Field of the Invention

The present invention relates to a hybrid energy storage system. In particular, a hybrid energy storage system for electricity storage is described.

Background

Large turboalternators in thermal power stations still underpin the electricity generation systems of today (2016). These machines provide energy storage at several different levels. The inductance of the stator windings smoothes out very high frequency variations in electrical power drawn (with time-spans in the order of 1ms or less). The inertia of the large spinning rotors provides for energy storage over time-frames of the order of 0.1 s - 10s. For longer duration variations in power, the steam-chest present in most systems provides a further buffer and beyond this, the control system on the boiler enables the system to draw upon massive energy stores (normally) in the form of fossil fuels. Properly understood, the generation systems that we have been using for over 100 years are, in fact, hybrid energy storage and generation systems. As the world migrates away from fossil-fuel combustion towards the low-carbon generation forms, the restoration of flexibility in the electricity system becomes a huge challenge. Most renewable energy forms (and especially the most common present-day forms of wind and solar power harvesting facilities) provide electrical power when the primary resource is present but have very little capability to deliver any power when that resource is not present.

Photovoltaic (PV) panels in particular have virtually no intrinsic energy storage so that if a cloud obscures the sun even momentarily, the energy output from a PV panel immediately drops. If controlled correctly, wind turbines are less of a problem in this respect since the rotating components of wind turbines do naturally store some energy but even these have so-called "inertia time constants" in the order of only a few seconds.

Whilst any one electricity system has only a small degree of penetration of renewable energy sources, the flexibility required by the system can be provided by the remaining fossil-fuelled generation. However, as the penetration of renewable energy generators rises, the flexibility problem rises more than proportionately. It is clear that the flexibility problem will become a serious issue in many locations worldwide as renewables continue to dominate all new power generation. New solutions are required.

Flywheels are one extremely attractive form of energy storage. Emulating the kinetic energy in generator rotors, a key advantage that they have is that they can experience very many energy cycles over their lifetimes with extremely high round-trip efficiency. Moreover, because the energy is already in the form of mechanical motion, it is relatively straightforward and direct to convert from the flywheel energy into electrical power (or vice-versa). A key problem with flywheels is that they are intrinsically relatively expensive. The amount of kinetic energy that can be stored in any spinning rotor is proportional to the volume of active structural material in that rotor and the maximum working stress that this rotor can withstand. The constant of proportionality varies over only a small range. With a given volume of structural material of known maximum working stress (i.e. a given total cost for the rotor structure), the maximum kinetic energy that can be stored is quite easily calculated even though the actual design could vary over a wide space.

To overcome the high expense of flywheels per unit energy stored, it is attractive to devise an energy storage system that can combine the excellent effectiveness of flywheels to manage many cycles with the much lower cost per unit energy capability of systems based on fluid flow. The resulting hybrid systems combine the best of both worlds - the high cycle, high turnaround efficiency aspects of the flywheel with the low cost-per-unit-of-energy stored feature of the fluid-flow based subsystem.

One important categorisation of flywheel energy storage systems is according to whether or not the motor/generator unit of a flywheel energy storage system is connected directly to the AC network (possibly via a transformer) or whether a set of power-electronics is located between that motor/generator unit and the AC network. An advantage of having the power-electronics in place is that the rotational speed of the magnetic fields within the motor/generator unit can be made independent of the AC frequency of the system served by the flywheel and this enables compact flywheels to be spun at very high speeds. If the spin speed of flywheel is restricted, then the diameter of the flywheel may need to be large relative to the axial length in order to achieve good utilisation of the material. Thus, the presence of the power-electronics removes one mechanical design constraint. However, there are disadvantages to having the power- electronics present and these disadvantages lie chiefly in the additional costs, losses and unreliability introduced by the power-electronics. The maj ority of all designs for flywheel energy storage systems interpose power-electronic converters between the generator and the AC grid and this is a maj or differentiation between such designs and the design of the present patent.

