WO2014181081A1 - Energy transfer apparatus and distribution control method therefor - Google Patents

Energy transfer apparatus and distribution control method therefor Download PDF

Info

Publication number
WO2014181081A1
WO2014181081A1 PCT/GB2014/051215 GB2014051215W WO2014181081A1 WO 2014181081 A1 WO2014181081 A1 WO 2014181081A1 GB 2014051215 W GB2014051215 W GB 2014051215W WO 2014181081 A1 WO2014181081 A1 WO 2014181081A1
Authority
WO
WIPO (PCT)
Prior art keywords
power
power transfer
voltage
module
converter
Prior art date
Application number
PCT/GB2014/051215
Other languages
French (fr)
Inventor
Nilanjan Mukherjee
Danielle STRICKLAND
Original Assignee
Aston University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to GBGB1308189.8A priority Critical patent/GB201308189D0/en
Priority to GB1308189.8 priority
Priority to GB1315787.0 priority
Priority to GBGB1315787.0A priority patent/GB201315787D0/en
Application filed by Aston University filed Critical Aston University
Publication of WO2014181081A1 publication Critical patent/WO2014181081A1/en

Links

Classifications

    • HELECTRICITY
    • H02GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02JCIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00Circuit arrangements for ac mains or ac distribution networks
    • H02J3/28Arrangements for balancing of the load in a network by storage of energy
    • H02J3/32Arrangements for balancing of the load in a network by storage of energy using batteries with converting means

Abstract

An apparatus for storing energy from and/or supplying energy to an electrical grid. The apparatus has a plurality of units for storing energy and/or: for supplying energy, A distribution of power conveyed between the eiectrical grid and the plurality of units is dynamically controlled based on the operation, states of the units:. Because the distribution of power is dynamically controlled based on the; operation states' of the units, units with different operation states, for example second-life batteries having different battery characteristics and/or different battery parameters, can be included in the apparatus.

Description

ENERGY TRANSFER APPARATUS AND
DISTRIBUTION CONTROL METHOD THEREFOR
Field of the invention
The present invention relates to an energy transfer apparatus for conveying power to and/or from an electrical grid.
The present invention also relates to a method of distributing power within an energy transfer apparatus for conveying power to and/or from an electrical grid.
Background of the invention Energy storage systems based on battery technologies are operating in existing electrical grid systems to improve the stability and reliability of grid/microgrid systems [1] . They can be used for grid ancillary services such as
primary/secondary frequency support applications and for peak load management (peak load lopping) , voltage support, renewable integration, amongst other applications [2] . The majority of current research on Battery Energy Storage Systems (BESS) is two-fold: a) electrical grid side control and/or energy management in renewable energy systems (mostly in wind and photovoltaic (PV) systems) to smooth out the power oscillation or power mismatch [3] -[5] and b) battery side control or Battery Management Systems (BMS) [6] -[7] to avoid cell degradation and to enhance the utilisation of a series connected string. Traditionally, a two-stage or single-stage converter is used with large numbers of series batteries, especially in low to medium voltage applications [8]-[9]. In all these cases, the same battery chemistry and sub module size is used. Furthermore, since there could be voltage and/or state-of-charge (SOC) imbalance among the cells, charge/SOC and/or voltage equalization is required. Previous research, has been carried out on balancing/equalization circuits (active and passive) of series connected cells [10]- [13] in order to avoid premature battery cell damage.
A new BESS can be a costly investment. To assist with cost reductions, there is significant interest in industry
(e.g. ABB and G ) in using Second Life Battery Energy Storage Systems (SLBESS) as a grid-tie energy storage system [14]- [15] . SLBESS include used batteries, for example
transportation batteries that have been used in Electric Vehicles (EVs) or Hybrid Electric Vehicles (HEVs) . However, these schemes all use batteries from a single source. As such, the batteries have the same chemistry and size and may also have similar performance.
At present, the recycling chain has not been established for second life batteries and is unlikely to be mature until the uptake on new EVs and HEVs is in sufficient numbers to warrant the investment necessary. This means that the SLBESS based on used transportation batteries available for grid applications in the years up to 2020 are likely to be formed from batteries from different manufacturers. These batteries are likely to have different battery chemistries and could be in significantly different conditions, depending on their respective driver and drive cycle behaviour in their first life. Around 2020 the availability of second life batteries is expected to be significantly improved [16] .
Different vehicles may use different battery chemistries (e.g. lead acid, nickel metal hydride (NiMH), lithium-ion) and may have batteries that come in a range of different sizes (in terms of power and/or energy) depending on the specific application (e.g. EV, HEV, bus or motorbike). Therefore, when transportation batteries from different vehicles become available for second use applications, they could be
significantly different from each other in terms of
chemistry/type, capacity, nominal voltage, initial state-of- charge (SOC) , internal impedance, charging-discharging limits, physical size, cooling system requirements etc. Moreover, hybrid/heterogeneous second life battery parameters are more prone to vary and fail (low reliability) during the second life operation. Therefore, in a BESS including second-life batteries a traditional large series string using only an equalization strategy based on cell voltage or SOC (as used in existing BESS and SLBESS) will not be suitable (e.g. will not be reliable or appropriate) .
Therefore, there exists a problem of finding a suitable way to include (i.e. to hybridise) different types of second- life batteries, for example different types of transportation batteries subsequent to their primary use in low and ultra-low carbon vehicles, in a Battery Energy Storage System. Summary of the invention
At its most general, the present invention relates to an apparatus for storing energy from and/or supplying energy to an electrical grid. The apparatus has a plurality of units for storing energy and/or for supplying energy. A
distribution of power conveyed between the electrical grid and the plurality of units is dynamically controlled based on the operation states of the units. Because the distribution of power is dynamically controlled based on the operation states of the units, units with different operation states, for example second-life batteries having different battery characteristics and/or different battery parameters, can be included in the apparatus .
According to a first aspect of the present invention, there is provided an energy transfer apparatus for conveying power to and/or from an electrical grid, the apparatus comprising :
a plurality of power transfer modules, each power transfer module comprising a power transfer unit for storing DC power and/or for supplying DC power; power conversion means arranged to:
convert between AC power in the electrical grid and DC power in the plurality of power transfer modules; and
convert between a local DC voltage across each power transfer unit and a respective power transfer module voltage on a grid-side of the respective power transfer module; and a controller communicably connected to the power
conversion means and arranged to:
determine an operation state for each power transfer unit; and
dynamically adjust a distribution of power conveyed between the electrical grid and the plurality of power transfer modules based on the operation states for each of the power transfer units.
A "grid-side" of the power transfer module (or other) may mean the side (e.g. an input and/or output side) of the power transfer module (or other) that is closest to the electrical grid, e.g. a side that receives an electrical input from and/or outputs an electrical output to the electrical grid (directly or indirectly) .
Therefore, with the apparatus of the present invention it is possible to include (i.e. to hybridize) power transfer units (e.g. generation and/or storage units) having different operation states in a single energy transfer apparatus, because the distribution of power conveyed between the electrical grid and the power transfer modules is dynamically adjusted based on the determined operation states of the power transfer units. In other words, the real-time (or other, e.g. recently determined) operation states of the power transfer units are used to dynamically control how power received from the electrical grid to be stored in the power transfer units is distributed between the power transfer modules, and/or how supply of power from the power transfer units to the
electrical grid is distributed between the power transfer modules. Therefore, with the present invention the distribution of power may be dynamically changed to take into account changes in the operation states of the power transfer units during the operation of the apparatus. Where the power transfer units are not homogeneous units, e.g. where they have different ages, sizes, etc., their operation states may vary in different ways during the operation of the electrical transfer apparatus. Such variation in the operation states of the power transfer units may be taken into account in the present invention by dynamically adjusting the distribution of power based on the determined operation states of the power transfer units. For example, the operation states of the power transfer units may be repeatedly determined on a regular (e.g. periodic) or irregular (e.g. non-periodic) basis.
The apparatus according to the first aspect of the present invention may have any one, or, to the extent they are compatible, any combination of the following optional
features .
The power conversion means may comprise a DC to DC converter in each of the plurality of power transfer modules arranged to convert between the local DC voltage across the power transfer unit and a DC power transfer module voltage on a grid-side of the DC to DC converter. Including a DC to DC converter in each of the power transfer modules may provide better control over the sizes of the power transfer module voltages, which may allow for better control of the
distribution of power conveyed between the electrical grid and the power transfer modules. Including separate DC to DC converters in each of the power transfer modules may also improve the reliability of the apparatus.
