GB2517009A

GB2517009A - Apparatus and method for storing energy

Info

Publication number: GB2517009A
Application number: GB1405928.1A
Authority: GB
Inventors: Jonathan Sebastian Howes; James Macnaghten
Original assignee: Isentropic Ltd
Current assignee: Isentropic Ltd
Priority date: 2013-04-05
Filing date: 2014-04-02
Publication date: 2015-02-11
Anticipated expiration: 2034-04-02
Also published as: GB2517009B; GB201306146D0; WO2014162129A1; GB201405928D0; US20160032783A1; CN105102768A; EP2981682A1

Abstract

A thermal energy storage system has a first vessel 10 and a second vessel 20 connected by power machinery 30. Working fluid is stored as a saturated liquid / vapour mixture. During storage, vapour from the first vessel is compressed, and then transferred to the second vessel via a regenerator 50. The temperature and pressure of the working fluid in the second vessel 20 increases, and the temperature in the first vessel 10 decreases during storage. The stored energy is recovered by the vapour returning through the regenerator 50 and power machinery 30 where it expands to produce work before condensing back into the first vessel, possibly through a second regenerator 40. The regenerator 50 comprises a gas permeable, solid thermal storage medium which, during storage, stores superheat and some latent heat from the vapour passing through it. The storage medium exhibits continuously increasing temperature profiles during storage and a small temperature difference with the surrounding vapour, thereby minimising irreversible losses during the thermal energy transfers.

Description

Apparatus and Method for Storing Energy

Field of the invention

This invention relates to a system and method for storing and recovering energy, especially for storing and recovering energy using thermal storage.

Background to the invention

The inflexibility in power generation provided by nuclear, solar and wind based renewable energy sources has posed a number of problems in using these technologies to provide a major part of a national or regional power grid. One such problem is the need to store and recover electrical energy to prevent disruptions to the electricity supply in the light of the variable energy supply, with particular issues being the cost of energy storage, the flexibility and the energy densities achievable. A number of energy storage technologies are now being developed to address such issues including pumped hydro storage, flywheel storage, compressed air energy storage CAES (including isothermal and adiabatic CAES) and pumped heat energy storage, as in accordance with Applicant's earlier application No. WO2009/044139.

In the 1920's and 1930's a number of energy storage and recovery systems based on steam were described in published patent specifications by Fritz Marguerre.

One of the earliest, GB167763, describes a system in which steam from a lower temperature water accumulator is forced by power machinery acting as a compressor into a higher temperature water accumulator, where it condenses to a liquid due to the the saturation temperature and pressure rising within that accumulator during storage as more energy is stored. The stored energy is recovered by the power machinery allowing steam from that accumulator to expand through it to produce work before returning to and condensing in the lower temperature water accumulator. One of the latest, GB423093, describes the same system augmented with a superheater accumulator disposed between the power machinery and the higher temperature water accumulator for storage and return of superheat to and from a liquid in the superheater accumulator.

The present invention is directed towards providing an improved energy storage and recovery system and method, and in particular, a system and method in which irreversible losses during thermal energy transfers are minimised.

Summary of the Invention

The present invention provides an energy storage and recovery system comprising: a first vessel configured to store a working fluid as a saturated liquid/vapour mixture L1 having a temperature TLI, a second vessel configured to store the working fluid as a saturated liquid/vapour mixture L2 having a temperature TL2, power machinery disposed between the first and second vessels, and a regenerator disposed between the power machinery and the liquid stored in the second vessel, wherein the system is configured such that: (i) in a storage mode, working fluid vapour passes from the first vessel to the power machinery where it is compressed before passing through the regenerator and condensing in the liquid [2 in the second vessel, so as to produce a progressive increase in 112 of L2and in its liquid/vapour equilibrium phase change temperature during the storage mode; and, (ü) in a recovery mode, working fluid vapour passes from the second vessel, through the regenerator to the power machinery where it is expanded to produce power before condensing in the liquid L1 in the first vessel, so as to produce a progressive decrease in TL2 of L2and in its liquid/vapour equilibrium phase change temperature during the recovery mode; characterised in that: the regenerator comprises a solid thermal storage medium through which the working fluid vapour passes for direct heat transfer between the vapour and solid medium so as to store and return superheat during the storage and recovery modes, respectively, and, wherein the system is configured such that some condensation takes place in the regenerator during the storage mode.

As thermal energy is stored in the saturated liquid/vapour mixture L2, its TL2 and vapour pressure increases progressively (i.e. gradually) with charging overtime, and decreases progressively with discharging overtime, with the result that the liquid/vapour equilibrium phase change temperature "marches" with time. In the present invention, this effect is utilised in a regenerator disposed on the higher pressure side of the system to achieve efficient thermal energy storage (e.g. of superheat) and round-trip efficiency, this being enabled by a through flow regenerator having a solid thermal storage medium allowing direct heat transfer between the vapour and solid (gas permeable e.g. porous) medium.

The regenerator/store of the present system stores both superheat and latent heat along the solid thermal storage medium of the regenerator in a highly controlled manner (a porous mass store, being in intimate contact with the vapour, always self-minimises temperature differences so as to maximise reversibility) in a respective upstream superheat transfer region and downstream latent heat transfer region, where the temperature profile of the solid thermal storage medium progressively increases in temperature in both regions during the storage mode. Thus, the entire store is actively storing heat ("thermally active") during storage, as compared to prior art stores where gas still has to pass through inactive regions with associated pressure losses (i.e. a fully charged region and fully uncharged region upstream and downstream of the thermal front respectively). Furthermore and importantly, the specific combination of a porous mass heat store with a marching temperature is able to encourage a minimum heat exchange temperature difference (AT) throughout the store, as discussed further in the specific embodiments below.

In one embodiment, the system is configured such that condensation takes place in the regenerator for the full running time of the storage mode, i.e. 100%, so that the storage mode always involves some condensation in the regenerator. In this way, a superheat transfer region and respectively downstream latent heat transfer region develop across the regenerator (the latter region progressing downstream with time) and the latter region is never allowed to leave the regenerator during the storage mode, so that thermal storage may occur with better efficiency for the entire storage mode.

However, if necessary, the system may be configured to run such that the latent heat transfer region is allowed to leave the regenerator so that superheated vapour exits the regenerator; in that case, the system is configured such that condensation takes place in the regenerator (in an optimised manner) for at least the first 50%, or 75%, or 90% of the full running time of the storage mode.

In one embodiment, the system is configured to cease operating in the storage mode when condensation is only occurring in the last 5% or less of the downstream length of the regenerator (i.e. adjacent the downstream outlet thereof), or even the last 2% or less of the downstream length of the regenerator.

In one embodiment, the system is configured to cease operating in the storage mode when condensation is about to finish in the regenerator such that some superheated gas is about to start exiting the regenerator.

Discussion of Thermal Transfer Processes The fact that the liquid temperature L2 marches during condensation (if at least some latent heat of condensation is allowed to reside therein), or during evaporation (if the liquid supplying the latent heat of evaporation is allowed to progressively cool) is highly significant. A marching, (preferably continuously) increasing liquid temperature during condensation inevitably produces a feedback reaction that increases the vapour pressure thereby facilitating the difference in temperature required for further condensation. A marching, (preferably continuously) decreasing liquid temperature during evaporation inevitably produces a feedback reaction that decreases the vapour pressure thereby facilitating further evaporation.

However, if a regenerator is placed in the path of the cooling and condensing vapour, the temperature of the regenerator will also similarly march (gradually increase or decrease with time), but will lag the vapour temperature. This means that it is always cooler than the vapour and can always provide the requisite cooling, but desirably with a matched very small temperature difference, which will only change slowly with time thereby reducing losses.

Similarly, when a regenerator is placed in the path of the evaporating vapour, the temperature of the regenerator will also similarly march (gradually decrease with time), but will slightly lag the temperature of the evaporating vapour. This conveniently always provides a superheat, either the small one desirable for the LP side, or a much larger degree of superheat on the HP side that will add significant superheating commensurate with an adiabatic expansion at constant entropy such that the post expansion condition is either just saturated or still slightly superheated.

In the case where, during condensation, TL2of liquid [2 increases during charging (or TLI of Liquid L1 increases during discharging), a regenerator installed in the path of the cooling condensing gas will have a temperature which also increases with time during charging or discharging, but wherein the temperature thereof lags that of the condensing vapour. (By lags is meant chronologically follows behind in time; by leads it is meant precedes chronologically in time.) In the case where, during evaporation, TL2 of Liquid [2 decreases during discharging (or TL1 of liquid L increases during charging), a regenerator installed in the path of the evaporating gas will have a temperature which also decreases with time during charging or discharging, but wherein the temperature thereof lags that of the evaporating vapour.

To explain in more detail: The heat store/regenerator, and indeed the liquid, temperatures always lag the vapour temperature (that is, chronologically in time) and it is this that causes heat to move between vapour and heat store. When temperatures are rising during condensation, the vapour will always be warmer than the liquid and regenerator, that is, it will move to a warmer temperature ahead of the regenerator/liquid.

When temperatures are cooling during evaporation, the vapour will always be colder and will move to a cooler temperature ahead of the regenerator/liquid.

Charging: As the cold liquid boils off, this is caused by progressive pressure reduction which, in turn, means the vapour is progressively cooler before it passes through the cold heat store, therefore this heat store is warmer than the vapour and so the vapour is heated relative to its temperature at boil-off, i.e., the heat store temperature is lagging the temperature change of the vapour, which is now superheated. The vapour is then compressed, gets hotter and the pressure on the cold side is further reduced (the compressor has taken vapour from this side, pressure and temperature reduce as above) and the pressure on the hot side is increased, i.e., the pressure ratio is increased over that at which the preceding bit of vapour entered the hot side. This means that the vapour leaving the compressor is now slightly hotter than the vapour that preceded it.

