluid with an intermediate oil loop, mainly in a co-generating arrangement.
{\bf STATE OF THE ART}\\
The first years of the new century were the start of a considerable acceleration of the development of ORC technology and of its applications. The technology is still evolving in various directions, and several new applications are proposed. Applications of ORC power systems differ both in terms of power capacity (from few kW up to almost 20 MW per unit) and in terms of maximum cycle temperature difference (from the tens of degrees of OTEC power plants, to more than $300~\celsius$ of biomass conversion). Based on a collection of data, whose summary will be reported in the full paper, it is estimated that currently the installed capacity of ORC power plants is approximately $1500~\MWe$. The installed capacity has gone from $\approx 500~\MWe$ in 2000 to $\approx 1500~\MWe$ in 2012. The number of installed units grew sharply starting from 50 in 2005 to approximately 370 in 2012. Both the installed capacity and the number of installed units are increasing at an impressive pace. The different technical solutions adopted or proposed for ORC power systems are briefly described here.
{\bf Technical options}\\
\emph{Cycle configuration}. The working fluid at the outlet of the evaporator can be superheated, currently the most commonly adopted solution, or saturated. The supercritical cycle configuration has been studied, \cite{Boretz1986_SuperCriticalOrganicRankine,Angelino1993_CyclicMethylsiloxanesas,Karellas2008, Schuster2010} and only recently it has been implemented in an experimental geothermal power plant in Italy, and in a biomass fired system \cite{Spadacini2011NEWCONFIGURATIONORGANIC}.
%http://www.enel.com/en-GB/media/news/geothermal-old-energy-for-the-future/p/090027d981d3cdfe/
The supercritical cycle configuration may improve the conversion efficiency and allow for a more compact primary heat exchanger. The regenerator is also commonly adopted in high- and medium-temperature cycle configurations, requiring heavier organic compounds as working fluids. Other configurations for cogeneration have been implemented, whenever a more flexible control of the heating/electricity ratio must be obtained \cite{Bini2011Highefficiency}.
\emph{Working fluid.} Medium-capacity high-temperature systems currently utilize linear siloxanes or perfluorocarbons as working fluids. Pentane is also used, but the turbine inlet temperature has to be kept at a lower value. Toluene is the working fluid of high-temperature ORC turbogenerators in the hundreds $\kWe$ range. Very recently a large-capacity ORC power plant ($\approx 17~\MWe$) has been commissioned, and it utilizes cyclopentane as the working fluid \cite{DelTurco2011}. Medium and low temperature ORC power systems utilize R245fa if non-flammability is a requirement and the capacity rather low, or light hydrocarbons. A non-flammable azeotropic mixture of 1,1,1,3,3-pentafluorobutane and perfluoropolyether has been adopted in medium-temperature applications.
\emph{Expander.} The axial and radial turbine configurations are adopted in medium and large-capacity power plants. Recently the radial-outlow turbine configuration has also been used in a high-temperature medium-capacity system. For turbogenerators in the few hundreds of $\kWe$ range the most commonly adopted expander configuration is the single-stage centripetal turbine. For even smaller capacity (tens of $\kWe$) volumetric expanders, i.e., scroll, screw, vane and piston expander are in the development phase.
\emph{Turbogenerator assembly.} The electrical generator may be connected directly to the expander shaft, or a reduction gear is interposed. The reduction gear demands for an oil-lubrication system, as working fluids are not viscous enough for this task. The generator can be connected directly to the grid (synchronous type) if the shaft speed is low. High rotational speed of the shaft demands for an inverter. The generator may be air-cooled or the working fluid vapor serves also as coolant. The bearings of the turbogenerator shaft may be lubricated with the working fluid, or with oil. Oil should be mixed with the working fluids, therefore tight seals are needed. Liquid or vapor lubricated bearings are often of tilting pad type, in order to obtain high rotordynamic stability. Alternatively, magnetic bearings, usually of the active type, provide several advantages, but are possibly more expensive. If the turbine, generator and feed pump are on the same shaft, their assembly can be fully hermetic, if working-fluid lubricated or magnetic bearings are used. The pre-feed pump must be hermetic and the electric generator cooled by the working-fluid vapor.