The power conversion means may comprise a power converter for converting between AC power in the electrical grid and DC power in the apparatus, and the plurality of power transfer modules may be connected in series with a power transfer module side of the power converter, so that a total DC voltage across the power transfer module side of the power converter is a sum of the DC power transfer module voltages. For example, the topology of the power conversion means may be a cascaded DC to DC converter to power converter (e.g. inverter and/or rectifier) topology. Including a single power converter (i.e. an inverter and/or rectifier) for converting between AC power on the electrical grid and DC power in the apparatus may mean that the conversion between AC power and DC power can controlled simply and easily. With this topology, each power transfer module may contribute a different DC power transfer module voltage towards the total DC voltage across the power converter. Therefore, with this arrangement it may be possible to control the distribution of power conveyed between the electrical grid and the power transfer modules by controlling the DC power transfer module voltages.
The controller may be arranged to dynamically adjust the distribution of power conveyed between the electrical grid and the plurality of power transfer modules to maintain the total DC voltage across the power transfer module side of the power converter at a constant level or within a predetermined range (e.g. a voltage range required by the power converter for connecting to the grid and within power converter switch ratings) . Maintaining the total DC voltage across the power converter at a constant level or within a predetermined range may allow for uninterrupted operation of the power converter. Therefore, the apparatus may be able to convey power to and/or from the electrical grid in a continuous manner, i.e. without interruption. Because the total DC voltage across the power converter is a sum of the respective DC power transfer module voltages, the total DC voltage may be maintained at a constant level or within a predetermined range without requiring that each of the respective DC power transfer module voltages is maintained at a constant level or within a predetermined range .
Alternatively, the power conversion means may comprise a power converter for converting between AC power in the electrical grid and DC power in the apparatus, and the plurality of power transfer modules may be connected in parallel with a power transfer module side of the power converter, so that the DC power transfer module voltages are the same as a DC voltage across the power transfer module side of the power converter. For example, the topology of the power conversion means may comprise a parallel DC to DC converter based topology.
The power conversion means may comprise a power converter in each of the plurality of power transfer modules arranged to convert between the DC power transfer module voltage on the grid-side of the DC to DC converter and an AC power transfer module voltage on a grid-side of the power transfer module. In other words, instead of a single power converter for converting between AC power and DC power there may be a plurality of power converters, one in each of the plurality of power transfer modules. For example, the topology of the power conversion means may comprise DC to DC converters in series with cascaded power converters (i.e. an inverter and/or rectifier) or in series with parallel power converters. This arrangement may allow for better control of the conversion between AC power and DC power. The inclusion of separate DC to DC converters and power converters for converting between AC and DC in each of the plurality of power transfer modules may allow for a high degree of control over the power
conversion, and may also improve the reliability of the apparatus .
The power conversion means may comprise a power converter in each of the plurality of power transfer modules arranged to convert between the local DC voltage across the power transfer unit and an AC power transfer module voltage on a grid-side of the power transfer module. The power converter may therefore perform a dual role of converting between DC power in the power transfer unit and AC power on a grid-side of the power transfer module and changing the magnitude of the voltage between the DC power in the power transfer unit and the AC power on the grid side of the power transfer module. For example, the topology of the power conversion means may comprise cascaded power converters (i.e. inverters and/or rectifiers) or parallel power converters.
The plurality of power transfer modules may be connected in series with the electrical grid, so that a total AC voltage across the electrical grid is a sum of the AC power transfer module voltages
Alternatively, the plurality of power transfer modules may be connected in parallel with the electrical grid, so that the AC power transfer module voltages are the same as an AC voltage across the electrical grid
Each DC to DC converter may controllable by a respective DC reference voltage, and the controller may be arranged to set each respective DC reference voltage based on the operation state for its respective power transfer unit. The DC to DC converter may control the DC power transfer module voltage of the respective power transfer module based on the respective DC reference voltage. For example, the DC to DC converters may control a DC voltage across a power transfer module capacitor of the power transfer module to be the same as the respective DC reference voltage. There may be a closed loop control system where the DC voltages across the power transfer module capacitors follow the desired reference voltages. For example, an outer DC voltage loop may control the DC voltages across the power transfer module capacitors while an inner loop may make sure that the power transfer unit currents remains within stable limits, e.g. in the case of a fault on one or more battery cells of the power transfer unit.
Where the power transfer modules are connected in series with a power converter for converting between AC power on the electrical grid and DC power in the apparatus, the DC to DC converters need to provide a combined DC voltage for the power converter (i.e. a converter and/or rectifier). Therefore, the DC to DC converters may need to operate within certain DC power transfer module voltages. This may be achieved by controlling the DC power transfer module voltages based on the respective DC reference voltages.
Providing each DC to DC converter with a respective reference voltage by which the DC to DC converter is
controllable may be a particularly suitable (i.e. efficient and/or easily controllable) way for the controller to control the distribution of power conveyed between the electrical grid and the power transfer modules. For example, the power transfer module power may be proportional to the DC power transfer module voltage of the power transfer module.
Therefore, by controlling (i.e. adjusting) the DC power transfer module voltages of the power transfer modules the power transfer module powers may be controlled (i.e.
adjusted) . In this manner, the distribution of power conveyed between the electrical grid and the power transfer modules may be controlled (i.e. adjusted). Because the reference voltages are set based on the operation states of the power transfer units, the operation states of the power transfer units are taken into account/used when determining the distribution of power, and changes in the operation states of the power transfer units may therefore lead to adjustments of the distribution of power.
The DC to DC converters of the plurality of power transfer modules may form a cascaded H-bridge DC to DC converter. A cascaded H-bridge DC to DC converter has been found to offer better value for money than a parallel DC to DC configuration. For example, in a parallel configuration with a DC bus voltage of 630V, to directly invert to a grid of 400V+10% a 24V sub-module would need to be boosted by a factor of 26 times to achieve this DC bus voltage, or the sub-modules would need to be connected together in series with a
subsequent loss in reliability. These problems may be avoided by using a cascaded H-bridge DC to DC configuration. The controller may be arranged to obtain a value representative of each of one or more characteristic
parameters of each power transfer unit in order to determine the operation state for that power transfer unit. Thus, the operation state of a power transfer unit may be determined based on a representative value of a single characteristic parameter of the respective power transfer unit.
Alternatively, the operation state of a power transfer unit may be determined based on representative values of more than one characteristic parameter of the respective power transfer unit. In other words, the operation state of the power transfer unit may collectively represent a number of different characteristic parameters of the power transfer unit. The representative value (s) may be obtained by performing one or more measurements on the respective power transfer unit or on its respective module and/or by performing a calculation or estimation.
The controller may be arranged to: calculate a weighting factor for each power transfer unit from the obtained values representative of the characteristic parameters of that power transfer unit, and apportion a DC (battery or other) current corresponding to DC power in the apparatus according to the weighting factor for each power transfer unit. Many
characteristic parameters of a power transfer unit may be relatable to the current of the power transfer unit through equations and/or lookup tables. For example, battery parameters such as state-of-charge (SOC) , voltage, impedance, state-of-health (SOH) or charge (Q) can be related to the battery current through equations and/or lookup tables.
Therefore, it may be appropriate (e.g. efficient and/or easily controllable) to control the distribution of the power between the modules by controlling the apportionment of a DC (battery or other) current corresponding to the power between the modules . Where the power transfer unit in one or more of the plurality of power transfer modules comprises one or more battery cells, the characteristic parameters of the one or more battery cells may include one or more of: a voltage of the one or more battery cells, a capacity of the one or more battery cells, a state-of-charge (SOC) of the one or more battery cells, a state-of-health (SOH) of the one or more battery cells, an impedance of the one or more battery cells, and a temperature of the one or more battery cells. These characteristic parameters may be indicative of the current operation state of the one or more battery cells, and
therefore they may be suitable parameters to take into account when adjusting the distribution of power. For power transfer units other than battery cells (e.g. other types of energy storage units or generation means), the characteristic parameters may include (instead of, or in addition, to the above) temperature, fuel flow, partial shading and pressure (depending on the type of power transfer unit.