The hottest part of the hot store (regenerator) is therefore cooler than this latest vapour and so passing the vapour through it now cools the vapour to the point of saturation (and then a little bit more usually such that a small fraction condenses) before it enters the hot side liquid. As this saturated vapour is now hotter, due to the increased pressure on the hot side, than the chronologically preceding hot side vapour it is also hotter than the hot side liquid and so passing it through the liquid will cause it to condense, whereupon it gives up its heat of vaporisation and heats the hot side liquid a bit more, increasing the pressure.

The heating or cooling of the vapour is an intrinsic result of the way in which the temperatures march during charge and vapour, liquid and heat store temperatures all march with the correct sign of temperature difference to make operation possible, both on charge and discharge. An exactly similar sequence can be presented for discharge.

While ideally TLI increases on discharging and decreases during charging, in practice, however, low temps and pressures of liquid LI can cause difficulties. On the LP side, the reducing pressure is likely to cause the compressor operating pressure ratio limit of operation to be reached more quickly than its upper pressure limit for operation and hence, steps are preferably taken to mitigate this so the temperature and hence pressure of the liquid on the LP side falls or rises at a much slower rate with gas mass transferred (by evaporation) change than for the H P side; for example, the rate of decrease of T[l with time may be managed so that it is significantly less than the rate of decrease of between TL2 (e.g. less than 50%, more preferably less than 25% or 15%).

In one embodiment, the system comprises additional thermal ballast in the first vessel or a temperature regulating sub-system associated with the first vessel, configured to reduce the respective rates of progressive decrease and increase in TLI of L1 during the storage and recovery modes, respectively. The thermal ballast may be solid ballast contained in the first vessel, or an excess of liquid may be kept as thermal ballast on the LP side such that, for example, no more than 40%, preferably no more than 20% of the liquid on the LF side evaporates upon fully charging. Alternatively, external heat may be added to the liquid on the [P side, or the liquid on [P side may be in communication with an external tank that allows only a small proportion of its content to circulate through the vessel where liquid L is undergoing evaporation/condensation.

In one embodiment, a solid thermal media or phase change media is provided in the liquid on the LP side in order to act as thermal ballast (supplementary heat capacity) providing some latent heat for evaporation such that the rate of reduction of the liquid temperature and hence pressure in the [P side is reduced.

If required in certain instances (for example, to give a slower rise of peak pressure), the system may comprise additional thermal ballast in the second vessel (which may be as described elsewhere for the first vessel) or a temperature regulating sub-system (e.g. heat exchange system) associated with the second vessel configured to reduce, but obviously not eliminate, the respective rates of progressive increase and decrease in TL2 of L2during the storage and recovery modes, respectively.

Turning to the regenerator, this may be made from a variety of solid materials that could include metals, refractory materials, ceramics or organic materials. For a very responsive regenerator it is preferable to use one containing a solid material that has a very high surface area to volume, such as a regenerator made using a metal foam or layers of fine woven metal cloth, although any suitable gas permeable solid substrate e.g. porous medium (e.g. particulate bed) or matrix may be used. The principle is that the vapour passes through the regenerator transferring heat to or from the regenerator material. The more responsive the regenerator, the lower the temperature difference that occurs between the vapour and the solid.

In one embodiment, the regenerator comprises a porous matrix of solid thermal storage medium. In one embodiment, the solid thermal storage medium comprises a particulate packed bed; where the particles may have an average width of less than 6mm, and may be continuous or segregated into layers, such as, for example, individual layers of particles supported on mesh so as to avoid ratcheting.

In a highly preferred embodiment, the system is configured such that there is a progressive decrease in TLI of L1 and in its liquid/vapour equilibrium phase change temperature during the storage mode, and, a progressive increase in TLI of Li and in its liquid/vapour equilibrium phase change temperature during the recovery mode. (This may also be expressed in terms of increasing and decreasing temperature and vapour pressure.) In one embodiment, a further regenerator is disposed between the power machinery and the liquid stored in the first vessel.

The further regenerator only stores and returns superheat during the recovery and storage modes, respectively.

In one embodiment, the regenerator is located inside the second vessel above the liquid and/or the further regenerator, if present, is located inside the first vessel above the liquid. The regenerator or further regenerator may be provided above the entire cross section of the liquid inside the respective vessel (as opposed to a necked location) and this will avoid pressure losses associated with a separately connected regenerator, as well as separate pressurisation and insulation for it. Regardless of whether the regenerator is inside or outside of a vessel, the system should be configured to ensure draining of condensate into the respective liquid.

The size of the hot store/regenerator may be selected to match the total sensible heat requiring storage during a full charge. In this way, the hot store/regenerator may be sized to the average total mass of fluid transferred normally during each mode. The size of the cold store/regenerator may be sufficient to provide superheating of not more than 10K or 20K, or even up to 100K or 200K or 300K, depending upon the choice of working fluid and overall peak pressure ratio.

Preferably, in the energy storage mode, the difference in temperature between TLI and TL2 increases with time with as more energy is stored; and/or, in the energy recovery mode the difference in temperature between TLI and TL2 decreases with time with as more energy is recovered.

During charging and/or discharging TL2 is usually greater than TL1, but they may be equal at the start of the storage mode and end of the recovery mode. The terms TLI and TL2 are conveniently used to refer to the varying temperatures of liquid L1 and liquid L2at a point in time. Except for the uncharged/discharged (starting) condition, TL2 is usually always greater than TLI because compression will increase vapour pressure and temperature of liquid L2 relative to liquid Li.

TL1 will naturally decrease progressively with charging over time and increase progressively with discharging overtime, unless steps are taken e.g. to add external heat or thermal ballast. A marching temperature is desirable on the LP side and HP side during condensation and vaporisation for a more reversible and hence efficient system.

As mentioned earlier, thermal ballast will reduce the rate of marching on either side, if required. External heat from an auxiliary system added during charging may also reduce the rate of marching, or, on the LP side, totally eliminate any marching. The system may therefore be configured such that TLI of Li is maintained substantially constant for some or all of the storage mode, and/or for some or all of the recovery mode. In that case, there is little advantage to providing a regenerator disposed between the power machinery and the liquid stored in the first vessel.

The LP side vessel may be connected to a supply tank (e.g. infinite supply) that feeds in and out working fluid (e.g. continuously) such that some can be replaced with working fluid at a different temperature; this involves mass exchange and temperature exchange with the surroundings and may be used to reduce or eliminate marching of the L1 temperature and pressure.

The LP side vessel may be connected to a larger supply tank that supplies a finite supply of working fluid, whereby marching of the Li temperature and pressure is merely reduced and pressure restrictions or a feed pump or the like between the supply tank and L1 vessel allow a marching pressure in that vessel.

The LP side vessel may be sealed but include a heat exchange sub-system (internal or external to that vessel) that reduces or eliminates marching of the L1 temperature and pressure; this will involve temperature exchange with the surroundings but no mass exchange.

In one embodiment, the system is configured to lose waste heat and may include a device to do this such as for example a heat exchanger in the circuit or auxiliary circuit.

In one embodiment, the system is configured to lose waste heat from the first vessel by allowing working fluid vapour L1 to vent to atmosphere or to connected apparatus, optionally, a waste heat recapture sub-system, when its vapour pressure exceeds atmospheric pressure or the pressure in the sub-system or connected apparatus, respectively. Usually, this will be at the end of the recovery stage.

In one embodiment, the working fluid liquids L1 and L2 are each initially preheated or precooled to respective selected temperatures before commencement of the storage mode, optionally by stopping a preceding recovery mode at a selected point. In this way the next storage mode will commence with the system part-charged and with the liquids in the first and second vessels at suitable temperatures. These may be linked temperatures that would eventually have been achieved starting with both liquids at the same temperature, for example, ambient temperature or another selected temperature or, for example, starting with a deliberately raised or lowered temperature depending upon the working fluid selected. In the case of a water/steam mixture, storage preferably starts with both liquids pre-heated to about 100°C.

The working fluid liquids L1 and L2 on the LF and HF sides may start charging at an equal temperature and vapour pressure.

In one embodiment, the working fluid liquids L1 and L2 are initially both pre-heated at the start of storage, usually with TL2 still greater than TLI (for example, where freezing of L1 during storage needs to be avoided). Starting warmer makes the whole process take place at an average higher pressure and hence higher power density and hence more work per cycle. This means that for the same energy stored, a lower final pressure ratio will be required.

Any working fluid must be kept above its triple point temperature. In a preferred embodiment, the working fluid comprises a water/steam mixture, preferably, a pure water/steam mixture. In the case of water, charging and/or discharging should be conducted such that both liquids are kept above 0° C. Alternatively, the working fluid could be nitrogen or air; in either case, charging and/or discharging is conducted such that both liquids are kept at cryogenic temperatures.

Preferably, the system is an evacuated environment with no other vaporisable or non-vaporisable species present. The working fluid liquid to be evaporated should exist as a saturated liquid/vapour mixture, so that it will readily boil off vapour if there is a decrease in pressure above the liquid (or other impetus supplied).

Usually, the working liquid on each side will have the same composition, and usually, no other liquids or vapours will be present on either side.

Usually, the working fluid on the [P side and/or the HP side will be sealed and hence retained in the respective side. Where the working liquid is sealed in either side, its mass will progressively change (march) during charging or discharging. If no steps are taken to adjust the working liquid temperature on a side, its temperature and vapour pressure will similarly march on that side.

The vapour pressure of L2, and preferably, the vapour pressures of liquids L1 and L2 vary during charging, and compression may be conducted using a variable pressure ratio compressor. At least one compressor, but two or three or more stages may be used depending on the operational range of the working fluid. The above comments apply equally to an expander in respect of the expansion during discharging.