The system of interest in this invention has direct electrical connection (possibly via a transformer) between the windings of the motor/generator unit and the AC network and that motor/generator unit is a "synchronous machine" in the sense that if this machine is delivering steady power into the AC network and if the frequency of that AC network is not changing, the rotor of that machine must be spinning at a constant speed. Some synchronous machines use a wound rotor arrangement to create a rotor magnetic field that can interact with the stator. Other synchronous machines simply use permanent magnets on the rotor to cause the necessary magnetic interaction with the stator. The simplest of synchronous machines are probably synchronous reluctance machines that have neither permanent magnets nor rotor windings and they simply use saliency of ferromagnetic "poles" on the rotor to interact with the stator magnetic field. All such varieties of synchronous electrical machine (and combinations thereof) are relevant to the present invention.

There are other types of electrical machine that could be used with flywheel energy storage including doubly-fed induction generators (DFIGs) [ 1 ], [2] and whilst these have certain attractions, they are not directly relevant to the present invention. In a later section detailing the advantages of the present invention, flywheel systems including DFIGs are mentioned again for comparison purposes.

When the stator windings of a synchronous machine are connected directly to the AC network (possibly via a transformer) then the rotor of that machine must run at a speed determined by the frequency of the AC network. If the synchronous machine is a 2-pole design, then synchronous frequency is exactly the same as the AC grid frequency (3000rpm for 50Hz or 3600rpm for 60Hz). If the synchronous machine is a 4-pole design, then synchronous frequency is exactly one half of AC grid frequency (1500rpm for 50Hz or 1800rpm for 60Hz) etc .

Although the frequency of the AC grid has a definite nominal value in any territory (usually 50Hz or 60Hz), this frequency does vary slightly over a narrow band of values. The rotor of a synchronous machine naturally has some inertia of its own and if it is mechanically coupled to more spinning mass, then it can be endowed with much more inertia than the bare electromagnetic design of the rotor would indicate. As the AC grid frequency falls, this rotor naturally feeds energy into the grid - drawing from the store of kinetic energy in the rotor inertia. As the AC grid frequency rises, this rotor naturally withdraws energy from the grid - depositing that energy into its rotor kinetic energy. In this way, the spinning rotor automatically and passively acts to stabilise the grid but because the variation in frequency is very small, the amounts of energy being exchanged between the rotor store of kinetic energy and the grid are much smaller than the total amount of kinetic energy stored in the rotor. Specifically, if the maximum excursion of grid frequency away from nominal value is only 0.1 %, then the maximum energy that could be given up by the rotor inertia would be approximately 0.2% of the total stored kinetic energy. This is a very significant issue in terms of trying to produce cost-effective energy stores. A related issue is that there is no independent control over when rotor kinetic energy would be exchanged with the grid. To overcome the problems of (i) being able to access only a small fraction of the kinetic energy stored in a large spinning inertia and (ii) having independent control over when energy is pushed into the AC grid or withdrawn from it, a facility is required that can drive the rotor of the synchronous machine (henceforth termed the "magnet rotor") at a speed that is not identical to the speed of the large spinning inertia (henceforth termed the "flywheel rotor") or any fixed multiple of that speed.

One solution to this problem is proposed by Turner in GB25191 16 where a variable-ratio gearbox is connected between the large spinning inertia and the synchronous machine and the speed ratio of that gearbox is varied using a motor in order to tend to increase or decrease the speed of the synchronous machine rotor. Because a "gearbox" is fitted between the flywheel rotor and the magnet rotor, the sum of the torque on the flywheel rotor and the torque on the magnet rotor is not zero. This is implicit in the definition of a gearbox and it is precisely contrary to the nature of the present invention.

Summary of the Invention

According to an aspect of the present invention, there is provided a hybrid energy storage system for electricity storage, said system comprising: a first rotor for acting as a maj or store of kinetic energy; a second rotor for carrying sources of magneto-motive force; a differential drive unit coupled between the first rotor and the second rotor for exerting a torque acting between the first and second rotors; and a fluid subsystem for exchanging power with the differential drive unit using a fluid flow, said fluid flow passing through at least part of one or both rotors; wherein the differential drive unit exerts a substantially equal and opposite torque to the first and second rotors.

The present invention provides a system for storing energy for use, for example, to help balance an electricity grid. In embodiments of the present invention, the spin speeds of the flywheel rotor and the magnet rotor may be caused to be different by a differential drive. The differential drive may exert a positive torque on one of those rotors and (almost) the exact negative of that torque on the other rotor. This differential drive does not transmit any (significant) net torque to ground.