Each of the plurality of power transfer modules may include a bypass switch controllable by the controller to disconnect the power transfer module so that power is not conveyed between the power transfer unit of that power transfer module and the electrical grid. Therefore, if one of the power transfer modules experiences abnormal conditions and/or displays abnormal behaviour, for example high
impedance, overcharge or deep discharge (e.g. where the power transfer unit of the module comprises a battery) , the
controller can control the bypass switch of that power transfer module to disconnect that power transfer module, e.g. to disconnect it from a power converter to which it is connected. The bypass switch may also be usable to disconnect generating units, e.lg. a generator, when the flow of power is from the electrical grid to the energy storage device, in order to prevent unwanted back-feeding of the generating units. Where the power transfer modules are connected in series, in the absence of a bypass switch for disconnecting (i.e. bypassing) an abnormal failure of a single power transfer module could interrupt a common DC link current and cause consequent tripping of the entire circuit. In addition, if one power transfer module became unstable the instability (e.g. a voltage or current oscillation) could propagate through all of the power transfer modules, particularly where the power transfer modules each have capacitors that are connected in series. These problems may be avoided in the present invention by providing each power transfer module with a bypass switch controllable by the controller.
The controller may be switchable between a first state where power is supplied to the electrical grid from the power transfer units and a second state where power is received from the electrical grid and stored in the power transfer units. Therefore, the apparatus may be used to temporarily store power from the electrical grid when operating in the second state and to then later return some or all of the stored power to the electrical grid when operating in the first state, i.e. the apparatus may function as an energy storage system that can temporarily store energy from the electrical grid. In addition, or alternatively, the apparatus may actively generate and supply power to the electrical grid in the first state, i.e. through the operation of one or more electrical generators. The apparatus of the present invention may therefore be suitable for use in one or more grid support mechanisms, for example: frequency response, voltage support, peak load lopping, aiding renewable generation integration, customer energy management and/or supporting islanded micro grids. Some embodiments of the present invention may relate to a Battery Energy Storage System, a Second-Life Battery Energy Storage System, Hybrid Battery Energy Storage System, a Hybrid Energy Storage/Generation System and a Hybrid
Generation System (first state only) . The DC to DC converters may be selectively operable in any one of a boost mode, a buck mode, and a buck-boost mode. In boost mode, the output voltage magnitude of the DC to DC converter is greater than the input voltage magnitude, i.e. the DC to DC converter is a "step-up" DC to DC converter. In buck mode, the output voltage magnitude of the DC to DC converter is less than the input voltage magnitude, i.e. the DC to DC converter is a "step-down" DC to DC converter. In buck-boost mode the output voltage magnitude of the DC to DC converter is either greater than or less than the input voltage magnitude, i.e. the DC to DC converter is a "step-up" and a "step-down" DC to DC converter. In addition, or alternatively, the power converters may be operable in any one of a boost mode, a buck mode and a buck-boost mode. In other words, the power converters may increase or decrease the voltage magnitude when converting between AC power and DC power.
The power transfer unit in one or more of the plurality of power transfer modules may comprise one or more battery cells. For example, the power transfer unit may comprise a battery module that comprises a plurality of battery cells. For example, the apparatus of the present invention may include (i.e. hybridise) a power transfer unit comprising one or more battery cells from a first battery with a power transfer unit comprising one or more battery cells from a second battery. The one or more battery cells from the first battery may have a different operation state to the one or more battery cells from the second battery. The first and second batteries may both be new batteries. Alternatively, the first and second batteries may both be second-life (i.e. used) batteries. Alternatively, the first battery may be a new battery and the second battery may be a second-life (i.e. used) battery. Alternatively, or in addition, the apparatus of the present invention may hybridise a power transfer unit comprising one or more battery cells with a power transfer unit comprising another type of energy storage unit, such as a fuel cell, or with a generation device, such as an electrical generator. Of course, in other embodiments of the invention none of the modules may comprise one or more battery cells and instead they may all comprise other types of energy storage unit and/or energy generation means .
The one or more battery cells include one or more transportation batteries. A transportation battery may be a battery that has been specifically designed to power a vehicle, e.g. as a primary or a secondary power source. More particularly, a transportation battery may be a battery that has been specifically designed to power an electric vehicle (EV) or a hybrid electric vehicle (HEV) such as a car, motorcycle, scooter or bus. The one or more transportation batteries may comprise new (i.e. unused) transportation batteries. Alternatively, or in addition, the one or more transportation batteries may comprise second-life (i.e. used) transportation batteries. For example, the transportation batteries may have been previously used in low or ultra-low carbon vehicles, such as EVs or HEVs . The apparatus of the present invention may include (i.e. hybridise) more than one new (i.e. unused) transportation battery and/or second-life (i.e. used) transportation battery in a single apparatus.
Alternatively, or in addition, the apparatus of the present invention may hybridise a new or second life transportation battery with another type of inverter connected energy storage, such as a fuel cell, and/or with generation means, in a single apparatus.
According to a second aspect of the present invention, there is provided a method of distributing power within an energy transfer apparatus for conveying power to and/or from an electrical grid, the method comprising:
determining an operation state for a power transfer unit in each of a plurality of power transfer modules of the energy transfer apparatus, the power transfer units being arranged to store DC power and/or to supply DC power; and dynamically apportioning the power conveyed between the electrical grid and the plurality of power transfer modules based on the operation states for each of the power transfer units .
Therefore, with the method of the present invention it is possible to include (i.e. to hybridize) power transfer units having different operation states in a single energy transfer apparatus, because the power conveyed between the electrical grid and the power transfer units is apportioned based on the operation states of the power transfer units. In other words, the real-time (or other, e.g. recently determined) operation states of the power transfer units are taken into account when dynamically apportioning power received from the electrical grid to the power transfer units and/or when dynamically apportioning the supply of power from the power transfer units to the electrical grid. Therefore, with the method of the present invention the apportionment of power conveyed between the electrical grid and the power transfer units may be dynamically updated to take into account any changes in the operation states of the power transfer units.
The method according to the second aspect of the invention may comprise any one, or, to the extent that they are compatible, any combination of the following optional features. Optional features of the first aspect of the invention specified above may also be applicable to the second aspect of the invention, where compatible.
The method may comprise converting between a local DC voltage across the power transfer unit and a respective DC power transfer module voltage in each of the plurality of power transfer modules.
The plurality of power transfer modules may be connected in series with a power transfer module side of a power converter for converting between AC power in the electrical grid and DC power in the apparatus, so that a total DC voltage across the power transfer module side of the power converter is a sum of the DC power transfer module voltages; and
dynamically apportioning the power conveyed between the electrical grid and the plurality of power transfer modules maintains the total DC voltage across the power transfer module side of the power converter at a constant level or within a predetermined range. Maintaining the total DC voltage across the power converter at a constant level or within a predetermined range may allow for uninterrupted operation of the power converter (i.e. an inverter and/or rectifier) . Therefore, the method may be able to convey power to and/or from the electrical grid in a continuous manner, i.e. without interruption.
The method may be a method for conveying power from the energy transfer apparatus to the electrical grid, and dynamically apportioning the power conveyed between the electrical grid and the plurality of power transfer modules may include setting the discharge rates of the power transfer units in a manner whereby they would all fully discharge simultaneously. This may ensure that each power transfer unit is undertaking a proportionate share of the power discharge relative to the other power transfer units, and that none of the power transfer units should stop contributing to the power discharge during the operation of the energy transfer apparatus.
Determining an operation state for each power transfer unit may include obtaining a value representative of each of one or more characteristic parameters of that power transfer unit.
One or more of the power transfer units may comprise one or more battery cells, and the characteristic parameters of the one or more battery cells may include one or more of: a voltage of the one or more battery cells, a capacity of the one or more battery cells, a state-of-charge (SOC) of the one or more battery cells, a state-of-health (SOH) of the one or
SUBSTITUTE SHEET RULE 26 more battery cells, an impedance of the one or more battery cells, and a temperature of the one or more battery cells. These characteristic parameters may be indicative of the current operation state of the one or more battery cells, and therefore they may be suitable parameters to take into account when apportioning the power.
The method may include calculating a DC current for each power transfer unit based on the operation state of the power transfer unit. Therefore, the apportionment of the power conveyed between the electrical grid and the power transfer modules may be achieved by controlling the DC currents for each of the power transfer units. Many characteristic parameters of a power transfer unit may be relatable to the current of the power transfer unit through equations and/or lookup tables. For example, battery parameters such as state- of-charge (SOC) , voltage, impedance, state-of-health (SOH) or charge (Q) can be related to the battery current through equations and/or lookup tables.
Dynamically apportioning the power conveyed between the electrical grid and the plurality of power transfer modules may include: setting a reference voltage for a DC to DC converter in each of the plurality of power transfer modules, whereby the DC to DC converter converts between the local DC voltage across the power transfer unit and a respective DC power transfer module voltage on a grid-side of the DC to DC converter. Setting a reference voltage for each DC-DC converter may be a suitable (i.e. efficient and/or easily controllable) way to control and therefore to dynamically apportion the power conveyed between the electrical grid and the power transfer modules. The DC-DC converter may control the DC power transfer module voltage of the respective power transfer module based on the DC reference voltage for that power transfer module. For example, the DC to DC converter may control a DC voltage across a capacitor of the power transfer module to be the same as the DC reference voltage.