Preferably, the compression is powered by electrical energy and/or the expansion generates electrical energy. This may be an a.c. input e.g. grid or d.c. input e.g. from photovoltaic panels. However, the energy storage system may be powered by mechanical energy e.g. wind turbine and may be used to generate mechanical energy e.g. ship propulsion.

The energy storage system will store and return energy (e.g. electricity) and be operable in both energy storage and energy recovery mode. Preferably, it will be capable of switching rapidly between those two modes. To that end, reversible compression/expansion power machinery may be used, preferably reversible positive displacement machinery. This may be, for example, a positive displacement device, preferably a linearly reciprocating piston based device. Alternatively, separate machines connected (e.g. in parallel) in the energy storage system may be used for the respective compressing and expanding functions.

Preferably, the compression during the charging mode and/or the expansion during the discharging mode is substantially isentropic (or reversibly adiabatic).

Preferably, substantially no heat transfer in or out of the gas (for example through the compressor walls) is allowed to take place, so that the work of compression or expansion corresponds to nearly the maximum temperature rise or fall achievable. In the energy storage and recovery methods above, preferably the compression and expansion are conducted using reversible machinery.

It will be evident that the pressure on the HP side when the system is fully charged should not exceed the critical pressure of the working fluid. The pressure and temperature on the HP side should not exceed compressor/expander operational range or the working temperature of the materials.

The above energy storage and/or energy recovery steps will usually occur as a continuous process, but may be repeated batchwise.

Preferably, liquid L1 and/or liquid L2 is caused to evaporate as a vapour by a reduction in pressure above the liquid. Ideally, the pressure reduction is carried out by the machinery that also carries out the subsequent compression step.

Preferably condensation of a vapour into a liquid in either or both condensation steps is promoted by introducing the vapour into the liquid from below its surface. Ideally, the vapour is caused to pass through a bubble-promoting forming device disposed in the liquid.

Preferably, respective one way valves are provided above liquid Li and/or liquid L2 (e.g. in a vessel containing the liquid provided with an outlet above the liquid) to respectively permit condensing vapour to re-enter the liquid from below its surface and evaporating vapour to leave the liquid from above its surface.

In accordance with the present invention, there is further provided a method of storing and recovering energy wherein, in a storage mode, working fluid vapour from a first vessel containing a saturated working liquid/vapour mixture is compressed by power machinery and passed, via a regenerator, into a second vessel, where it condenses into a saturated working liquid/vapour mixture, the temperature and vapour pressure of which increases as more energy is stored therein, and wherein, in a recovery mode, stored energy is recovered by evaporation of vapour from the saturated working liquid/vapour mixture in the second vessel such that the temperature and vapour pressure of the mixture decreases, the vapour returning through the regenerator and expanding in power machinery so as to produce work before condensing back into the saturated working liquid/vapour mixture of the first vessel, wherein the regenerator comprises a gas permeable, solid thermal storage medium, and wherein during the storage mode superheat and latent heat are stored along the solid thermal storage medium in a respective upstream superheat transfer region and downstream latent heat transfer region, and wherein the temperature profile of the solid thermal storage medium increases in temperature in both regions during the storage mode.

In accordance with the present invention, there is further provided a method of operating an energy storage and recovery system as described above, wherein during the storage mode superheat and latent heat are stored along the solid thermal storage medium of the regenerator in a respective upstream superheat transfer region and downstream latent heat transfer region, and wherein the temperature profile of the solid thermal storage medium progressively increases in temperature in both regions during the storage mode.

In one embodiment, during the storage mode the temperature difference AT between the vapour and the solid thermal storage medium that it contacts is generally less than 15°C, or preferably less than 10°C, or even less than 5°C, at at least one selected downstream position in the superheat transfer region and/or the latent heat transfer region, and preferably at all downstream positions.

As indicated above, the stored thermal energy is be stored by direct transfer to thermal media in a thermal store/regenerator through which the gas passes for subsequent return, and the heating or cooling steps are usually conducted isobarically.

In accordance with the present invention, there is further provided a method of operating an energy storage system using a working fluid that undergoes a phase change, the method comprising storing energy in an energy storage/ charging mode comprising the steps of: i) evaporating an amount of a saturated liquid L1 having a temperature TLI to form a vapour, ii) optionally, heating the vapour to superheat it, iU) doing work by compressing the vapour to a higher temperature and pressure, iv) cooling and condensing an amount of the compressed vapour to a liquid [2 having a temperature TL2 so that there is a transfer of thermal energy from the vapour to the liquid, and further comprising recovering the energy in an energy recovery/discharging mode comprising the steps of: v) evaporating an amount of saturated liquid [2 at a temperature TL2 to form a vapour, vi) heating the vapour to superheat it, vii) expanding the vapour to a lower pressure and temperature to generate work, viii) optionally, cooling the vapour, ix) condensing an amount of the expanded vapour back to a liquid L1 having a temperature TLI; wherein the energy storage system comprises a lower pressure LP side in which the working fluid is present as a liquid/vapour mixture L1 at a lower vapour pressure, and a higher pressure HP side in which the working fluid is present as a liquid/vapour mixture L2 at a higher vapour pressure, separated by at least one compressor/expander operating so as to transfer vapour between the respective sides at the respective pressures, wherein during charging and discharging TL2 is greater than TL1, wherein heating step vi) uses stored thermal energy, wherein the stored thermal energy is stored by direct transfer to thermal media in a regenerator through which the gas passes for subsequent return, the regenerator comprising a through-flow regenerator having a solid thermal storage medium so as to allow direct heat transfer between the vapour and solid medium, and wherein (substantially) all sensible heat transfer upon cooling prior to condensing occurs within the regenerator, and preferably a minor amount of condensation occurs there too (e.g. at least 20% or less or even 10% or less of total condensation during the storage mode).

Brief Description of the Figures

Specific embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings, in which: Figure 1 illustrates a comparative energy storage system; Figures 2a and 2b illustrate the system of Figure 1 respectively charging and discharging; Figures 3a and 3b are 1-s (Temperature-Entropy) diagrams for the system of Figure 1 respectively charging and discharging; Figure 4 illustrates an energy storage system in accordance with a first embodiment of the present invention; Figure 5 illustrates an energy storage system in accordance with a second embodiment of the present invention; Figures 6a to 6c are graphs showing temperature profiles of comparative thermal energy stores; Figures 7a to 7c are graphs showing temperature profiles of the regenerator of the system according to the first embodiment of the present invention; Figure 8 is the T-s diagram for the system according to the second embodiment of the present invention; Figure 9 is a schematic showing the energy storage system according to the second embodiment divided into sections 1 to 4, while Figures 10 and 11 are the respective T-s diagram and temperature graphs illustrating the temperatures in those sections; Figure 12 illustrates an energy storage system in accordance with a third embodiment of the present invention; Figure 13 illustrates an energy storage system in accordance with a fourth embodiment of the present invention; Figure 14 illustrates an energy storage system in accordance with a fifth embodiment of the present invention; Figure 15 illustrates an energy storage system in accordance with a sixth embodiment of the present invention; and, Figure 16 illustrates an energy storage system in accordance with a seventh embodiment of the present invention.

Detailed Description of Specific Embodiments

Figures Ito 3 (Comparative) As referenced above, GB167763 (to Marguerre) describes a system in which steam from a lower temperature water accumulator is forced by power machinery acting as a compressor into a higher temperature water accumulator, where it is forced to condense, the temperature and pressure in this accumulator rising during storage. The stored energy is recovered by the power machinery allowing steam from that accumulator to expand through it to produce work before returning to and condensing in the lower temperature water accumulator. The advantages of such a system is that work is performed directly on the working fluid and that energy is conveniently stored in it as a liquid in a small volume.

Figure 1 Figure 1 illustrates a similar energy storage system embodying the same prior art concept where the working fluid undergoes a dual phase change so as to store energy.

In this preferred system, each tank has an upper outlet for evaporating fluid and a lower inlet for condensing vapour. System 200 is shown with its valves 36a-d in a charging configuration and comprises a lower pressure LP side with a lower pressure or colder tank or vessel hereinafter the "cold tank", and a higher pressure HP side with a hotter tank or vessel hereinafter the "hot tank" and an intermediate compressor/expander 30.

The valves are preferably one way valves configured so as to allow evaporating liquid (shown at 6) in each tank to leave the liquid as vapour from its upper surface, and to allow condensing vapour in each tank to re-enter the liquid from below its surface and condense (shown at 8).

The working fluid liquid in both cold and hot tanks may initially be at the same temperature and pressure. When charging the compressor/expander machine acts as a compressor. It draws working fluid vapour from the cold tank thus reducing the pressure in that tank. As all spaces within the system contain either working fluid vapour or liquid at an initially similar temperature, the reduction in pressure will immediately cause the (saturated) liquid to boil. This is a preferred method of initiating a liquid to vapour phase change, although other less efficient means may be used such as the supplying of external heat to the liquid to initiate boiling. The boiling (shown by bubbling) takes heat of vaporisation from the cold tank liquid 11 resulting in a drop in temperature thereof and a corresponding reduction in its vapour pressure. The vapour produced passes along duct 37 through open valve 36 and is compressed adiabatically within the compressor 30 raising the temperature and pressure thereof, and, as it left the cold tank at the saturation condition, thereby superheating the vapour. It is then forced by the valving, open valve 36d, to re-enter the bottom of the hot tank where it passes through the liquid. Being somewhat warmer than that liquid it loses heat and condenses while simultaneously raising the temperature of the liquid in the hot tank 20.

Charging may continue until the temperature of the working fluid 11 in the cold tank drops to a point where the vapour pressure is so low that the compression ratio to pass fluid to the hot tank becomes unviable, or a maximum operating temperature is reached in the hot tank.

Discharge is the reverse of the charge process and may be achieved by allowing a reversible expander/compressor (e.g. a reversible positive displacement machine such as a reciprocating piston-based machine) to operate as an expander, or by switching the compressor for an expander provided in a parallel connection (i.e. disconnecting the former and reconnecting the latter).