\emph{Feed pump.} Single or multi-stage centrifugal pumps are mostly adopted. Volumetric pumps are often the choice in mini-ORC systems. In all cases but the hermetic turbogenerator assembly, the pump is powered by an electric motor. Very often a prefeedpump is also part of the system in order to avoid cavitation. Placing the pumps at a certain depth below the ground is also a common solution for cavitation.
\emph{Evaporator.} The working fluid maybe directly heated by the thermal energy source, or a diathermic oil loop is interposed. The intermediate loop allows for easier control and a higher degree of standardization of the ORC power block, at the cost of complication and thermodynamic losses. The evaporator is most often of the once-though type. Many heat exchanger configurations are adopted, i.e., shell \& tube, plate, etc. depending on requirements.
\emph{Condenser.} The most frequently adopted types are the finned-tube and the plate heat exchanger. The thermal energy may be rejected directly to the a ambient air or an intermediate glycole-water circuit is added if the air-coolers must be located at a certain distance from the power block. If available, lake, river or sea water are preferable as coolant, but often regulation limits their use. Especially in case of air-cooling, the power consumption of the auxiliaries is relevant and must be included in the evaluation of the plant efficiency. ORC power conversion becomes often more profitable if cogeneration of heat and/or cooling is possible. In this case the condenser is cooled by process water.
If all the alternatives briefly outlined here are considered for the design of a new system, hundreds of thousand of options are available, in principle. This shows that the optimal design of an ORC power system is rather complex. Examples of applications and related technical solutions are described and discussed in the full paper.
{\bf Applications}\\
\emph{Conversion of geothermal heat.} If the temperature level of the reservoir and/or the amount energy is limited, and if the goethermal steam is contaminated or sour, an ORC power plant is the preferred solution for electricity generation. This is the commercial application featuring the largest installed capacity of ORC power plants, because it was the first and because geothermal ORC power plants tend to be larger than for other applications.
\emph{Biomass from wood-industry, forestry and agricultural residues.} Transportation costs of the fuel dictates small capacity for this kind of power plants, therefore ORC systems are almost always chosen if the power capacity is below $\approx 3 \MWe$. The number of these ORC power plants is growing rapidly and it is propagating from central and northern Europe to other parts of the world.
\emph{Waste heat from recip engines and gas turbines.} In analogy with combined-cycle power plants, an ORC bottoming cycle can greatly improve the conversion efficiency of medium and small gas turbine and recip engines, whereby a traditional steam power plant is not feasible because of the small capacity. Up to now, bottoming ORC power plants have been installed for gas turbines in the range from approximately 0.5 to few $\MWe$ and for reciprocating engines from 1 to few $\MWe$. Very recently ORC power plants have been proposed as an alternative to steam power plants for medium-size gas turbines (25 to 45 $\MWe$), because of better operational characteristics and lower projected life-cycle costs.
\emph{Industrial waste heat recovery.} The potential for heat recovery from industrial processes is enormous. Regulatory, and technical issues are often obstacles to its implementation. Currently, several example of heat recovery by means of ORC power plants can be found in the cement, glass, and steel industry.
\emph{Biogas from industrial and municipal waste.} The gas produced in landfills is similar to biogas, but may contain chemicals and particles that would be damaging if the gas powered directly gas turbines or recip engines. In small and medium-capacity plants of this type, ORC power systems are powered by burners fuelled by biogas of municipal waste, and gas cleaning occurs at the stack. Few examples of this tyep of small power plants are in operation.