SUBSTITUTE SHEET RULE 26 For example, there may be a closed loop control system where the DC voltages across the power transfer module capacitors follow the respective reference voltages and there may also be an inner control loop that makes sure battery current remains within stable limits, in the case of a fault on one or more battery cells. Where the DC to DC converters are connected in series with a power converter for converting between AC power on the electrical grid and DC power in the apparatus, the DC to DC converters may have to provide a combined DC voltage across the power converter (e.g. and inverter and/or a rectifier) . Therefore, the DC-DC converters may need to operate within certain DC power transfer module voltages.
This may be achieved by controlling the DC power transfer module voltages based on the DC reference voltages.
The DC power in the power transfer module may be proportional to the DC power transfer module voltage of the power transfer module. Therefore, by controlling (i.e.
adjusting) the DC power transfer module voltages of the power transfer modules the DC power transfer module powers may be controlled (i.e. adjusted). In this manner, the distribution of power conveyed between the electrical grid and the power transfer modules may be controlled (i.e. adjusted).
Determining an operation state for each power transfer unit may be performed rapidly enough for dynamic apportioning to be performed in real time. Performing the dynamic apportioning in real-time means that the real-time (or other, e.g. recently determined) operation states of each of the power transfer units are taken into account at all times when apportioning the power. The method may therefore be able to rapidly change/adapt the apportioning of power in response to changes in the operation states of the power transfer units. This may prevent failure or abnormal behaviour in one or more of the power transfer units and may help to maintain the stability of the energy transfer apparatus.
SUBSTITUTE SHEET RULE 26 The method may comprise repeatedly, and optionally periodically, determining an operation state for each power transfer unit.
Brief description of the drawings
Embodiments of the present invention will now be discussed, by way of example only, with reference to the accompanying Figures, in which:
FIG. 1 shows module integrated multi-modular topologies that can be used in embodiments of the present invention, including: a) cascaded inverters and b) parallel inverters;
FIG. 2 shows module integrated multi-modular topologies that can be used in embodiments of the present invention, including a) DC to DC converters in series with cascaded inverters and b) DC to DC converters in series with parallel inverters ;
FIG. 3 shows module integrated multi-modular topologies that can be used in embodiments of the present invention, including a) cascaded DC to DC converter to inverter and b) parallel DC to DC converter to inverter;
FIG. 4 shows an example of a fault-tolerant multi-modular topology for l-φ second-life battery storage systems in a low- voltage grid system;
FIG. 5 shows three different battery models: a) resistive model, b) RC-model, c) simplified model;
FIG. 6 shows a method of online impedance estimation based on high frequency ripple;
FIG. 7 shows an experimental result of online impedance estimation for a 24V lithium titanate battery using 10 kHz switching frequency: a) measured voltage ripple and b) estimated impedance;
FIG. 8 shows a distributed control structure for a cascaded DC to DC converter;
SUBSTITUTE SHEET RULE 26 FIG. 9 shows an example of module DC link voltage based control per module;
FIG. 10 shows two control loops: a) module voltage control loop, and b) module current control loop;
FIG. 11 shows a frequency response plot of three modules for a particular set of controller parameters in discharging mode;
FIG. 12 shows examples of the overall control structures for a distributed SLBESS: (a) battery side distributed control, and (b) grid side control;
FIG. 13 shows experimental results of power sharing at time of connecting to grid in discharging mode;
FIG. 14 shows experimental results of distributed power sharing at time of connecting to grid in discharging mode;
FIG. 15 shows experimental results of transition from charging to discharging mode - case - 1;
FIG. 16 shows experimental results of transition from charging to discharging mode - case - 2;
FIG. 17 shows experimental results of transition from discharging to charging mode;
FIG. 18 shows experimental results of voltage dynamics on transition from charging to discharging mode;
FIG. 19 shows experimental results of voltage dynamics on transition from charging to discharging mode;
FIG. 20 shows experimental results of dynamic variation of module current;
FIG. 21 shows experimental results of battery impedance variation: discharging mode;
FIG. 22 shows experimental results of battery impedance variation: charging mode;
FIG. 23 shows experimental results of module bypassing variation: charging mode; and
FIG. 24 shows experimental results of module bypassing variation: discharging mode.
SUBSTITUTE SHEET RULE 26 Detailed description of the preferred embodiments and further optional features of the invention
In some embodiments of the present invention, one or more of the power transfer units may comprise one or more battery cells, e.g. one or more battery cells from an ex- transportation battery. For the purposes of clarity and conciseness, in the following discussion battery cells of an ex-transportation battery are used as an example of power transfer units. However, it is to be understood that the present invention is not limited to the power transfer units being battery cells. Instead, the present invention is applicable to other types of power transfer unit, either in combination with or instead of battery cells, such as other types of energy storage units (e.g. fuel cells) and/or generation means, such as an electrical generator (e.g. a photovoltaic module) .
The likely recycling route for ex-transportation
batteries is that the vehicle containing the transportation battery will be returned to the manufacturer. The
manufacturer will remove the battery from the vehicle and will supply the battery to third parties for second-life
applications, either directly or through an intermediary (e.g. via a battery re-cycler) .
The battery will be stripped down into sub-modules and tested before being leased or sold on for a second-life application. It may be impractical to strip the batteries down to cell level. However, within the battery there are likely to be modularized units which the battery can easily be reduced into. These sub-modules are likely to be minimally tested and sorted at the recycler or manufacturer before being sent on for a second-life application.
The voltage associated with these sub-modules may be anywhere in the range sub-12V (e.g. for the battery of the Honda insight vehicle) to upwards of 200V DC (e.g. for some Renault vehicles) depending on the manufacturer's
configurations. The battery sub-modules from a single battery will have seen similar life profiles and will likely contain battery cell balancing (either active or passive) , perhaps in conjunction with an available Battery Management System (BMS) .
Where the manufacturer refrains from passing on the BMS system protocols to the second life developer, a universal type BMS may be needed to ensure safety. CONVERTER TOPOLOGY
Some embodiments of the present invention may relate to combining different battery sub-modules together in a modular power electronic topology with a suggested distributed control strategy to take into account different voltages (and limits), state-of-charge (SOC) , state-of-health (SOH) and/or battery capacity during charging and/or discharging.
Key challenges to overcome in SLBESS are: a) poor battery reliability and b) the heterogeneous nature of the batteries. Therefore, converter topology plays an important role in integrating such an energy storage system. Previous work has shown that from a reliability perspective a multi-modular converter with module redundancy is better for such
applications [17] . To completely optimize the use of each cell would require a converter per cell. However, this approach would reduce the converter efficiency and increase the control complexity (and cost) [18] .
Module integrated multi-modular converters have been the focus of research into grid-tie photovoltaic (PV) applications where power imbalance, partial shading, cost/reliability and module mismatch are handled using such converters . Such converters can be broadly divided into two groups: DC side modular topologies and AC side modular topologies. An AC side modular converter or cascaded multilevel converters (cascaded H-bridges) are reported in high/medium voltage grid (3.3kV, 4.16kV and above) and/or high power applications (> lOOkW or MW levels) [19] -[20] because it can reduce switch stress, enhance the efficiency and increase reliability. However, in applications where the optimal utilisation of the batteries is desired, such a structure requires a dedicated DC to DC converter along with each H-bridge [21] . This demands a large number of overall switches, drivers, sensors, and components that can increase the overall system cost and controller complexity in low voltage grid applications. Therefore, a DC side modular converter which uses multiple DC to DC modules on the DC side and a single inverter module on the grid side are considered to be more suited in low-voltage (LV) grid applications, for example PV applications [22]-[25]. This architecture can also integrate heterogeneous DC sources, and the modularity on the DC side can operate with poor battery reliability.
Examples of module integrated multi-modular topologies that can be used in embodiments of the present invention are illustrated in FIGS. 1 to 3.
FIG. 1 a) shows a cascaded inverter/rectifier topology.