Figures 2a and 2b Figures 2a and 2b show the charging and discharging gas flow directions. In this instant the variation between the two conditions has been achieved by means of non-return valves; valves 36a and d are open on charging and valves 36 b and c are open on discharging. Selectable valves, rather than automatic valves as shown, are another alternative.

During discharging, working liquid 21 in the hot tank 20 boils in response to a drop in pressure caused by the expander 30 drawing vapour from that tank and along duct 38. Work is released during expansion and the vapour after expansion is passed via duct 32 to the bottom of the cold tank 10 where, in a process similar to that found in the hot tank during charge, the vapour is cooled and condenses while warming the cold tank Figures 3a and 3b Figs 3a and 3b are T-s diagrams for the system of Fig. 1 respectively showing showing charge and discharge on temperature-entropy coordinates. As the vapour leaves the hot tank in a saturated condition, the expansion will involve wet vapour inside the liquid-vapour curve (shown as the downward arrow in Fig. 3b). It is also evident that the superheated vapour leaving the compressor on the hot side during charge (downward arrow from C in Fig. 3a) has considerably more enthalpy than the wet vapour entering the expander on discharge and that this must result in a very poor round trip efficiency. The heavy lines show the charge and discharge paths as each process proceeds.

Taking Fig. 3a, the uncharged system is, in this instance, at 1 bar pressure (as shown by the horizontal plot between A° and E°. The thin arrows indicate the increasing area of the T-S diagram as charge proceeds (see intermediate charge position from A° and F°) and it can be observed that an appreciable enclosed area is defined by the saturated vapour line A to E and the part of the charge path in the superheated region.

On discharge, superheat is not possible as the hot side boils off at the saturation condition. The discharge paths therefore show a significantly reduced 1-s area and hence significantly less energy is released on discharge than on charge. This represents a rather poor overall round trip efficiency. A further result of this poor efficiency is that the final temperature of the two liquids in the hot and cold tanks when discharged is likely to increase after each cycle unless steps are taken to remove excess heat.

It is possible for the hot tank to start with no working fluid present if, on charge, the thermal mass of the hot tank is sufficient to initiate condensation. As further vapour is added to the hot tank it will then condense within the liquid from this first condensation and the charge can proceed as before.

Figure 4 Figure 4 illustrates an energy storage system in accordance with a first embodiment of the present invention.

This system 202 may comprise all the components of the basic system as described in relation to Figures 1 to 3, but further comprises a regenerator 50 on the HP side of the system comprising a solid gas permeable thermal storage medium through which the gas passes for direct transfer of thermal energy to and from the solid medium.

The function of this regenerator is to capture superheat on charging and return it on discharging and this is done in accordance with the present invention in a manner that reduces irreversibilities in the respective heat transfer processes.

The regenerator 50 is inserted between the compressor/expander 30 and the hot tank 20 so that it is located in the path of both the condensing vapour on charging and the returning, evaporating vapour on discharging, with the arrangement configured to allow any condensing vapour to reach the hot tank 50. In the Figure 4 system, where each tank has an upper outlet for evaporating fluid and a lower inlet for condensing vapour, the regenerator is therefore located before the junction leading to those, and is preferably vertically disposed above the tank. Alternatively, the regenerator may be disposed vertically above the liquid surface inside the hot tank, as shown in Figure 16 below.

In the Figure 4 system, the sensible heat of the superheated region shown on the charge diagram of Figure 3a is now stored within the regenerator (hot side heat store) for later recovery, so that Figure 3a represents both the charge and discharge pathway. The charge and discharge paths are therefore similar (notwithstanding small variations due to real process irreversibilities) and so the work of charge and discharge (the area under the T-s diagram) is similar.

Ideally, even by the fully charged condition, the regenerator 50 should be of a sufficient size that it can capture all the sensible heat (superheat) such that the vapour attempting to condense in the hot tank is a saturated vapour, as opposed to a dry vapour containing superheat. This requirement may be alternatively specified by stating that some condensation should always take place in the regenerator during charging and preferably even up to the fully charged condition.

GB423093 (to Marguerre) also teaches the use of a regenerator or "superheat accumulator" in this location to store and return the superheat of the steam in order to improve the amount of energy stored and returned during the round trip. However, Marguerre suggests this is done using indirect heat transfer using coiled pipe heat exchangers with the superheat eventually stored in a liquid in the superheat accumulator.

Since the liquid is neither static (e.g. due to convection effects) nor in direct contact with the vapour, thermal energy is stored crudely with little control, responsivity or efficiency leading to a high degree of thermodynamic irreversibility. Thus, while Marguerre recognises the need to store and return superheat, there is no recognition that the marching inlet temperature and marching phase change temperature associated with this particular liquid/vapour/liquid energy cycle can be harnessed to achieve very efficient thermal storage.

Before discussing the operation of the Figure 4 system and the role of the regenerator (i.e. heat store), it is helpful to consider how comparative heat stores operate.

Figures 6a to 6c (Comparative) In Figures 6a to 6c the graphs represent the temperature profile along the length of a conventional heat store comprising a porous solid mass (e.g. matrix) at progressively advancing states of charge. The flow is in the direction of the arrow.

Figure 6a -No marching, no phase change Referring to Figure 6a, gas enters the heat store from the left at a constant temperature above that of the porous mass through which it flows. Sensible heat is transferred to the mass, thereby cooling the gas, but it does not condense in the store.

The profiles represent the temperature profile in the porous mass at different states of advancing charge with this profile taking the form of an advancing front, or thermocline, moving through the store in the direction of the flow. The porous mass downstream (ahead) of the thermocline remains substantially at the original store temperature, whereas the porous mass upstream (behind) the thermocline is substantially at the gas inlet temperature. Both ahead of, and behind the thermocline, the store is thermodynamically inactive, ie, no heat transfer is taking place although gas is still passing through this material and piessure losses, and hence entropy generation will occur, which will reduce the overall reversibility of the process.

The evident lengthening of the thermocline region as the charge progresses also results in a reduced proportion of the porous mass achieving the inlet gas temperature, if irreversible losses are to be avoided as a result of running all, or part of the thermocline out of the store.

Figure Gb -No marching, with phase change This figure shows the effect of passing a dry vapour (gas) into the porous mass heat store at constant pressure and temperature, where the gas cools and condenses in the store.

The graph shows distance along the store (x axis) versus temperature within the store (y axis) for both the store material and for the vapour. Three consecutive states of thermal charge are shown with the vapour flowing through the storage mass in the direction indicated by the arrow. Each stage is shown by a solid and dashed line representing the temperature profile along the store of the vapour and the porous mass respectively. The lowest dotted line depicts the original (uncharged) temperature of the heat store at the start of charging.

The leftmost pair of lines represent the temperature profiles shortly after the start of the thermal charge process. The porous mass temperature (dotted line) is at the same temperature as the entering vapour at the left hand edge of the graph. This temperature drops in the direction of the flow before abruptly changing the sign of the gradient and proceeding to warm for the remainder of the distance through the store. The vapour temperature (solid line) initially follows a similar profile, but, due to the need for a temperature difference between solid and vapour for heat exchange to occur, is above the storage mass temperature by some temperature increment, this increment increasing with distance upstream along the store until the point at which the storage mass line exhibits the change in sign of slope. At this point, the vapour has reached the condensation point (dew point), ie, it has started to condense to droplets, and since the pressure is constant, this is a constant temperature process, ie, the line representing vapour temperature continues to the right hand side at a constant temperature. At the condensation point, the temperature difference between the vapour and the storage mass is significant. Since this difference is a result of irreversible heat transfer, it results in significant entropy creation and is highly undesirable.

The pairs of lines representing the two later stages of charge show similar characteristics with the temperature differences at any given vapour temperature getting progressively smaller with later states of charge. This has the unfortunate effect of also resulting in a significant slackening of the temperature gradient associated with both sensible and latent heat transfer. As the temperature profile may be regarded as a travelling thermal front (a thermocline), the pad of the store that is fully charged, ie, substantially at a temperature equal to the inlet temperature of the vapour, is a fraction of the entire length of the store. A slackening gradient therefore results in a lower utilisation of the store as a result of the consequent lengthening of the thermal front.

Once any region of the store upstream of the thermocline achieves the inlet vapour temperature, no further heat transfer takes place in that region. However, gas must still pass through it, which means that pressure losses will accrue over this region without any useful charging activity taking place.

Figure 6c -Marching, no phase change This figure shows the effect of a gas passing through the solid mass store that has a marching (i.e. progressively increasing) inlet gas temperature. In this case, sensible heat is transferred to the mass, thereby cooling the gas, but it does not condense in the store.

Some processes can result in a continually rising (or falling) temperature of a gas or vapour. If there is a need to store thermal energy under these conditions, a porous mass store has some specific advantages by comparison with a conventional heat exchange process.

Figure 6c shows the effect of dry vapour (gas) arriving at the inlet to a store with a temperature rising (marching) at a constant rate. The pressure is constant as is the mass flow rate. The thermocline now comprises, in the early stages of the charge, a shod region of heated store with a negative temperature gradient in the direction of flow, followed by a region in which this gradient blends out to the original, un-charged store temperature.

As the charge progresses, this temperature profile advances along the store with an unvarying form, however, as the inlet temperature is now increasing, an approximately linear, negative gradient part of the thermocline enlarges in scale.

Eventually, the blend region of the thermocline passes out through the exit of the store (right hand side of figure) and the gas exit temperature starts to rise. Once the blend region has fully exited the store, the thermocline tends towards a linear, negative gradient temperature profile and the temperature everywhere in the store rises at a constant rate equal to the temperature rise rate at the store inlet.

Every part of the store is now thermally active and undergoing heat exchange because each new "packet" of gas encountered has a higher temperature than the last one.