When the flywheel is being discharged, the torque can continue to push the magnet rotor forward at synchronous speed whilst decelerating the flywheel rotor. Evidently, the differential drive is inj ecting power into the magnet rotor during the flywheel discharging process. When the flywheel is being charged again, the differential drive absorbs power from the magnet rotor. The fluid subsystem, which may be considered to be an external power source (and sink) for the differential drive is a fluid system and when the flywheel rotor has decelerated to a standstill, this fluid system can continue to supply rated power to the motor/generator unit through the differential drive and the magnet rotor.

Embodiments of the present invention combine the advantage of having high real inertia with the intrinsic capability to generate full power from a fluid source. With the intended definition of fluid power, this encompasses all of the thermomechanical energy storage solutions including pumped-hydro, compressed air energy storage and pumped thermal energy storage so this invention embraces a wide class of hybrid energy storage systems in which a spinning inertia forms one energy store and the second energy store is compatible with very high capacities. The fluid power aspects of this invention may aid to permit the realisation of energy stores having extremely low costs per unit of total energy storage capacity. The flywheel aspects of this invention may mean that the response times to frequency-changing (or loss of power) are effectively instantaneous and the turnaround efficiency for fast alternations in power flow is extremely high. The simplicity and robustness of the flywheel aspect of this invention further may make it naturally suitable for use as an uninterruptable power supply (UPS). The present system may provide noteworthy advantages compared with energy storage (or UPS) systems based on flywheels where all of the power drawn from the generator passes through power-electronics. Obviously, the losses, additional unreliability and capital costs of the power-electronic converter are avoided but there are losses, potential unreliabilities and capital costs associated with the differential drive in the present invention. However there are other points. Because the power generated from the present system flows directly from the stator windings of a synchronous electrical machine, the total harmonic distortion is very low. In flywheels connected to the AC network via power-electronic converters, the finite torque rating of the electrical machine mean that it is not possible to achieve rated power output for flywheel speeds all the way down to zero. Moreover, there is also a minimum flywheel speed below which the electrical machine is simply not able to develop sufficient voltage to drive electrical power into the AC network. Thus it is not possible with those systems to release all of the kinetic energy stored in the flywheel rotor into the AC network. By contrast the present invention can exploit all of the kinetic energy stored in the flywheel rotor.

The present invention may also have noteworthy advantages compared with flywheel energy storage systems where the electrical machine is a doubly-fed induction generator (DFIG). In those systems, as the flywheel rotor decelerates, larger and larger fractions of the power being output from the generator stator are fed into the rotor. It is not practical for those systems to extract a high fraction of the kinetic energy stored in the flywheel. If the flywheel was run down to 50% of synchronous speed, then 25% of the kinetic energy that was present originally in the flywheel would still be present and 50% of all of the power emerging from the generator stator windings would be passing into the generator rotor. Thus, to supply a l OOkW load would require that the generator was rated at 200kW to cater for the low flywheel speed, increasing capital expenditure.

The present invention may also compete with systems comprising batteries and power-electronics sets. A list of advantages over these sets is threefold: (a) the cost per unit of energy storage is far lower than batteries can achieve, (b) most elements of the system have a near-infinite lifetime in stark contrast to batteries and (c) real inertia exists which is more valuable than "synthetic inertia" provided by batteries. Only one system element (the differential drive) has clearly a finite lifetime and the condition of that one element can be monitored very easily.

It may also be useful to compare this invention with energy storage systems where the only significant energy store is from a fluid subsystem of some sort. This might be a compressed air energy storage system, although the hydraulic fluid may also be pressurised using a liquid pump driven by an electric motor or a mechanically coupled shaft. In the case of the present invention, when quantities of energy are exchanged with the electricity system that are small relative to the total kinetic energy that can be stored in the flywheel, the energy may then be exchanged predominantly with the flywheel and the effective turnaround efficiency for these energy exchanges is very high. To illustrate this point, consider that the efficiency of the pure flywheel action is 98% and the efficiency of conversion of the fluid power is 70%. If we have one exchange of energy with the grid where the total energy exported to the grid is identical to the total kinetic energy storable in the flywheel, then the 75% of that energy has come from the flywheel and only 25% has been drawn from the fluid subsystem. The losses in this case would be estimated as (2% x 0.75 + 30% x 0.25) = 9% - much lower than they would have been for an energy storage system based purely on the conversion of energy from a fluid subsystem.