In this topology the apparatus comprises a plurality of modules 1, each of which comprises a power transfer unit in the form of a battery module 3 (of course, in other
embodiments of the invention other types of power transfer unit may be present instead of a battery module 3, such as other types of energy storage units or types of generation means) and a power converter 5 (e.g. an inverter and/or a rectifier) . The power converter 5 converts between DC power in the battery module 3 and AC power on the electrical grid 7 side of the power converter 5. The power converter 5 also converts (i.e. changes) between a DC voltage across the battery module 3 and an AC voltage on the electrical grid 7 side of the power converter 5 (an AC module voltage) . The power converters 5 are connected in series across an
electrical grid 7, so that a total AC voltage across the electrical grid 7 is a sum of the AC voltages produced by the power converters 5.
FIG. 1 b) shows an alternative topology to that shown in FIG. 1 a) in which the power converters 5 are connected in parallel across the electrical grid 7, so that the AC voltages on the electrical grid 7 side of the power converters 5 are the same as an AC voltage across the electrical grid 7.
FIG. 2 a) shows a topology comprising DC to DC converters in series with cascaded inverters/rectifiers. The apparatus comprises a plurality of modules 9, each of which comprises a power transfer unit in the form of a battery module 3 (of course, in other embodiments of the invention other types of power transfer unit may be present instead of a battery module 3), a DC to DC converter 11 and a power converter 13. Each DC to DC converter 11 converts between a local DC voltage across the respective battery module 3 and a DC voltage on the electrical grid 7 side of the DC to DC converter 11 (a DC module voltage) . The DC module voltage may be greater than or less than the local DC voltage. Each power converter 13 (e.g. an inverter and/or a rectifier) converts between the DC module voltage on the electrical grid 7 side of the DC to DC
converter 11 and an AC module voltage on the electrical grid side of the power converter 13. The power converters 13 are connected in series across an electrical grid 7, so that a total AC voltage across the electrical grid 7 is a sum of the
AC voltages produced by the power converters 13.
FIG. 2 b) shows an alternative topology to that shown in FIG. 2 a) in which the power converters 13 are connected in parallel across the electrical grid 7, so that the AC voltages on the electrical grid 7 side of the power converters 13 are the same as an AC voltage across the electrical grid 7.
FIG. 3 a) shows a topology comprising cascaded DC to DC converters across an inverter/rectifier. The apparatus comprises a plurality of modules 15, each of which comprises a power transfer unit in the form of a battery module 3 (of course, in other embodiments of the invention other types of power transfer unit may be present instead of a battery module 3) and a DC to DC converter 11. Each DC to DC converter 11 converts between a local DC voltage across the respective battery module 3 and a DC voltage on the electrical grid 7 side of the DC to DC converter 11 (a DC module voltage) . The DC to DC converters are connected in series across a power converter 13 (e.g. an inverter and/or a rectifier), which converts between AC power in the electrical grid and DC power in the modules 15, so that a DC voltage across the power converter 13 is equal to a sum of the DC module voltages. A common DC-link current flows between the DC to DC converters 11.
Fig. 3 b) shows an alternative topology to that shown in FIG. l a), in which the DC to DC converters 11 are connected in parallel across a power converter 13 (e.g. an inverter and/or a rectifier) , so that the DC module voltages are equal to a DC voltage across the power converter 13.
There could be two main forms of modular DC to DC converters. Parallel DC to DC converters step up the low module voltage to a high DC link voltage around 400 or 600- 700V depending on single or three phase grid connection. This inherently requires an extremely high step-up ratio converter (around 20-30) to integrate the battery modules to the central DC link. This high step-up ratio reduces the converter efficiency and increases ripple on the battery side.
Moreover, high voltage devices increase the size and cost of the overall converter. For these reasons, in some preferred embodiments of the present invention there is adopted a series connected/cascaded DC to DC structure with smaller DC to DC converters, as illustrated in FIG. 3 a) . This architecture can also use low-rated semiconductor devices and smaller inductors. This provides not only cost, size and integration related advantages but also provide very high efficiency using appropriate semi-conductors (very low Rds (on) and fast switching) such as, CoolMOS, Trench MOSFET, OptiMOS, super- junction MOSFET etc. in each module.
The cascaded converter module can be mainly buck or boost type. Buck type module at the input side inherently demands a large number of modules irrespective of power levels to integrate low voltage battery modules to the DC link along with a large input capacitor in each module. Therefore, in preferred embodiments of the present invention the boost type module is considered at the input side, as this allows for a much smaller number of modules and means that large capacitors can be avoided. However, a problem with modular cascaded boost converters is a lack of fault-tolerance/module bypassing with respect to source. This is an important issue in SLBESS because second life batteries possess variable parameters and limited lifespan.
To handle the reliability issues of SLBESS, in preferred embodiments of the present invention the cascaded DC to DC structure illustrated in FIG. 3 a) is modified using a bypass controlled switch network in parallel with each module capacitor as shown in FIG. 4 . As shown in FIG. 4 , the apparatus has a similar topology to that shown in FIG. 3 a) , in that is comprises a plurality of modules 17 connected in series across an inverter/rectifier 19, which converts between AC power on an electrical grid and DC power in the apparatus. As above, each of the modules 17 comprises a DC to DC
converter 21 and a power transfer unit in the form of a battery module 23 (of course, in other embodiments of the invention other types of power transfer unit may be present instead of a battery module 23) .
The topology illustrated in FIG. 4 employs efficient and fast acting low voltage semiconductors to assist with module bypassing, since these devices have extremely low Rds (on) , which do not incur significant losses and are low-cost switches. Moreover, they tend to operate with low junction temperatures, which makes them less prone to failure compared to traditional switches. Due to the controlled nature of these switches, faulty modules can be taken out and healthy modules could be inserted during the operation. The switches Si and Sii act in pulse-width modulation (PW ) mode when a battery module 23 is charging or discharging which controls the power flow and the switches ΤΊ and T operate in bypass mode (Ti is ON and Tu is OFF) . Under abnormal condition such as overcharge, deep-discharge or high impedance failure mode, a controller (not shown) turns T ON and ΊΊ OFF, which in turn bypasses the ith module. The topology shown in FIG. 4 has been designed in this fashion such that switches Ti and T£i may be used in conjunction with an inductor in series with the DC to DC converters and the inverter/rectifier to operate the topology in buck or boost/buck mode.
Preferably, there is some redundancy in the modular DC side because of the reliability of second life batteries, 'k- out-of-n' (V k < n) redundancy is used in some embodiments of the present invention. Balancing may also be used in each module between the similar cells to equally utilise them.
Much of the following description assumes that the topology shown in FIG. 4 is used. However, the present invention is not limited to the topology shown in FIG. 4 and instead includes, among others, the other topologies
illustrated in FIGS. 1 to 3. Some or all of the following description also applies to those other topologies, where compatible.
SUB-MODULE CHARACTERISATION
Prior to operation of the apparatus, an initial
characterisation of the power transfer units (i.e. generation or energy storage device) is needed to provide baseline or nominal values from which degraded performance may be calculated. Nominal design values could be used with new devices. However, an exception is ex-transportation batteries, where after getting these batteries from different vehicles, their past history will most likely be unknown.
Information available from initial strip down and preliminary testing/characterization may include: a) battery sub module type/chemistry, b) battery sub module capacity, c) battery sub module nominal open circuit voltage d) battery sub module lumped internal impedance or possible state-of-health (SOH) and e) initial battery sub module SOC. Additional information that it may be advantageous to know about the battery may be BMS interface variables and protocols and/or
temperature/voltage limits with other safety dependent information. Each sub-module of the converter may contain similar batteries (in terms of capacity, nominal voltage etc.) with proper balancing among them.
POWER SHARING STRATEGY FOR SLBESS
In the apparatus of the present invention, a distribution of power conveyed between the electrical grid and the power transfer units is dynamically adjusted based on the operation states of the power transfer units. The following section describes power sharing strategies which may be used in some embodiments of the present invention to distribute the total power among the different power transfer units (e.g. among battery sub-modules) . The power sharing strategy may be based on the topology in FIG. 4, where heterogeneous modules with different types, capacities, SOC, and voltage have been included. Again, battery cells are used as an example of the power transfer units, but it is to be understood that the present invention equally includes other types of energy storage units and/or types of energy generation means.
There are many ways the grid side power/total power can be distributed among the different modules, these include, a) sub module voltage (Vbatt,i), b) sub module capacity (Qi) , c) sub module impedance (Zi) or state-of-health (SOH) , d) initial sub module state-of-charge (SOC) (SOCi) or e) any possible combination of the above (15 different combinations) . Many or all of the (battery) parameters under investigation including initial SOC, voltage, SOH (or Z) and Q can be related to the (battery) current through equations/lookup tables. Therefore, it may be appropriate to distribute the total power according to the individual module current ratio. Therefore, a current sharing strategy is adopted in preferred embodiments of the present invention.