The conditions depicted for comparative stores in the graphs of Figures 6a-6c are summarised in Table 1 below.

Figures Gas Inlet Temp. Sensible Heat Transfer Latent Heat Transfer Oa Constant Yes No 6b Constant Yes Yes, phase change 6c Marching Yes No 7a-c Marching Yes Yes, phase change

Table 1

By contrast, returning to the through-flow solid medium regenerator 50 of the system of Figure 4 according to the present invention, all three conditions in Table 1 take place, as described below with reference to Figures 7a-c.

Figures 7a to 7c Referring to Figure 7a, this is a graph showing the gas and HP side regenerator temperature profile over the length of the regenerator at an instant early during charging, where the superheated gas is entering the cooler regenerator and cooling and starting to condense, and where the gas has a progressively increasing inlet temperature. Ts represents the heat store material temperature and Tg represents the vapour temperature. Thus, the initial negative gradient upstream corresponds to a sensible heat transfer region, and blends into a flat downstream region corresponding to a latent heat transfer region where the gas has lost all its superheat and is starting to condense at its (constant) liquid/vapour equilibrium phase change temperature.

Figure 7b shows the progression of the temperature profile of the vapour and HP side regenerator with time during the storage mode (charge). Both vapour and solid temperatures are shown and the marching of the [-V phase change temperature is evident. Also evident is that the vapour temperature (dashed line) is leading the regenerator (and, implicitly, the liquid temperature) in this condensation mode of operation. If vaporizing, the solid and dashed lines would change over and the vapour will again be leading the regenerator (and the liquid) in the cooling sense.

The marching condition is achieved if the low temperature end of the regenerator/heat store is directly connected to a closed receiving vessel into which the vapour is condensed. This is typically achieved by passing the vapour though a body of the liquid condensate such that further condensation occurs at the bubble boundaries. As this is a closed vessel, as energy in the form of latent heat is added to this body of liquid, the temperature of the liquid rises. This also raises the equilibrium pressure between the vapour and the liquid, ie, the pressure in the vessel increases as further condensation takes place.

The vapour transiting the heat store, to continue to flow through the store, must therefore be pumped at progressively higher pressure as the pressure in the liquid vessel is increased. Since the vapour entering the store is both dry and superheated, this means that the compressor that is feeding the store, if it takes input vapour at a constant pressure, must operate over progressively increasing pressure ratio. This pressure ratio is entirely dependent upon the vapour-liquid equilibrium pressure in the liquid vessel.

This is also the case if the vapour entering the compressor is supplied at a steadily decreasing pressure and temperature.

The use of a porous solid mass heat store in this situation has several very significant and non-obvious advantages. Firstly, the heat exchange temperature difference is smaller than would be the case for a conventional heat exchange process.

Secondly, condensation that occurs in the store takes place at a constant, and small temperature difference (delta T). This should be compared with the situation within a non-temperature marching store with a phase change as described earlier in which the condensation temperature difference is both non-uniform and significant.

The above-referenced small delta T effect is explained in terms of a solid thermal medium as follows:-Any thermodynamic cycle that involves a phase change as a working fluid is heated or cooled is vulnerable to high irreversibility. This occurs as a result of the phase change taking place at constant temperature while the body from, or to which the heat is flowing does not undergo a phase change and so has a varying temperature. Under these conditions, the temperature difference between the phase change material and the non-phase change material cannot be constant over the duration of the phase change and so must therefore, at some point, be greater than the minimum possible. As heat transfer requires a temperature difference, the temperature difference is non-reversible and results in an increase in generated entropy.

In the present invention, the liquid into which a saturated vapour is condensing exhibits a "marching" (i.e. progressively increasing) temperature i.e. this problem is overcome by constantly changing the temperature and pressure at which the phase change occurs. This causes the heat transfer involved in the phase change process to take place over a constant temperature increment, since this implies that this will equal the average temperature difference, the scale of the maximum temperature difference is reduced and the reversibility of the process is significantly improved. As a temperature and pressure dependence of phase change temperature is essential for this condition to be achieved, this can only be applied to the vapour-liquid or liquid-vapour transitions.

The effect of rate of end state temperature change on heat exchange during a phase change is examined in the following discussion.

Heat exchange between a gas and a matrix comprising a solid energy storage material is via direct contact between the gas and the storage mass. Heat exchange can only occur if a temperature difference exists between gas and solid and the heat flow is directly proportional to this temperature difference. A thermal store typically consists of a porous matrix of solid material through which a mass flow of gas passes. If, during a phase change within the gas from vapour to liquid, the rate of change of temperature at one point is forced to a particular value, the temperature difference between vapour and solid at that point becomes that existing over the entire phase change region of the store.

To show why this is the case: Notation:-S Entropy Tg Gas temperature Solid temperature Porosity li Film heat transfer coefficient I,, Particle scale dimension in Gas mass flow rate A Store cross sectional area p8 Store material solid density Store material heat capacity x Dimension along store in flow direction The rate of entropy change from gas to solid at a given point in the store is given by; dS (T -T5)(1-8)hA ci IT (1) And the rate of change of solid temperature is given by; dTs(Tg_Ts)(1_E)h 2 cit -pscsIp Setting AT = (Tg -T5) and re-arranging equations 1 and 2 to make AT the subject gives; dS thT ciT. AT(1 -E)h A9±PSCS [p (3) Equation (3) is equivalent to: dS ciT5 a-T =b-=cLIT cit where a, b and care constants.

It can thus be observed that if heat transfer is taking place, AT must be non-zero.

If a phase change is occurring then Pg will be constant and so a given value of AT requires a proportional value of. The effect of this is that if, for example, the phase change is vapour to liquid, the solid temperature must be less than the gas temperature but the solid temperature must be rising. If the store exit temperature is made to rise continuously then the result is that the vapour temperature, and hence the temperature of the phase change, will rise at a similar rate. This also requires the pressure to rise as the condensation temperature cannot rise without this occurring. The result is that the phase change heat exchange takes place within the store with a constant, and therefore minimum AT and with a constant reduction of vapour entropy with distance () but with a flat temperature profile along the store over the phase change region (97-a). This is described from here on as a "marching phase change temperature" and the resulting condition of smallest possible temperature difference results in the lowest possible generation of entropy and hence the highest possible reversibility. This is a very significant improvement to the effectiveness of any energy storage cycle in which it is incorporated.

Simulation of a heat store charge, using a real gas equation of state, confirms the constant AT between gas and heat store material over the phase change region. Figure 7b is a typical plot of the gas and solid material temperature profiles over the length of the store taken from this simulation for a charging process for a hot store (ie, hot gas entering a cooler store). The dotted lines represent the vapour temperature profile while the solid lines represent the temperature of the porous storage mass.

It is undesirable to allow full condensation of the vapour within the heat store as this will then prevent the condensing vessel from rising in temperature in exact parallel with the condensation temperature in the porous mass store. This will result in the exit flow from the porous mass store entering the liquid store at a different temperature to that of the liquid already in the store and hence irreversible thermal mixing will then take place.

A further advantage is that all parts of the store may be thermally active at all times, ie, every part of the store may undergo simultaneous heat exchange so as to achieve 100% store utilisation (as compared to prior art stores where gas still has to pass through inactive regions with associated pressure losses.) A final advantage is due to the variation in heat capacity of most materials associated with temperature variation. In a conventional counter flow heat exchange situation, As temperature changes, if one flow is gaseous and the opposing flow is liquid, the heat capacities of the two flows usually tend to vary in opposition to each other, as the mass flow in either of these flow must be instantaneously constant at all locations within the heat exchanger, the resulting heat capacity mismatch results in a very non-uniform, and hence non-minimum heat exchange temperature difference. A porous mass store, being in intimate contact with the vapour, always self-minimises this temperature difference and preserves the maximum reveisibility.

In conclusion, the specific combination of a porous mass heat store with a marching temperature is able to produce a minimum heat exchange temperature difference throughout the store, may result in 100% of the store actively exchanging heat at all times, may provide a constant and minimum phase change temperature difference between the condensing vapour and the porous heat storage mass (i.e. in the downstream phase change region of the store) and may result in 100% utilisation of the store. If combined with a closed condensing vessel on the high side the marching becomes automatic as the feeding compressor responds to the steadily rising temperature within the condensing vessel. The heat store thus responds exactly and immediately to pressure and temperature changes within the liquid store.

Referring to Figure 7b, it should be noted that the profiles showing the constant but steadily increasing vapour temperature (i.e. [-V phase change temperature) also reflects the liquid temperature in the hot tank and that the liquid temperature "marches" as the vapour pressure progressively increases. This has the very significant benefit that it allows a small amount of liquid in the hot tank (because it has a steadily marching/increasing temperature) to absorb the energy of the phase change (i.e. absorb the heat of condensation) in an efficient (i.e. highly irreversible) manner that minimises entropy increase due to the fact that a small delta T is maintained even as the phase change progresses.

If, as in the case of Figure 6b, there is at any point no condensation in the hot store, for example, if charging continues for so long that the phase change region exits the hot store, then instead of saturated vapour exchanging its latent heat with the liquid in the tank with a small delta T, dry vapour containing superheat will be exchanging heat with the liquid over a larger delta T. Figure 7c shows how only the HP regenerator temperature profile changes with time during charging. The temperature at the right hand side of the figure can be seen to increase with increasing state of charge whereas a thermal front can be seen to progress from left to right. The phase change region is the horizontal portion of each trace although it should be noted that the degree of phase change within the heat store is low with the vapour, at departure from the store, typically having a quality (dryness fraction) of greater than 0.8 i.e., typically at least 80% of the phase change takes place downstream of the regenerator (within the vapour receiving tank), usually significantly more, but always less than 100%.