In an embodiment, the fluid flow may be a liquid. The fluid may be a hydraulic fluid, such as hydraulic oil or distilled water.

Additionally or alternatively, the first rotor may be a flywheel rotor having a higher inertia than the second rotor. By a flywheel rotor is intended to mean a rotor having a relatively high inertia as described previously. The flywheel rotor may be configured to rotate with an average speed substantially equal to the, or an integer sub-multiple of, the nominal alternating current (AC) electrical frequency of an AC electrical system.

In embodiments, the second rotor may rotate relative to an electrical machine stator. The stator may have electrical windings in direct electrical connection to the/an AC electrical system. Alternatively, the stator may have electrical windings connected to the/an AC electrical system via one or more transformers. In embodiments, the fluid subsystem may be external to the first and second rotors and to the differential drive unit. Accordingly, the fluid subsystem may sends power into, and/or absorbs power from, the differential drive system.

It may be appreciated that the differential drive unit may be considered to act to cause a controllable difference between spin speeds of the first and second rotor.

In a preferred embodiment the differential drive unit may comprise one or more reversible positive-displacement pump units such that a pressure difference across one or more of said pump units generates torque within the differential drive unit. A positive pressure difference across one or more of said pump units may then generate a positive torque within the differential drive unit, accelerating the second rotor and decelerating the first rotor. Additionally or alternatively, a negative pressure difference across one or more of said pump units may generate a negative torque within the differential drive unit, decelerating the second rotor and accelerating the first rotor. Each pump unit may be associated with a control valve set for producing said pressure difference across said pump unit. A positive, negative and zero pressure difference may be generated across each of said one or more pump units.

In further embodiments, the fluid subsystem may comprise a high pressure manifold and a low pressure manifold. The fluid subsystem may then provide power into the differential drive as the fluid enters the differential drive unit from the high pressure manifold and the fluid exits the differential drive unit to the low pressure manifold. Additionally or alternatively, it may provide power when the power flows from the differential drive, the fluid enters the differential drive unit from the low pressure manifold and the fluid exits the differential drive unit to the high pressure manifold.

A frequency control system may be provided for monitoring an electrical frequency. The frequency control system may communicate with the differential drive unit to draw power from, or transfer power to, the fluid subsystem depending on a desired electrical frequency.

In some embodiments an internal brake to prevent relative rotation between the first and second rotors is provided. The internal brake may be included in the differential drive unit. The internal brake may be configured to engage on the differential drive unit for reducing a difference in rate of rotation between the first and second rotors. The internal brake may lock the differential drive unit during normal operation of the system. Optionally the first and/or second rotors may have hollow shafts through which the fluid flow passes. In an alternative embodiment, the fluid flow may be compressed air.

The differential drive unit may alternatively comprise a plurality of hydraulic rams acting for controlling the fluid flow by one or more digitally controlled valves.

Typically the second rotor can be supported on bearings, said bearing supported on a stationary frame. Alternatively a shaft of the first rotor may protrude under the second rotor, and wherein the second rotor may be supported on bearings relative to the shaft of the first rotor. The differential drive unit may then occupy a space radially located between the shaft of the first rotor and second rotor. Fluid flow may then pass radially outward from the fluid subsystem into the differential drive unit. A second brake may also be provided to prevent the first rotor from reversing direction relative to an intended direction of rotation. A sprag clutch or other one-way clutch may be provided.

In an alternative example, a hybrid energy storage system comprising a first rotor acting as a maj or store of kinetic energy, a second rotor spinning about the same axis carrying the rotating element of a normal synchronous electrical motor/generator, a differential drive unit connected between the first rotor and the second rotor such that it can exert a torque acting between the two rotors to affect the difference in spin speeds of those rotors and an external energy store capable of sending power into the differential drive unit or absorbing power from that differential drive unit with said power being carried by fluid flow passing through one or both rotors is provided.

The fluid used to deliver power into the differential drive unit and to absorb power from the differential drive unit may be a liquid. The stator of said normal synchronous electrical motor/generator may have windings in direct electrical connection to the/an AC electrical system. The stator of said normal synchronous electrical motor/generator may alternatively have windings connected to the/an AC electrical system via one or more transformers.

It can be appreciated that, although certain examples and embodiments described above have been primarily described with respect to a single aspect, the features described are also applicable to the other aspects defined herein.