Preferably, none of the modules should be bypassed unless under a fault condition (or if they are power generation sources when the flow of power is from the grid to the DC side) , so that each module is undertaking a proportionate share of the contribution relative to each other. One method of ensuring this is to control the sub-modules so that they would aim to be fully charged or fully discharged at the same time for a particular time of grid support. Therefore, in some embodiments of the present invention the charging or discharging of the sub-modules (e.g. battery units) is controlled so that all of the sub-modules are fully charged or fully discharged at the same time.
Various different possible control scenarios for carrying out a current sharing strategy (based purely on battery energy storage, by way of example) will now be discussed. The control scenarios discussed below can also be used with other types of energy storage units and/or generation means.
Control scenario 1:
In this scenario, the sub-module capacities are assumed to be different, but the sub-module impedances, SOCs and voltage ranges are assumed to be identical.
It is assumed that a battery is to be discharged to a particular minimum SOC (SOCmin) for Γ2 time from an initial SOC0 and to be charged up to a particular SOC {SOCmax) for T2 time. Equations (1) and (2) provide the relation between capacity and SOC during discharging and charging respectively for a constant current. Re-arranging these to give equations (3) and (4) provides the required expressions of capacity {Q) during discharging and charging. This strategy allows the higher capacity sub-module to provide the higher share of the current during both charging and discharging.
SOCnax = SOC0 + ^ l f^ for charging {2 )
G =™rbattTslnr ' £'^« * <? ^charging (3 )
Q = soc baU-soc · α Q char9ing ( )
Control scenario 2:
In this scenario, the nominal voltage and the voltage ranges (i.e. the voltage range between Vmax and Vrain of the sub- module) of the sub-modules are assumed to be different, but the sub-module SOCs, impedances and capacities are assumed to be identical. In this case, where the nominal voltages are different, the charging and discharging limits (Vmaii and V^) could be used to optimally utilise the modules. Assuming the battery voltage varies linearly from fully charged to fully discharged, {VbattL - Vmin) or {Vmax - Vbact) provides the
information of how far a module can be discharged and charged respectively before reaching its minimum or maximum point with respect to individual modules. Therefore, a module with a higher {Vbatt - Vmln) can be discharged at a higher rate compared with a module with lower (Vbatt - Vmln) , if every other parameter was to remain the same. Similar logic is applicable during charging, depending on {Vmax - Vbatt) . Therefore, the current sharing can be written according to equations (5) and (6) for discharging and charging respectively for individual modules. lbatt,l « (Ybatt.i - Vmin discharging (5) lbatt,i K (Υπιαχ,ι - Vbatcd charging (6)
Control Scenario 3:
In this scenario, the sub-module impedances are assumed to be different but the sub-module voltages, SOCs and capacities are assumed to be the same.
Battery characteristics can be represented as a variety of models, some of which are shown in FIG. 5. In this scenario, FIG. 5c is assumed sufficiently accurate for the purposes of this study. It is to be noted that the current command is always inversely proportional to the module impedance irrespective of charging and discharging for a fixed terminal voltage and OCV as shown in equation (7) . However, the sharing preferably should not be in exact ratio of impedance because different module chemistries could have different nominal or steady state impedances. Therefore, the current sharing is preferably calculated using per-unit impedance . This ratio also acts as an indication of
Znom
module state-of-health (SOH) because it indicates the
percentage change in internal impedance.
1 z
Vdc,i = Vbat i ^battX^i > i-batt.i *· ^αη,ί K "l™'1 C7)
Control Scenario 4:
In this scenario, the sub-module SOCs are assumed to be different but the sub-module voltages, capacities and
impedances are assumed to be the same.
The desired charging and discharging command can be found from the coulomb counting equation as shown in equations (8) and (9) . The SOCmln generally lies around 20% and SOCmax is dependent on the degree of capacity fade, but is considered to be 80% of the remaining capacity. However, it is sufficiently accurate to assume that SOCmln and SOCmax are 0 and 100% respectively for all cases as long as the SOC value is limited appropriately .
S0Cmin = S0C0 - during dischoring
SOCmin « 0 → SOC0 = i^ , ibatt oc SOC0
SOCmax = SOC0 + during charging
SOCmax * 1 → ibatt oc (1 - SOCo)
This sharing strategy requires information of initial module SOC. Equations (10) and (11) can be modified using OCV in equation (3) if a linear relationship is assumed between SOC and OCV. In the case where OCV and SOC relationship is not linear, the sharing would be according to the SOC which can be found from an SOC-OCV look-up table.
Figure imgf000033_0001
Generalised Sharing Strategy In the case, where two or more sub-module parameters differ between sub-modules, e.g. the sub-modules comprise batteries that differ in more than one of SOC, SOH, voltage range and capacity, the sharing strategies discussed above which assume differences in only a single sub-module parameter may not be appropriate. Therefore, in some embodiments of the present invention a generalized strategy may be generated by combining each of the cases above to give the relationships of equation (12) for discharging and equation (13) for charging. „ QZnom (Ybatt-Vmin) (PCVo-OCVmin) , ,„ ,
Ibatt Z (OCVmax-OCVmin) )
, o- QZnom (Vmax-VbattKOCVmax-OCVmin) .
Therefore, for n' number of modules the current sharing ratio is : i = ibatti = fQr discharging i = = itat^ fgr cft .
ωι ω2 ωη a a ωιΛ ω2 2 α>„,„
Figure imgf000034_0001
It can be seen the generalised state-weightage depends on all the battery parameters. Therefore, all the parameters may have to be measured or estimated to properly distribute the power among the modules. Only Vbatt can be directly measured. However, capacity, impedance etc. can be estimated online, e.g. in the manner discussed below.
ONLINE IMPEDANCE AND CAPACITY ESTIMATION
In some embodiments of the present invention the proposed sharing strategy may depend on the terminal voltages, the capacities, the impedances and the OCVs . The following section outlines methods of estimating online (i.e. during operation of the energy transfer apparatus) impedance (¾) , open circuit voltage (Vbattji) and capacity (Qs) .
Impedance estimation
Impedance is a very important parameter which determines state-of-health (SOH) of a battery [27] -[28] . In second life applications this parameter tends to go high from the nominal value. Some embodiments of the present invention may use high-frequency ripple based impedance estimation to estimate the online impedance. The idea behind this method is to use the high-frequency inductor ripple current of the DC to DC converter, which corresponds to the high-frequency ripple of the respective battery terminal voltage, to calculate the internal impedance of the battery. The ripple current and ripple voltage are illustrated schematically in FIG. 6 for a 24V lithium titanate HEV battery when an external 0.03 Ohms resistor was connected in series with the battery whose nominal impedance is about 0.02 Ohms. Since the switching frequency is 10' s of kHz, it can be assumed that the SOC does not change significantly during each small switching interval. The OCV is also considered constant within that interval.
Therefore, two different equations can be written, a) at t=0 and b) at t=dTs as shown in equations (15) and (16) .
The ripple current and ripple voltage can be extracted from the current and the voltage using a LPF (low pass filter) with cut-off frequency l/10h of the switching frequency, as illustrated in FIG. 6. This is because the IZ voltage drop occurs at every instant and the battery terminal voltage reaches a local minimum point when the current reaches a local maximum during discharging, and vice-versa. After extracting the ripple parts of the voltage and the current, the
magnitudes of both ripple parts are calculated. The online impedance can subsequently be determined from the magnitudes of the ripple parts according to equation (17) . max = OCV ± ibattmax Z at t = 0 (15)
Figure imgf000035_0001
(Vbatt;
Z = max -vbattmln) \ _ I (Afftatt)
(17) max The experimental setup illustrated in FIG. 6 may not be sufficiently accurate to accurately measure the phase, and therefore to split R and X (real and reactive components of impedance) for the battery into their component parts.
However, this method is able to detect the online variation of impedance because the battery voltage ripple always varies when the impedance varies, since the current ripple is fixed by the switching frequency and by the inductor value. This is a valid approach when the impedance is used as a comparative value (as in this case) . This method assumes that the temperature of the cells is approximately constant across each sub module.
An experimental trace showing the results of impedance estimation using this method is illustrated in FIG. 7. FIG. 7 shows the variation of the impedance determined using such a ripple-based strategy for a 24V lithium titanate HEV battery when an external 0.03 Ohms resistor was connected in series with the battery, whose nominal impedance is about 0.02 Ohms. It can be seen from FIG. 7 that the voltage ripple shows a significant change when the impedance varies. However, there are some assumptions in using this method:
i. The switching frequency is set to a value high enough to ensure good dynamics of the DC to DC converter but at a value sufficient to get accurate results through the sensors. High switching frequency >50 kHz is difficult to measure because of the effects of noise, ii. The impedance of the battery changes with frequency.