Figure 5 Figure 5 illustrates an energy storage system 204 in accordance with a second embodiment of the present invention, comprising a further regenerator 40 on the lower pressure [P side of the system to further improve round trip efficiency. The addition of this LP side regenerator ensures that all vapour compressed or expanded is dry, i.e., contains no liquid droplets.

Storing some superheat from the vapour on the [P (cold) side is very desirable.

This is achieved by the addition of a sensible heat store (regenerator) between the cold liquid vessel and the compressor/expander. Consider the effect of reducing cold side pressure by drawing some vapour from the cold side with the compressor. The liquid in the cold vessel starts in equilibrium with the vapour that is also present within the vessel.

As vapour is drawn off by the compressor, the liquid in the vessel boils and the latent heat required for this process, coming from the body of the liquid, results in a fall in the liquid temperature. The sensible heat store is still substantially at the original equilibrium temperature and so the passage of the vapour through this heat store warms the vapour.

Since this vapour is initially at the saturation temperature, any warming results in some degree of superheat, the vapour entering the compressor is thus entirely dry in nature and so the compression process can be made highly reversible as there are no droplets of liquid within the charge of vapour to be compressed. The presence of droplets within a vapour under compression or expansion result, due to the finite size of droplets, in the creation of a temperature mismatch between the droplet and the surrounding vapour and so irreversible heat exchange will naturally result.

During a discharge cycle, as the hot side of this system is highly reversible and the condition of the vapour is similar to that found in a charge cycle in all pads of the system at a given state of charge or discharge (ie, if the system is charging or discharging at an instantaneously similar percentage of full charge), the vapour leaving the expander will still be superheated to a similar degree as it was during the corresponding charge process. The vapour within the compressor/expander will thus always be dry throughout the whole compression/expansion process. This makes this process very highly reversible and the presence of even a small superheat store unit on the cold side of the system is very desirable indeed and will improve round trip efficiency.

The LP regenerator does not need to be large as any amount of heat storage will result in some degree of superheat of the vapour (which is always cooler than the LP regenerator).

Figure 8 The charge cycle for the Figure 5 system is shown in Figure 8. The discharge cycle is similar and, with idealised "perfect" machinery, would overlay the charge cycle exactly. Note that the arrows show the direction in which the T-S loop expands during charging as before. The direction of these arrows is reversed for discharge.

After compression the vapour is warmer and so, since the hot store temperature represents the status of the system at a lower state of charge than that represented by the state of the fresh vapour, it is now warmer than the hot store 50. Passage through the hot store 50 therefore cools the vapour, ideally to the point where saturation is just achieved within the store. Passage to the hot tank is via a pipe 34 to the bottom of the tank where the vapour is nd (8) through the liquid already present and the phase change back to liquid continues within the tank. This warms the hot tank liquid due to absorption of the heat of vaporisation of the vapour, as the vapour entering the tank is always slightly warmer than the liquid within the tank the temperature of this tank continually rises during charge. This rising of temperature also results in an increase of vapour pressure and hence the pressure ratio of the system increases steadily. The rising hot tank temperature also provides the marching [-V phase change temperature at the exit of the hot store that is necessary for a phase change at minimum T, and hence maximum reversibility, as described above.

Figures 9-11 The following Figures 9-11 annotate the key temperature points in the cycle.

Referring to the schematic two heat store (regenerator) system shown in Figure 9: 1 Cold tank temperature 2 Temperature at compressor/expander cold side 3 Temperature at compressor/expander hot side 4 Hot tank temperature Regions 1-4 are annotated on the T-s diagram of Figure 10 as Temperature-Entropy (T-S) coordinates. The time history of each point is illustrated on a state-of-charge versus temperature plot in Figure 11.

From Figure 11 it is clear that for a packet of vapour, the temperature rises from point ito 2 on charging due to the additional superheating provided by the LP side regenerator, there is a large increase in temperature at point 3 after compression, but the temperature at point 4 has dropped due to some cooling through the HP side regenerator, prior to condensation, although point 4 on the "hot side" is still higher in temperature than the LP side "cold side" points.

Regardless of which side of the apparatus, or charge mode, both regenerators always act to donate (regeneratively stored) heat to an evaporating vapour, and to remove and store heat from a vapour that is about to condense. However, because the cycle is not symmetric, it is the sensible heat removal and storage from the compressed gas on the HP side during charge (from 3 to 4 of Fig. 9) that is crucial, as the return of that thermal energy (4 to 3 on discharge) provides the energy for a large expander power stroke.

Notably from Figure 11, it is the exit temperature of the vapour leaving the compressor on charge that marches the most significantly.

Figures 12 to 14 Figures 12 to 14 illustrate third, fourth and fifth embodiments according to the present invention comprising systems in that limit the extent to which the pressure and temperature drop in the working fluid in the LF vessel. All three embodiments are based, merely by way of example only, on the system of Figures 4 and 5.

If the cold tank temperature is allowed to fall excessively, ie, close to the triple point temperature, the vapour pressure will drop such that the compression ratio required to achieve delivery of vapour to the hot side of the system will become unfeasibly large.

This high compression ratio will also be associated with an excessive temperature ratio and, as the hot side pressure has increased during charge, the peak temperature may become higher than may be tolerated by the machinery within the system. Since the excessive compression ratio is a direct result of the falling cold side temperature, a simple way to avoid, or ameliorate this problem is to prevent the cold tank from cooling excessively. This may be achieved by (i) adding thermal ballast to the cold tank such as making the body of liquid on the cold side greater, or by adding a solid inert thermal ballast to the cold tank, or by (ii) adding heat from the surroundings to the cold tank, or by (iii) feeding the cold tank from a supply such that only a small percentage of the liquid passing through the tank is vaporised at any given time.

The first option of increasing the amount of working fluid on the cold side offers the highest reversibility as the marching temperature effects are achieved with minimum temperature variations at all times and with no heat exchange between the tank and the environment beyond the system.

The other two options both require external heat exchange although in the case of the continuous liquid feed to the cold tank, the time allowable for heat exchange is indeterminate and may take place over a considerable time with correspondingly very small AT.

Figure 12 illustrates a system that includes a continuous liquid supply on the LP (colder) side.

The [P tank 10 is provided with an inlet 213a for supplying working liquid to the tank (e.g. from an infinite supply) and beneath it an outlet 213b for withdrawing working liquid. With a continuous cold tank feed (the system, the liquid feed and exit pressures must be similar to the cold tank pressure. If this is a constant temperature feed, no marching temperature condition will exist and hence, a cold sensible heat store /regenerator would not experience any temperature variation during charge or discharge.

It is therefore of little use in this configuration (as it will have risen to the temperature of the approaching vapour shortly after start-up).

Referring to Figure 13, if, however, the working liquid supply on the LP side is allowed to reduce in temperature during charge then a cold heat store is still effective.

One way of achieving this is to feed from a larger supply tank (i.e. as a finite supply).

Figure 13 therefore illustrates a system 214 including an auxiliary supply tank 57 on the LP side that supplies working fluid via an upper inlet 215a to the [P vessel 10 and withdraws it from a lower outlet 215b. The temperature variation is achieved by mixing of the return flow from the LP vessel 10 (cold tank) with that of the larger supply tank 57, and hence, a [P side regenerator 40 is again of use. The temperatures of both supply tank 57 and cold tank 10 will fall during charge, but at a diminished rate as compared to a sealed LP vessel; as the vapour pressure is also falling a pump 53 and/or a restriction should be placed (in outlet duct 215b and in inlet duct 215a, respectively) between the supply tank S7and cold tank 10 to facilitate transfer of liquid against the ensuing pressure difference.

For the case where water is the working fluid, the supply tank 57 may typically start the charge process at close to ambient temperature. As the supply tank will not freeze, the vapour pressure in the cold tank 10 may drop to the point at which the system pressure ratio becomes unviable for the machinery.

Turning to Figure 14, alternatively, if cold tank temperature regulation is sought by means of external heat exchange this may be achieved by passing heat exchange members directly through the tank. These may be in the form of metallic leaves which are distributed within the liquid and pass through the walls of the tank to exchange heat with the surroundings. Another method, as shown in Figure 14, is to provide a heat exchanger 315 which passes a secondary, heat exchange fluid through working liquid heating/cooling coil within the tank via inlet 313b and outlet 313a, to regulate temperature in the LP vessel 10.

If the heat exchanger maintains a constant working liquid temperature on the cold side, again no regenerator is required on the cold side. Alternatively, by graded control of the heat exchange fluid a marching temperature condition can also be achieved in which case, unlike the system shown in this figure, a cold side sensible heat store again becomes effective.

If used in a domestic environment (for example) the secondary fluid may be the domestic hot water supply and the temperature of the cold tank could then be at a similar temperature to the domestic hot water system. Waste heat build-up due to irreversibilities would then be passed directly to the domestic hot water supply and would provide a useful additional heat input to that system: A further benefit of the cold side heat exchange method described above is that the heat exchange fluid does not need to be at the same pressure as the cold tank. This allows the system pressure to be set to suit a working fluid with vapour pressure characteristics that require pressures either above or below ambient.

Figure 15 Figure 15 shows an energy storage system in accordance with a sixth embodiment of the present invention, which illustrates integration of a waste heat recapture sub-system.

Irieversibilities in an energy storage and recovery system will result in the generation of entropy and hence an overall increase in the mean temperature of such a system. In the illustrated system, water/steam is used as the working fluid and the storage system is integrated with a domestic hot water system. This takes advantage of the ambient pressure boiling characteristics of water to dispose of any waste heat build-up that arises due to such irreversibilities within the system.

The cold liquid store 10 is connected to the domestic hot water tank 321 via a non-return valve 323 that only allows flow to run towards the hot water tank. The domestic tank 321 is at close to atmospheric pressure and so, as the pressure of the cold liquid tank 10 approaches the pressure within the domestic water tank, if the liquid tank absorbs excessive latent heat due to entropy build up within the working fluid, it will start to boil. The boil-off vapour will then transit the non-return valve 323 and enter the domestic hot water tank 321 where condensation will occur, thus adding heat to the domestic hot water supply. In this way the heat build-up due to irreversibilities within the energy storage system is put to useful effect by heating the hot water system and the heat build-up is controlled and removed from the storage system.