These and other aspects of the invention will be apparent from, and elucidated with reference to, the embodiments described hereinafter. The above discussion is not intended to represent every example embodiment or every implementation within the scope of the current or future Claim sets. The Figures and Detailed Description that follow also exemplify various example embodiments. Various example embodiments may be more completely understood in consideration of the following Detailed Description in connection with the accompanying Drawings. Detailed Description

The invention is described in further detail below by way of example and with reference to the accompanying drawings, in which:

Figure 1 shows a simplified schematic of a cross-sectional view of a hybrid energy storage system according to an embodiment of the present invention;

Figure 2 shows a stylised view within a differential drive used with the system of figure 1 ;

Figures 3a-3c show valve arrangements of the differential drive of Figure

2;

Figure 4 is a graph outlining how power within the system of Figure 1 changes over time; and Figure 5 shows an alternative view more detailed view of the system of Figure 1 .

The basic format of the invention is described with the aid of Figure 1. Four maj or components are involved. A first rotor 1 , henceforth referred to as the flywheel rotor forms the main store of kinetic energy for the system. In normal operation, this rotor spins with an average speed equal to the AC frequency (or some integer sub-multiple of that like one-half of that or one-third of that respectively). A second rotor 2, henceforth referred to as the magnet rotor, carries sources of magneto-motive force (MMF) thus creating a magnetic field that always spins with the magnet rotor. These sources of MMF would most often be simple permanent-magnet pieces but in some cases (especially for very large machines) it could be appropriate to have electrical windings on this rotor. In some cases, the magnet rotor might simply be the rotor of a synchronous reluctance machine. The electrical and magnetic design of the magnet rotor 2 would follow long-established principles of synchronous machine design that need not be restated here since any skilled machine designer would know these principles. The flywheel rotor 1 and the magnet rotor 2 are coupled via a differential drive 3 which can exchange power with a fluid subsystem. In most embodiments of this invention, the fluid in the fluid subsystem is a liquid such as hydraulic oil or distilled water and the differential drive 3 comprises positive-displacement machine elements such as gear-pumps or screw-pumps that can naturally act either as hydraulic motors or hydraulic pumps. The purpose of the differential drive 3 is to cause a controllable difference in spin speeds between the magnet rotor 2 and the flywheel rotor 1.

Figure 5 indicates the presence of an internal brake 38 is present within the differential drive unit to prevent relative motion between the magnet rotor 2 and the flywheel rotor 1 in a controlled manner. The differential drive unit 3 would normally contain at least two (usually three or more) discrete motor/pump units 3 1 , 32, 33 ... all coupled directly to the differential drive unit 3 such that when there is a positive pressure difference across any one of these motor/pump units (e.g. 33), that motor/pump unit (e.g. 33) develops some positive torque within the differential drive unit 3 tending to drive the magnet rotor 2 forwards and tending to decelerate the flywheel rotor 1. Similarly, when there is a negative pressure difference across any one of these motor/pump units (e.g. 33), that produces a torque in the opposite direction - tending to decelerate the magnet rotor 2 and tending to accelerate the flywheel rotor 1. There is one control valve set 51 , 52, 53 etc. associated with each one of the motor/pump units 3 1 , 32, 33 etc.. Each control valve set 51 , 52, 53 etc. is configured such that it can produce a positive or zero or negative pressure across its respective motor/pump unit 3 1 ,

32, 33 etc..

Figure 2 illustrates a possible arrangement for one control valve set (e.g. 5 1). Although this comprises numerous different controllable open/close passages { 51 a, 51b, 51 c, 51 d, 5 1 e}, it is clear that apart from the one-way valve elements { 51f,51 g,51i and 51h} , all can be operated by turning a single spool. Figures 3a, 3b and 3c show the control valve set 51 in three different configurations corresponding to positive pressure difference, zero pressure difference and negative pressure difference respectively. Evidently, these configurations correspond to torque states of positive, zero and negative respectively between magnet and flywheel rotors. Note that a fourth configuration of each control valve set is possible that is not shown in Figure 3. This is the locked state where all the controllable open/close passages { 51 a, 51b, 51 c, 51 d, 51 e} are closed. If the corresponding motor/pump unit 3 1 had zero leakage, this locked state would prevent any relative motion within the differential drive unit 3.