It is assumed that the measurements at the switching frequency are sufficiently close to DC to give an indication of the impedance and therefore SOH of the battery.
Capacity (Q) estimation Battery capacity can also vary online. This is
particularly important in second life battery applications because the capacity may have already degraded to 70-80% compared to the list value. The control system needs to track the capacity to assist with power sharing. A method similar to that described in [29] may be used, in which the capacity estimation can be done using the following steps:
i) Obtain the internal impedance using the ripple-based method.
ii) Find the corresponding OCV using OCV = Vbatt + ibattZ when discharging and OCV = Vbatt - ibattZ for charging.
iii) Find the SOC using an OCV-SOC look-up table.
iv) After obtaining the SOC from the look-up table, the capacity can be found from sample-by-sample using equation (18) where ti and t2 are the two sample instances. ll ib * ctdt
Q = jsocCt -socctz)] :i8)
DISTRIBUTED CONTROL STRUCTURE
The control of modular energy storage systems has been previously undertaken using a single type of battery with equal sub-modules in terms of voltage, capacity etc.
Equalization and balancing control has been considered to be relevant to BESS, for example to balance the SOC among the modules using a modular DC to AC converter [26] . However, in second life applications independent module control
(distributed control) according to battery parameter is an important function and has not been dealt with before. In some embodiments of the present invention, the main objective of the distributed control is to control each module according to instantaneous battery state so that each module is utilised optimally.
Earlier work previously untaken relates to distributed MPPT control of PV systems [24], which is solely based on different radiation conditions. However, the power sharing strategy and control system for a PV system is unidirectional from PV module to grid and the requirements are significantly different from a BESS.
The following describes a bidirectional distributed control system using a cascaded DC to DC converter. The control strategy assumes that the overall battery side voltage is less than the desired DC link voltage {∑Vbatti≤ V^c) , which is the most likely the case in an energy storage system.
Modelling of cascaded DC to DC converter
The dynamic equations of the converter can be written module-by-module as in equations (19) -(21) for the topology illustrated in FIG. 4, assuming that ± is on and is OFF
(i.e. that the module is operating in boost mode and that the module is not bypassed because of a fault) . Since J-out-of-n redundancy has been assumed (2-out-of-3 in the present case) the converter still operates if one or more modules fails/is bypassed. The overall control structure forces extra stress on the other modules when one module is bypassed. Each converter module may be designed considering such margins.
Module current dynamics : L?i!2 +(1_di)Vdci = ivi = l...n (19) Module voltage dynamics:
Figure imgf000038_0001
Total dc-link voltage:
Vdc = VdCil + VdCi2 + -VdCin (21)
The power balance equations of the modules are given in equation (22), where lr η2 etc. are the efficiencies of the modules. It should be noted that these efficiencies are very- high ( « 1 ) for this converter because of the OSFET
characteristics. It is also to be noted that for a constant Vdc and grid power, Idc also remains constant. Therefore, the module voltages, VdCtl become proportional to the module power
( ¾att,i lbatt
Vdc dc = Vi Vbatt,iibatt,i V £ = 1... n (22 )
The overall power balance equation is given by equation (23 ) which relates the grid voltage and central dc-bus voltage where η1ην stands for the inverter efficiency.
Vdc c = invVs = P ( 23 ) Small Signal Modelling
Since the battery terminal voltage, battery current, module DC link voltage etc. may vary during the control operation, it may be useful to investigate the small signal perturbation and the transfer function in order to accurately predict the dynamics of the control system. Equations ( 24 ) and ( 25 ) provide small signal equations for the ith module, where D± is the steady state duty ratio and dx is instantaneous duty ratio.
L 1 - D )V , - = Vbattii (24 )
Figure imgf000039_0001
There are two state variables such as i and VdCri per module, one control input di and power input Vbatt/1. The state- space equation of ih module therefore becomes the following:
Figure imgf000039_0002
where, X (26)
Figure imgf000040_0001
The rank of the controllability matrix is 2. Therefore, each converter module is fully controllable. The transfer functions of interest are: a) tft°flL ° and b) Vd^ . These transfer functions can be derived from the state-space equation as shown in equations (27) and (28) . These
expressions may help to design the current and voltage controller for a particular module.
Figure imgf000040_0002
Distributed Control Structure
The objectives of the distributed control system in some embodiments of the present invention may be: a) control of the total DC link voltage to a constant value or within a
predetermined range for uninterrupted inverter operation, and b) distribution of the total power/current among the modules according to battery instantaneous state-weightage as
described above. In some embodiments of the present
invention, all the half-bridges (Sif Sa) are controlled using two cascaded control loops, namely an outer module DC link capacitor voltage loop and a fast inner inductor current loop.
This may ensure stability of each module and consequently stability of the entire converter.
In some embodiments of the present invention, the control system controls the module capacitors to different voltage levels to distribute the total power among the modules (because the same DC link current (Idc) flows through all the cascaded modules) . Therefore, the control structure
calculates the different module voltage references (Vdc, , Vdc,2 * - Vdcr*) and controls each module DC capacitor voltage to be the same as the respective DC reference voltage. These voltage references may be dependent on instantaneous battery state (SOC/SOH) . These may be calculated using the power balance equations (22) and the weighting functions which were explained above (see equation (14) ) . It is to be noted that for the same Vbattrl all the currents become in the ratio of module DC link voltages, but in the case where Vbatt,i are different the currents get shared according to (22) . FIG. 8 shows a general form of voltage based distributed control. In the arrangement illustrated in FIG. 8 a notch filter is used to remove the 100Hz ripple on the DC bus voltage measurement (twice grid frequency) . This is then subtracted from the reference voltage and rate limited (to ensure no instantaneous changes) to the energy storage device. The outer DC link voltage loop controls output voltage while the inner loop makes sure battery current remains within stable limits, in the case of a fault on one or more battery cells.
The required module voltage references (VdC 1 *, VdCr2" ... Vdcn*) are generated to allow the control system to share the current according to the desired ratio. This reference generation is based on equations (12) and (13) as well as on the power balance equations as shown in equations (29) and (30) for discharging and for charging respectively. ibatu K wi → vdc,i <*■ vbatt,iWi for discharging (29)
i-bata <* w → Vd*c,i * VbatuWu for charging (30)
FIG. 9 shows how the module voltage references {VdCrl*, Vdc,2 '.··¾η*) are modulated as a function of battery parameters. In more detail, FIG. 9 shows the basic control structure from FIG. 8 including information on how the module DC voltage Vdc,i reference is generated from the impedance, capacity, SOC and battery voltage where equations (29) and (30) are represented by the function "f". The module voltage references continue to vary because the state-weightage varies when some of the battery parameters change. This makes it an adaptive
strategy. This control is very similar to droop control, but is a function of battery voltage, impedance, SOC and capacity. Since this control strategy may depend on all the battery parameters, the sudden/slow variation of any parameter can be dealt with quickly.
The module voltage (Vdoi) may exceed the maximum switch rating during the operation and can cause failure of the module. This situation may only occur under poor design of the power electronic topology and refers to the rating of the inverter switches. To avoid this, the control system needs to know the maximum duty ratio of a module so that such a situation can be avoided. Equation (31) provides guidance on the module duty ratio (dj_) which is dependent on maximum switch stress (Vswmax) and which is dependent on the redundancy present in the converter. It can be seen that the lower voltage module has wider range of duty ratio and vice-versa.
Figure imgf000042_0001
swmax
Module Controller Design
The controller design may be performed in two-stages: a) outer voltage loop design, and b) inner current loop design. FIG. 10 a) shows the voltage control loop assuming inner current loop delay (Td) is around 4-times the sample time (Ts) . The open loop transfer function GHv(s) may be written as in equation (32) to help design suitable control parameters. It can be noticed that GHv(s) is dependent on all the battery parameters such as voltage, capacity, impedance etc. The control parameter can be designed using a symmetric optimum method keeping the same PM for all the modules ( 'a' depends on the desired PM) . ™»( S ) = ^ )(^)(¾ )G½
The current loop can be formed as shown in FIG. 10 b) . Proportional control (i.e. the transfer function for the internal current loop) has been chosen in the current loop to improve the speed of response and the stability. Kc can be set from the desired bandwidth. In this case, it is taken to be 2 kHz for all the modules. The transfer function of equation (27) has been approximated at high frequency to design Kc as shown in equation (33) . ru f„\ — is 1 'ftatt,i(s) _ v 1 vdc,l
(BW)LG
VdcX The frequency response plot in discharging mode is shown in FIG. 11. It can be seen that the modules have the same frequency response and phase margin (PM) using different control parameters for different modules. Moreover, these parameters change dynamically when a mode changes (charging to discharging) to keep the phase margin the same.