The venting process will result in a nett loss of working fluid from the storage system and this must be replenished. For this purpose a connection 325 to the mains water supply is fed to the cold liquid storage tank 10 via a valve 327 so as to replenish missing water. As a certain level of liquid must be present in both hot and cold liquid tanks at any state of charge defined by the temperature in both tanks, a mismatch between the liquid level that should be present for correct charge-discharge cycling will be evident if level and temperature sensing is provided. If the actual level is deficient then the feed valve may be opened to correct this and so preserve the correct working fluid level in the system. The system may also be topped up after full discharge.

It is desirable for the water used in the storage system to be as clean as possible to avoid scale build up. As the process from cold to hot tank during charge is a boil-off-condense process, the cold sensible (superheat) store, the hot sensible heat store and the hot tank in addition to the compressor/expander only encounter vapour or distilled water, ie, the entire process can be seen as a distillation process and any solid matter will therefore only collect in the cold tank. As this tank is either at ambient pressure or below, the provision of access for de-scaling is therefore a trivial problem and the use of domestic water as feedstock for the process, rather than purified water, is acceptable.

A further benefit of using a domestic water feed is that clean, sterilised water may be extracted from the hot liquid tank via a suitable outlet. This may be of use in areas where the domestic water supply is not potable although extraction of water from this point in the system will result in a reduction in energy storage round trip efficiency.

A waste heat recapture sub-system may alternatively involve merely heat exchange from the vented working fluid, where the latter is directed to a heat exchange sub-system to provide indirect heating, as opposed to direct mixing of the working fluid with a heat receiving medium as described above. Extraction of condensate with extraction apparatus after this heat exchange process may also provide cooled purified water without impact on round-trip efficiency.

Figure 16 Figure 16 shows an alternative system in accordance with the present invention with a single inlet/outlet for vapour entering and leaving each tank and where a regenerator is located inside each tank above the respective liquid surfaces.

Referring to Figure 16, the energy storage system 210 is shown in a charging mode and comprises insulated hot vessel 120 containing a regenerator 150, hot working liquid 121, plenums 122, 123, hot inlet valve 124, hot outlet valve 125, transfer pipe 126, porous screen 127, hot liquid surface 128, as well as a cold vessel 110 containing a regenerator 140, cold working liquid 111, plenums 112, 113, cold outlet valve 114, cold inlet valve 115, transfer pipe 116, porous screen 117, cold liquid surface 118, compressor/expander means 130 and pipes 101 and 102.

By locating each regenerator inside each tank, it can benefit from the insulation and pressurisation of the tank, especially on the HP side. Both regenerators extend across the entire cross section of the respective hot and cold vessels located above the liquid. Each regenerator may comprise, for example, a particulate bed or solid porous matrix. The HP side regenerator 150 is shown much larger reflecting the fact that it needs to be able to store all the sensible heat of the compressed vapour emerging from the compressor 130 during the whole charge cycle, up to the maximum operating temperature, preferably so as to avoid any vapour leaving the HP side regenerator in a superheated condition. To achieve this, some phase change should start to take place in the regenerator 150.

General Conditions: The maximum vapour pressure in hot vessel 120 should be below the critical pressure for the working fluid and the corresponding maximum temperature should be within the operational range of the materials selected for the apparatus on the HP side. (Obviously, the liquid in the cold vessel must be kept above its triple point.) Starting conditions: Hot working liquid 121 would normally be at the same temperature or hotter than cold working liquid 111. Pressure in the hot vessel 120 would normally be the same or higher than the pressure in the cold vessel 110.

Hot liquid 121 and cold liquid 111 may be pre-heated by some external energy source, such as an electric element heater or gas powered heater. In this way the system can start at the correct conditions for the compression/expansion machine or machines; for example, it is often desirable to start with a pressure difference between the HP and LP sides so as to limit the variation in pressure ratio required. Alternatively, the system may start with the hot and cold working liquids at the same temperature (and pressure), which will result in the highest energy density upon full charging (largest time-integrated area under T-s diagram) but means that the compression/expansion machine or machines need to operate over a larger pressure ratio (with little work achieved in the early charging cycles). Where water is the working fluid, both sides may conveniently be pre-heated to about 100°C.

Usually, it will be desirable to start each storage mode with the hot liquid 121 and cold liquid 111 at pre-selected heated and cooled temperatures. In practice, these may be the respective temperatures normally achieved when the system is operating and is part-charged, so that, in effect, each recovery mode is terminated prematurely at that chosen part-charged condition such that the next storage mode commences at that beneficial condition.

In charging mode, compressor/expander 130, acting as a compressor, draws gas/vapour from plenum 113 via pipe 101. Plenums 112 and 113 are connected via cold outlet valve 114, which is designed to only allow gas/vapour to flow through in one direction, for example it could be a non-return flap valve. Cold inlet valve 115 is designed to let gas/vapour flow only in the opposite direction. During charging this cold inlet valve 115 would normally be closed. As gas/vapour is drawn from plenum 113 the pressure drops slightly and gas/vapour is drawn from plenum 112. This drops the pressure in plenum 112 also. This process will continue until the pressure in plenum 112 has fallen to a pressure where the cold working liquid 111 starts to boil. At this point the rate of fall in pressure will reduce as the boiling liquid supplies gas/vapour to plenum 112 to substantially maintain the pressure. As the cold working liquid 111 boils the temperature of the cold working liquid 111 will start to drop. The rate of this temperature drop is related to the mass of the cold working liquid 111 and the amount of liquid that is boiling off. Obviously a larger mass means that the temperature will fall at a lower rate for a given quantity of boil off'. If desired, a deliberate excess of working fluid may be kept in the vessel on the LP side. Alternatively! a solid thermal media or phase change media is provided in the liquid on the [P side in order to act as thermal ballast (supplementary heat capacity) providing some latent heat for evaporation such that the rate of reduction of the liquid temperature and hence pressure in the LP side is reduced.

On charging: The role of the compressor is to compress the dry vapour from low pressure to high pressure. The actual low pressure is determined by the temperature of the liquid in the low pressure side. The actual high pressure is determined by the temperature of the liquid in the high pressure side and how well the vapour condenses as it is bubbled through the liquid and should reflect the higher pressure in plenums 122 and 123. Compression (and expansion) should be as near to isentropic as possible. (For example if the process is carried out with an axial flow compressor it is likely that for a given pressure rise the temperature of the gas will rise above that which would be expected for a purely isentropic process.) The cold side regenerator should be slightly hotter than the low pressure vapour and therefore ensure that the vapour boiling off is given a further injection of thermal energy so as to superheat it (i.e. ensure it is dry'). The dry vapour is compressed from a temperature very close to that of the low pressure liquid up to the high pressure. At this point, it is hot dry superheated vapour and needs to be cooled down to near the saturation temperature of the hot liquid. This is done in the hot regenerator where a small amount of condensation also occurs.

The vapour is then bubbled through the hot liquid as follows. The hotter higher pressure superheated vapour passes via pipe 102 into plenum 123. Plenums 122 and 123 are separated via hot outlet valve 125, which is designed to only allow gas/vapour to flow through in one direction, for example it could be a non-return flap valve. During charging the hot outlet valve 125 is normally closed. This means that as the pressure rises in plenum 123 the level of the liquid in transfer pipe 126 falls until superheated vapour can exit the bottom of the transfer pipe and travel across the base of the porous screen 127. The purpose of the porous screen 127 is to encourage the superheated vapour to bubble up through the hot working liquid 121 and transfer heat to the hot For heat transfer to occur and the vapour to condense it must be hotter than the liquid it is passing through (i.e. temperature gradient must exist). If it is not hotter it will not condense and the pressure in the hot part of the system will rise. This rise in pressure means that the vapour leaving the regenerator (and also the compressor) gets hotter and should now be hotter than the liquid it is bubbling through (so it will condense).

The result is a simple feedback loop that ensures that most of the vapour that comes out of the regenerator condenses, but a little always adds to the vapour pressure so that the temperature in the vapour is rising ahead of the liquid. (Note the liquid has to be getting hotter if the vapour is condensing out as heat transfer is occurring from the vapour to the liquid.) In this way energy is stored by evaporating liquid contained within the cold vessel, compressing this gas/vapour to a higher temperature and pressure where it is condensed back within the hot vessel. The compressor/expander 130 is used to input the energy and may be driven by an electrical motor or a mechanical device or some other method. The compressor/expander 130 may be a single machine or two separate machines or some other combination, and may be connected to a motor/generator, usually one capable of being synchronised with a local grid.

Discharging: this is the reverse process. Vapour boils off from the hot side and is then heated further as it passes through the hot side regenerator. It is then expanded in the expander so that the temperature and pressure drops and the vapour does work" on the expander machinery. The vapour then passes through the cold side regenerator which should cool it very slightly (the pressure in cold side is rising at this point). The cooler vapour is bubbled through the liquid, which is at a yet lower temperature, warming the liquid back up and condensing the vapour. During a full cycle there is likely to be a build up of waste heat related to real process losses and it is referable that there is some mechanism for rejection of waste heat via an external heat exchanger.

In this system, the terms hot and cold are meant in a purely relative sense. The system is based around a liquid/vapour/gas process and consequently the choice of liquid will affect the actual temperatures of each side of the system. If nitrogen is chosen then both the hot and cold working liquids will be at cryogenic temperatures. If water is chosen then it is possible that both the hot and cold working liquids are well above ambient temperature.