The magnet rotor 2 spins relative to an electrical machine stator 4 in exactly the same way that all synchronous machines operate. Indeed, if the magnet rotor 2 and the electrical machine stator 4 are considered in isolation, they form a normal synchronous electrical machine. In most embodiments the electrical machine has a cylindrical airgap with the magnet rotor 2 spinning within the stator of an electrical machine 4.

The fluid subsystem has two manifolds - a high pressure manifold 5 and a low pressure manifold 6. When the fluid subsystem is providing power into the differential drive 3, the fluid enters the differential drive at the pressure of the high pressure manifold 5 and leaves at the pressure of the low pressure manifold 6. When power flows from the differential drive 3 back into the fluid, the reverse occurs - fluid is forced back into the high pressure manifold 5 being drawn from low pressure manifold 6.

In normal operation, the differential drive 3 is locked by the internal brake 38 so that the flywheel rotor 1 and the magnet rotor 2 form a single inertia in effect. In this normal operation mode, the system acts as a simple synchronous flywheel. As slight variations occur in the frequency of the AC system to which the electrical machine stator 4 is connected, packets of energy naturally flow into and back out of this combined single inertia.

When some event occurs such as a serious outage where the energy storage system is called upon to deliver high power for an extended period, the flywheel action occurs completely automatically. As the flywheel rotor 1 surrenders some of its kinetic energy via the magnet rotor 2 into the stator 4, the spin speed of the combination gradually falls. This spin speed is easily detected by measurements of voltages at the terminals of the electrical machine stator 4 using some established frequency-sensing device 41. A frequency control system 39 continuously monitors the electrical frequency sensed by the frequency-sensing device 41.

When this electrical frequency has fallen below some acceptable frequency threshold, this frequency control system 39 first operates control valve sets 51 , 52, 53, etc. within the differential drive unit 3 exposing high pressure fluid from the fluid subsystem to one side of one or more of the motor/pump units 3 1 , 32, 33 ... etc.. It also releases the internal brake 38 in the differential drive 3 so that the fluid power acts to drive the magnet rotor 2 forwards whilst simultaneously acting to decelerate the flywheel rotor 1. The frequency control system 39 then continuously monitors the electrical frequency - adjusting the control valve sets 51 , 52, 53, etc. as necessary to increase the torque being supplied by the fluid flow if the magnet rotor 2 is spinning below rated speed or to decrease that torque if the magnet rotor 2 is exceeding its rated speed. At any one instant, if the flywheel rotor is turning at, say, 55% of rated speed, then 55% of the power passing across the airgap between the magnet rotor 2 and the electrical machine stator 4 is being drawn from the flywheel kinetic energy and the remaining 45% is being provided by fluid power driving the differential drive 3. Figure 4 shows how power from the flywheel rotor 1 and power from the fluid stream vary with time during a system discharge event in which the flywheel decelerates completely to zero speed. It is seen that both powers vary approximately linearly with time and the sum of the two remains constant (and equal to the output power). After the flywheel has stopped turning, all subsequent power is supplied from the fluid subsystem.

If the serious outage continues to the point where the flywheel rotor 1 has decelerated completely to zero speed, then a second brake 1 1 (which may be a sprag clutch or other one-way clutch in some instances) acts to prevent the flywheel rotor 1 from reversing direction relative to the stator frame of reference. At this time the flywheel has no further kinetic energy to surrender and all power travelling across the airgap between the magnet rotor 2 and the electrical machine stator 4 is being provided by fluid power driving the differential drive 3. That situation can continue for as long as there is energy left in the external source providing the fluid power.

When the system is no longer required to deliver power from the motor/generator stator 4, the magnet rotor 2 continues to spin at its synchronous speed and no significant torque is required to maintain its speed. The control valve sets 51 , 52, 53 etc. are all set to the "zero-torque" state (as indicated in Figure 3b). The speed of the flywheel rotor 1 must be returned to its normal value of spinning synchronously with the magnet rotor 2. At the appropriate time, the control valve sets 51 , 52, 53, etc. are placed into the "negative torque" state so that the flywheel is accelerated positively. The slip-power (the product of torque in the differential drive and the difference in speeds between the magnet rotor 2 and the flywheel rotor 1) is absorbed by the fluid and this forms a part of the recharging process for the fluid subsystem.