\emph{Concentrated solar power.} Also in this case, if the design capacity of the system is small to medium, ORC power plants can be optimal energy converters. In recent years photovoltaic panels have been more successful because of more favorable economics. The advantage of a thermal conversion system is that cogeneration and storage of thermal energy are possible, thus increasing the utilization factor of the installation. Few examples of solar ORC power plants exist and they are powered by flat-plate, parabolic trough, or parabolic mirror collectors.\\
{\bf FUTURE SCENARIOS}\\
Predicting the future without a crystal ball is a difficult task, and here the attempt is to provide some suggestions about future directions by looking at the past. However, we also try not to be too conservative, with the hope of inspiring new ventures. One thing is sure: ORC technology development is correlated to oil prices, and policies. This makes of course any prediction unsure.
The margin for efficiency improvement and cost reduction is large. Thermodynamics shows that considerably higher efficiency, of the order of $30~\%$ is achievable with available working fluids, but improvements are hampered by the competition with fossil-fuel energy conversion, therefore ultimately by the price of fossil fuels. In addition, the investment cost for ORC power systems does not benefit yet from economy of production, which seems at times to close the loop. Gradually conditions, which are not only dictated by the economy, are becoming more favorable, and the visible trend toward efficiency improvement of existing and new products will continue, possibly with a steepest increase, if boundary conditions become quickly more favorable. As for applications, the growth of geothermal and biomass conversion applications might be linked to the gradual geographic extension of favorable conditions.
Industrial heat recovery might lead to a spike in the growth of the installed capacity of ORC power systems, if high-potential sectors like oil and gas, will include ORC heat recovery as an integral part of their process, either by retrofitting existing plants, or by appropriately design the new ones. Very large ORC power plants in this framework are possible, and the capacity is pushed by the economy of scale. New installations might open the path also to smaller industrial heat recovery power plants, and the processes that might benefit from them are numerous.
Promising research and development efforts are directed toward the current extremes of the power capacity range. The advent of mini-ORC power systems might open up new markets and a different economy, that of production. Small standard systems produced in large series could drastically increase the efficiency of small recip engines and gas turbines on board of trucks, ship, trains, but also all stationary applications of these prime movers. The ultimate goal might be a small combined-cycle power system, whereby the prime mover and the ORC bottoming system are optimized together from the start of the design process. An argument can be made also for the use of ORC power systems in combination with concentrated solar technology. The advantage with respect to competing photovoltaic technology is that the thermal system can operate in cogeneration mode, can make use of thermal storage, and of a fossil-fuel back up system.
A particular mention must refer to supercritical $\CO2$ (sc$\CO2$) closed-cycle turbogenerators. $\CO2$ is a simple organic compound, and the thermodynamic cycle can be a highly regenerated closed Brayton cycle, or a highly regenerated supercritical cycle, thus making its classification as an ORC power system disputable. The conceptual and technological similarities between ORC and sc$\CO2$ power systems are however apparent. Advantages in term of compactness and efficiency of the sc$\CO2$ system are well understood and interesting R\&D programs are active in various parts of the world. Its higher turbine inlet temperature ($500-650~\celsius$) make it attractive for coupling with solar towers or high temperature heat recovery. It was first studied for next generation nuclear reactors, but the fate of nuclear energy is quite uncertain.
In general terms, research should be devoted to efforts aimed at the development of specialized components. Up to know ORC development has been successful because very often components are ``taken'' from other technologies, and wisely assembled, with special intervention only on the most critical components, chiefly the expander. The success of new developments and the improvement of existing technology will increasingly rely on the development of dedicated components. An area in which considerable improvement might happen is that of working fluids. Up to now there is no working fluid (pure or mixture) specifically designed for an ORC application, contrary for instance to what happens for refrigeration systems. Fluid manufacturers started only recently to consider this option. Especially designed expanders and heat exchangers are needed for some new applications, i.e., mobile applications.
An undoubted strength animating the evolution of this technology, that is, inspiring the professionals that push it, is the strong social motivation related to the intrinsic goal of promoting the spread of a renewable energy technology. This inner drive pushed many of those that are successful in this sector today also during the years of low-cost oil. They are an example for the young generation in this community.
|