In this system the terms high and low pressure are also meant in a purely relative sense. The system is based around a liquid/vapour/gas process and consequently the choice of liquid will affect the actual pressures of each side of the system. It is conceivable that both the high and low pressure parts of the system are below atmospheric pressure or that both the high and low pressure parts are well above atmospheric pressure.

Such an energy storage system could be used for the storage of electricity. In this scenario, the energy input to the system would be electrical energy that would come from an electricity grid to compressor/expander 130 and the output from the system would also be electrical energy back to the same electricity grid. An alternative scenario might have two different grids, for example, the storage system might be charged by a dc input from a set of photovoltaic panels and then discharge energy directly to an ac electricity grid.

An example of an energy storage system that does not use electricity might be on a ship, where the storage system input energy is as mechanical energy in the form of a direct input from a wind turbine on the ship and where the output from the system is linked to the ship's propulsion. In this example, there is no electrical input or output, purely mechanical connections.

As mentioned above, the compressor/expander 130 may be a single machine and may be a reversible, piston based compressor/expander. This may be coupled to a (preferably grid synchronised) motor/generator and local grid and may be capable of instantly switching from a compression mode to an expansion mode. Thus, while the system may be run from fully uncharged to fully charged, and vice versa, there may be applications where the system needs suddenly to switch partway through charging in energy storage mode to instead recovering energy and contributing to grid demand. It is known in the art to use a reversible compression/expansion positive displacement based apparatus capable of such fast switching between compression and expansion. For example, this may be achieved by changing the activation timing of valves, preferably merely the closure valve timing, and which can seamlessly switch between compression and expansion modes while maintaining grid synchronisation.

As the pressure in the system rises so does the work per unit mass of vapour processed. One option is to keep constant power by processing less gas with variable speed machinery. Most machinery processes a fixed volume of gas at inlet, hence if the inlet pressure drops the mass flow also drops. Consequently, while the work per unit mass of gas drops, the power of a reciprocating compressor/expander may remain quite constant for a large part of the charge/discharge cycle, if the speed is varied using control apparatus. The pressure ratio is normally varying as well. A variable speed reciprocating machine could therefore provide constant power over a wide pressure range.

Ideally, a liquid is provided initially in both vessels and is preheated and/or precooled using heaters or heat exchangers to suitable starting conditions. If the liquid was water then one preferred starting condition of both stores might be 100°C. However, if the system is used regularly then there may be excess waste heat around (generated by previous cycles) to maintain an above ambient temperature, like 100°C, in normal operation.

For example, the system may start part charged eg on the HP side with the working fluid vapour pressure set at 10 bar and the temperature of the liquid at the associated saturation temperature and on the LP side with the working fluid vapour pressure set at 1 bar again with the associated saturation temperature. As the system charges, the pressure on HP side will slowly rise from that pressure to, say, about 25 bar at full charge, where the HP side apparatus needs to withstand the appropriate maximum superheat temperatures after the compressor. The system pressure should not go over the critical pressure, but can go over the critical temperature in the superheat region.

(The vapour will not condense to a liquid above the critical pressure.) As the skilled person will appreciate, modification may be made to the embodiments described above while remaining within the scope of the invention as claimed. In particular, while the use of water as a working fluid is highly preferred other working fluids may be used requiring operation within alternative pressures and temperature ranges.

Claims

CLAIMS1. An energy storage and recovery system comprising: a first vessel configured to store a working fluid as a saturated liquid/vapour mixture L1 having a temperature TL1, a second vessel configured to store the working fluid as a saturated liquid/vapour mixture L2 having a temperature TL2, power machinery disposed between the first and second vessels, and a regenerator disposed between the power machinery and the liquid stored in the second vessel, wherein the system is configured such that: (i) in a storage mode, working fluid vapour passes from the first vessel to the power machinery where it is compressed before passing through the regenerator and condensing in the liquid [2 in the second vessel, so as to produce a progressive increase in 1L2 of L2and in its liquid/vapour equilibrium phase change temperature during the storage mode; and, (ü) in a recovery mode, working fluid vapour passes from the second vessel, through the regenerator to the power machinery where it is expanded to produce power before condensing in the liquid Li in the first vessel, so as to produce a progressive decrease in TL2 of L2and in its liquid/vapour equilibrium phase change temperature during the recovery mode; characterised in that: the regenerator comprises a solid thermal storage medium through which the working fluid vapour passes for direct heat transfer between the vapour and solid medium so as to store and return superheat during the storage and recovery modes, respectively, and, wherein the system is configured such that some condensation takes place in the regenerator during the storage mode.
2. A system according to claim 1, wherein the regenerator comprises a porous matrix of solid thermal storage medium.
3. A system according to claim 1 or claim 2, wherein the system is configured such that condensation takes place in the regenerator for the full running time of the storage mode.
4. A system according to any preceding claim, wherein the system is configured to cease operating in the storage mode when condensation is only occurring in the last 5% or less of the downstream length of the regenerator.
5. A system according to any preceding claim, wherein the system is configured to cease operating in the storage mode when condensation is about to finish in the regenerator such that some superheated gas is about to start exiting the regenerator.
6. A system according to any preceding claim, wherein the system is configured such that there is a progressive decrease in TLI of L1 and in its liquid/vapour equilibrium phase change temperature during the storage mode, and, a progressive increase in TLI of L1 and in its liquid/vapour equilibrium phase change temperature during the recovery mode.
7. A system according to claim 6, wherein a further regenerator is disposed between the power machinery and the liquid stored in the first vessel.
8. A system according to claim 6 or claim 7, wherein the system comprises additional thermal ballast in the first vessel or a temperature regulating sub-system associated with the first vessel, configured to reduce the respective rates of progressive decrease and increase in TLI of L1 during the storage and recovery modes, respectively.
9. A system according to any preceding claim, wherein the regenerator is located inside the second vessel above the liquid and/or the further regenerator, if present, is located inside the first vessel above the liquid.
10. A system according to any preceding claim, wherein the system is configured to lose waste heat from the first vessel by allowing working fluid vapour L1 to vent to atmosphere or to a waste heat recapture sub-system, when its vapour pressure exceeds atmospheric pressure or the pressure in the sub-system, respectively.
11. A system according to any preceding claim, wherein the working fluid liquids L1 and L2 are each initially preheated or precooled to respective selected temperatures before commencement of the storage mode, optionally by stopping a preceding recovery mode at a selected point.
12. A system according to any preceding claim, wherein the working fluid comprises a water/steam mixture.
13. A method of operating an energy storage and recovery system according to any preceding claim, wherein during the storage mode superheat and latent heat are stored along the solid thermal storage medium of the regenerator in a respective upstream superheat transfer region and downstream latent heat transfer region, and wherein the temperature profile of the solid thermal storage medium progressively increases in temperature in both regions during the storage mode.
14. A method according to claim 13, wherein during the storage mode the temperature difference AT between the vapour and the solid thermal storage medium that it contacts is generally less than 15 at at least one selected downstream position in the superheat transfer region and/or the latent heat transfer region, and preferably at all downstream positions.
15. A method of operating an energy storage system using a working fluid that undergoes a phase change, the method comprising storing energy in an energy storage/ charging mode comprising the steps of: i) evaporating an amount of a saturated liquid L1 having a temperature TL1 to form a vapour, ii) optionally, heating the vapour to superheat it, iii) doing work by compressing the vapour to a higher temperature and pressure, iv) cooling and condensing an amount of the compressed vapour to a liquid L2 having a temperature TL2 so that there is a transfer of thermal energy from the vapour to the liquid, and further comprising recovering the energy in an energy recovery/discharging mode comprising the steps of: v) evaporating an amount of saturated liquid [2 at a temperature TL2 to form a vapour, vi) heating the vapour to superheat it, vU) expanding the vapour to a lower pressure and temperature to generate work, viU) optionally, cooling the vapour, ix) condensing an amount of the expanded vapour back to a liquid Li having a temperature TLI; wherein the energy storage system comprises a lower pressure [P side in which the working fluid is present as a liquid/vapour mixture Li at a lower vapour pressure, and a higher pressure HP side in which the working fluid is present as a liquid/vapour mixture L2 at a higher vapour pressure, separated by at least one compressor/expander operating so as to transfer vapour between the respective sides at the respective pressures, wherein during charging and discharging TL2 is greater than TLI, wherein heating step vi) uses stored thermal energy, wherein the stored thermal energy is stored by direct transfer to thermal media in a regenerator through which the gas passes for subsequent return, the regenerator comprising a throughflow regenerator having a solid thermal storage medium so as to allow direct heat transfer between the vapour and solid medium.wherein all sensible heat transfer upon cooling prior to condensing occurs within the regenerator, and preferably a minor amount of condensation occurs there too.
16. A method of storing and recovering energy wherein, in a storage mode, working fluid vapour from a first vessel containing a saturated working liquid/vapour mixture is compressed by power machinery and passed, via a regenerator, into a second vessel, where it condenses into a saturated working liquid/vapour mixture, the temperature and vapour pressure of which increases as more energy is stored therein, and wherein, in a recovery mode, stored energy is recovered by evaporation of vapour from the saturated working liquid/vapour mixture in the second vessel such that the temperature and vapour pressure of the mixture decreases, the vapour returning through the regenerator and expanding in power machinery so as to produce work before condensing back into the saturated working liquid/vapour mixture of the first vessel, wherein the regenerator comprises a gas permeable, solid thermal storage medium, and wherein during the storage mode superheat and latent heat are stored along the solid thermal storage medium in a respective upstream superheat transfer region and downstream latent heat transfer region, and wherein the temperature profile of the solid thermal storage medium increases in temperature in both regions during the storage mode.
17. A method substantially as described with reference to Figures 4, 5, 7a to 7c, and 8 to 16 of the accompanying drawings.
18. An apparatus or system substantially as described with reference to Figures 4, 5, 7a to 7c, and 8to 16 of the accompanying drawings.