An alternative method for accelerating the flywheel rotor 1 again involves engaging the internal brake 38 on the differential drive 3 gradually - tending to reduce the speed differential between the flywheel rotor 1 and the magnet rotor 2. This must be done gradually to prevent the speed of the magnet rotor from dropping unacceptably (possibly leading to "pole-slip"). A substantial amount of energy can be lost in this operation and this energy is dissipated as heat and wear in the brake 38. In some embodiments, this may be acceptable and it has the slight attraction that the control valve sets 51 , 52, 53 etc. can be simpler because the negative torque state (shown in Figure 3c) does not have to be provided. Embodiment #1.

In a first embodiment of the system, the magnet rotor 2 has a 2-pole field created by permanent magnets and thus its rated spin speed is identical to the nominal electrical frequency (50Hz = 3000rpm in Europe and 60Hz = 3600 rpm in the USA). The drive end of the magnet rotor 2 is coupled to a first side of the differential drive 3. A large flywheel rotor 1 is coupled directly to the other side of the differential drive 3. Each rotor has a hollow shaft through which a hydraulic fluid can pass. At the non-drive end of the flywheel rotor 1 , there is a rotating union 19 enabling the fluid to pass between the high pressure manifold 5 (in the stationary frame) and the rotating frame of the flywheel rotor 1. At the non-drive end of the magnet rotor 2, there is a rotating union 29 enabling the fluid to pass between the low pressure manifold 6 (in the stationary frame) and the rotating frame of the magnet rotor 2. The bearings at the drive ends of both the flywheel rotor 1 and the magnet rotor 2 are on flexible supports so that small degrees of misalignment of the bearing centres can be tolerated. Figure 5 shows this arrangement. In this embodiment the fluid subsystem that can provide power to the differential drive unit 3 is a compressed air energy storage system in which a multiplicity of simple cylinders with controlled valve-gear at the top of each one allows the exchange of energy between compressed air and the hydraulic fluid. When pressurised air pushes the hydraulic fluid out of one of these cylinders, the air has transferred work to the fluid. When the fluid entering a cylinder raises the pressure of air in a cylinder and then discharges that air into a pressurised reservoir, that fluid has transferred work into the air. Patent US5073090A describes such a single-stage compression/expansion machine where air is compressed (or expanded) by the induction (or expulsion) of a liquid at the bottom of a cylinder.

An electrical machine stator 4 surrounds the magnet rotor such that the combination of the magnet rotor 2 and the electrical machine stator 4 forms a complete synchronous machine with radial airgap. A differential drive 3 is present comprising at least three independent positive-displacement hydraulic motors/pump units 3 1 , 32, 33 of different capacities with each one directly coupled to the relative motion within the differential drive 3. When the differential drive is operating at any one differential speed between the two rotors, the first motor/pump 3 1 will be passing a flow rate equal to one half of the flow rate being passed by the second motor/pump 32 whilst the second motor/pump 32 is passing a flow rate equal to one half of the flow rate being passed by the third motor/pump 33. Each motor/pump unit 3 1 , 32, 33 within the differential drive 3 has a respective control valve set, 51 , 52, 53. By switching different motor/pump units between the positive-torque and the zero-torque states using the control valve sets, seven different levels of positive torque can be achieved with approximately equal increments of torque between them. Similarly, by switching different motor/pump units between the negative-torque and the zero-torque states, seven different levels of negative torque can be achieved.

A small exciter 36 is incorporated onto the same side of the differential drive 3 as the magnet rotor 2. The purpose of this exciter is to provide a small amount of power into that unit for the purposes of operating control valve sets. Since this side of the differential drive will always be spinning at synchronous speed, the required power will always be present. A small transmitter 35 also incorporated into the differential drive 3 enables control signals to be transmitted into that unit for the purposes of dictating when the control valves should be operated.

The differential drive 3 always has a positive or zero slip speed (with the magnet rotor 2 always travelling at least as fast as the flywheel rotor 1 and sometimes faster).

This system has three main modes of operation: (i) normal operation as a pure high-inertia idling synchronous machine, (ii) operation in discharge mode and (iii) operating in recharge mode. In normal operation, the internal brake 38 within the differential drive unit 3 is engaged and the magnet rotor 2 is hard- coupled to the flywheel rotor 1. Discharge mode may be instigated when the electrical frequency detected by a frequency-sensor 41 fitted to the electrical machine stator 4 has fallen below some defined threshold value. Alternatively it may be instigated separately. Once discharge mode is instigated, the internal brake 38 within the differential drive unit 3 is released and a controller begins to act on the control valve sets 51 , 52, 53, to place some or all of the motor/pump units into positive torque state. In this state, the flywheel rotor 1 is being decelerated and its energy is being transmitted into the magnet rotor - together with energy from the fluid subsystem powering the differential drive unit 3. Discharge mode continues with the flywheel rotor 1 continually surrendering kinetic energy until finally it has no kinetic energy left. At that point, a sprag clutch 1 1 fitted to the flywheel rotor 1 prevents the rotor from spinning in the reverse direction and the fluid subsystem continues to supply all power to the differential drive unit 3. When the discharge event is over, the flywheel rotor 1 is caused to accelerate again by placing one or more of the control valve sets 51 , 52, 53 into negative torque state. This has the effect of pumping some energy back into the fluid subsystem.

Embodiment #2.

A second embodiment of the system is closely related to embodiment #1 except that the differential drive unit 3 achieves control over the torque developed in that unit using methodologies now commonly referred to as "digital hydraulics". Thus, in place of having multiple motor/pump units all connected in parallel to the relative motion in the differential drive unit there is a set of hydraulic rams acting against a cam-plate or crankshaft and the flow of hydraulic fluid into or out of each hydraulic ram is controlled by one or more digitally-controlled valves that can expose the cylinder to low pressure and by one or more digitally-controlled valves that can expose the cylinder to high pressure.

Embodiment #3.

In a third embodiment of the system, the magnet rotor 2 has a 4-pole field created by permanent magnets and thus its rated spin speed is equal to one half of the normal electrical frequency (50Hz ->1500rpm in Europe and 60Hz - >1800rpm in the USA). One attraction of a 4-pole field is that the flux density at the centre of the rotor is naturally zero in stark contrast to a 2-pole field where the flux ideally passes straight through the centre of the rotor. This in turn makes it more feasible to introduce a hole at the centre of the rotor (for conveying fluid to/from the differential drive. The main disadvantage of this choice of pole-number for the machine is that the nominal diameter of the flywheel must be larger than it would have been for a 2-pole machine. In this embodiment, the flywheel rotor 1 comprises a large cylindrical body with a relatively narrow shaft protruding underneath the magnet rotor 2. In this embodiment, the magnet rotor 2 is supported on bearings relative to the shaft of the flywheel rotor 1. This is in contrast to embodiments 1 and 2 where the magnet rotor 2 was supported on bearings that were in turn supported directly off the stationary frame. The differential drive unit 3 in this case occupies a space radially located between the shaft of the flywheel rotor 1 and the magnet rotor itself 2. The shaft of the flywheel rotor 2 is hollow but has an internal barrier separating a high-pressure chamber of that shaft from a low-pressure chamber. This system operates on the same principles as outlined earlier but now when the system is discharging, fluid passes radially outward from the high pressure chamber of the shaft of the flywheel rotor 1 into one part of the differential drive unit 3 where it does some work turning some machine elements there before finally re-emerging into the low-pressure chamber of the rotor.

From reading the present disclosure, other variations and modifications will be apparent to the skilled person. Such variations and modifications may involve equivalent and other features which are already known in the art of building cladding structures, and which may be used instead of, or in addition to, features already described herein. Although the appended claims are directed to particular combinations of features, it should be understood that the scope of the disclosure of the present invention also includes any novel feature or any novel combination of features disclosed herein either explicitly or implicitly or any generalisation thereof, whether or not it relates to the same invention as presently claimed in any claim and whether or not it mitigates any or all of the same technical problems as does the present invention.

Features which are described in the context of separate embodiments may also be provided in combination in a single embodiment. Conversely, various features which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable sub-combination. The applicant hereby gives notice that new claims may be formulated to such features and/or combinations of such features during the prosecution of the present application or of any further application derived therefrom.

For the sake of completeness it is also stated that the term "comprising" does not exclude other elements or steps, the term "a" or "an" does not exclude a plurality, and reference signs in the claims shall not be construed as limiting the scope of the claims.

Other embodiments are intentionally within the scope of the invention as defined by the appended claims.