Transient response
The response time/rise time (time to reach to 90% of the steady-state value from the 10%) is important to determine the speed of response of the control system. It is directed by the closed loop bandwidth of the system. The closed loop bandwidth can be taken as being equal to/proportional to gain crossover frequency if the system frequency response curve cuts the OdB axis at - 20dB/decade (as shown in FIG. 11) . In most stable systems, the product of rise time and bandwidth is constant and is given by equation (34) .
In this case, the response time of each DC to DC converter depends on the closed loop bandwidth of the outer module voltage control loop. However, the overall response time of the energy storage system is approximately the sum of the response time of the modular DC to DC converter and the inverter on the grid side. However, the response time of inverter is much faster than that of the DC to DC converters (in the order of 100' s of \is) . Therefore, the response time of the energy storage system may be mainly governed by the response time of the DC side.
The gain crossover frequency can be found by solving equation (35) for this case. Due to the presence of a higher order equation, the gain crossover frequency is found using a numerical program (written in MATLAB) and directly entering the measured battery values and DC link voltages found in the experimental setup. After finding the closed loop bandwidth, the speed of response or rise time can be found from equation (34) .
0.35 2.2
Trd,i— (34) few Vgci
Figure imgf000044_0001
In some embodiments of the present invention an aim is to keep the phase margin or gain crossover frequency the same to maintain the stability. Therefore, the same rise
time/response of all the modules is expected.
Overall control structure for SLBESS An example of the overall control structure is shown in FIG. 12. The grid side control depends on the type of application of the energy storage system, e.g. voltage control or frequency control. The output of the voltage or frequency controller provides the reference for the inner current loop for the inverter. It is noted that the battery side control is different in charging and in discharging mode. The mode selection depends on the sign of the line side current reference isq * as shown in FIG. 12, because this active current reference is positive for discharging and negative for charging. In some embodiments of the present invention, the DC link voltage references are dynamically changed through the sign of isq * from charging to discharging and vice-versa. The initial SOC of the individual modules are taken as the SOC just before switching the modes.
EXPERIMENTAL VALIDATION
In order to verify the proposed control and power sharing strategy, experiments were performed on a hybrid multi-modular prototype in a grid connected condition. The sharing strategy has been validated experimentally on a three module based single phase prototype running into a 100V grid system. Table I shows the different components and their specifications used in the experiments .
Type/Name Rating/specification
LV Trench MOSFET for H-bridge dc- 100V 40Ά - Rds(onl =6mQ, ton dc modules +toff =75ns (FDPF085N10A)
Field Stop IGBT for inverter VcEfsat) - 2.2V, Eon + E0ff =
0.53mJ
Boost inductors of dc-dc modules 1.5mH, 15A, RL = 20mQ
dc-link capacitor in each module 1200pF, 100V
dc-link capacitor for inverter 2200pF, 400V
Line side inductors 3mH, 15A
Line side capacitor 10pF
Switching frequency of the dc-dc 10kHz
modules Switching frequency of inverter 10kHz
PM of the three modules 75°
Voltage loop BW, current loop BW 35 Hz, 2 kHz
Operating central dc-bus voltage 150V
Nominal Grid voltage 100V (peak)
Grid current 7A (peak)
Battery module -1 12V, 16Ah lead acid - Vmax =
14V Vmln = 9.5V, Znom = 0.015Ω
Battery module -2 24V, lOAh lithium titanate - Vmax = 27V Vmin = 20V, Znom - 0.02Ω
Battery module -3 7.2V, 6.5Ah NiMH - Vmax =
8.5V Vmln = 5V, Znom = 0.03Ω
Digital controller Multiple FPGA based OPAL-RT
TABLE I - COMPONENTS AND THEIR SPECIFICATIONS IN EXPERIMENTAL
VALIDATION The following experimental work has been undertaken to show: a) power sharing strategy/distributed control, b) dynamic parameter variation such as, voltage, impedance, and c) module bypassing or converter redundancy. All the experimental values (both transient response and steady state values) have been compared with calculated values for completeness of the study.
Mode - 1 zero to discharging mode (current dynamics) : FIG. 13 shows the experimental results for a distributed control scheme at the moment of connecting the apparatus to the grid. It is noted that the currents in the modules are significantly different. The module currents do not undergo any overshoot when responding to load. The first module produces the highest current while the third module outputs the lowest current among the three, because of the differences in instantaneous state-weightages . It can also be observed that the three currents have different dynamics, because of differences in controller parameters needed to make similar response times for the modules e.g. the first module has the highest rate of rise of current because it has the highest steady-state value among the modules. The measured and calculated steady-state and response times are provided in Table II.
Figure imgf000047_0001
TABLE II - COMPARISON OF CALCULATED AND EXPERIMENTAL RESPONSE ( STEADY-STATE AND TRANSIENT) OF MODULE CURRENTS IN DIFFERENT MODES Mode - 2 zero to charging mode (current dynamics) :
FIG. 14 shows the experimental results when the converter switches to charging mode at the moment of connecting the apparatus to the grid. The first module is charged at a higher current than the remaining modules because the first module has the highest state weightage during the charging mode. It can be seen that the module currents in the charging are less than that of discharging for a fixed grid side current. This is because the magnitudes of the battery voltages are higher in charging mode. Importantly, it can be seen that the three modules undergo different dynamics to keep the response time nearly the same. The calculated values of the steady-state and the transient response times are given against measured values in Table II. Mode - 3 charging to discharging mode (current dynamics) - case - I ;
FIG. 15 shows the experimental results when the converter switches from charging mode to discharging mode. It is important to note that the sharing between the modules is different in charging and in discharging. The first module takes higher current compared to the rest of the modules because of the higher state-weightage . The steady state values are provided in Table II with their corresponding state weightages. It can be noticed that none of the modules takes any overshoot and the response time (switching from one mode to another) remains similar compared with earlier modes.
Mode - 3 charging to discharging mode (current dynamics) - Case 2 - effect of initial SOC: FIG. 16 shows the experimental results of the converter switching from the charging mode to discharging mode. It is important to note that the current sharing is different in this case compared to case - 1. It can be seen that module -
2, which was sharing the second highest current during charging, carries the lowest current in discharging mode and also that module - 3, which was carrying the lowest current during charging mode, dominates over the second module in discharging mode. This is because the initial SOC of the third module was at a very high value before switching to discharging mode, which allows it to share a higher current compared to the module - 2, even though it has the lowest capacity among the units. The numerical comparison is given in Table II.
Mode - 4 discharging to charging mode (current dynamics) :
FIG. 17 shows the experimental results when the converter switches from discharging mode to charging mode. It is to be noted that module - 2 and module - 3 share nearly similar currents while module - 1 shares the highest current.
Mode - 5 charging to discharging mode (voltage dynamics)
FIG. 18 shows the experimental result of module DC link voltage dynamics in the transition from charging to
discharging. This result is presented to show how the module DC link voltages are used to change the power sharing. It is important to note that the steady state voltages are different when a mode switches as described above. It is noticeable that module - 2 has the highest voltage even though the corresponding current is not the highest. This is because the nominal voltage of module - 2 is much higher than those of the remaining modules. The values of the DC link voltages depend on state-weightage and corresponding battery voltage which are shown in equations (29) and (30).
Mode - 6 discharging to charging mode (voltage dynamics)
FIG. 19 shows the experimental results of module DC link voltage dynamics in the transition from discharging to charging. It can be seen from the result that VdcZ remains almost unchanged while Vdcl shows an increase and Vdc3 shows a decrease .
Validation of parameter variation
The validation of online parameter variation is presented for two cases: a) due to battery terminal voltage/SOC
variation and, b) impedance variation.
Case - 1: dynamic voltage variation:
The battery terminal voltage is a slow changing variable, therefore, a result has been presented over a wide time scale (slow acquisition) . Fig 20 shows the experimental results of voltage variation. It can be clearly seen from the current trace that module - 3 decreases gradually with time while current in module -2 increases very slowly. Since module - 3 has the lowest capacity, it shows faster voltage variation among all the modules. This result shows that the proposed strategy utilises the modules according to their instantaneous states, gradually providing the low stress on the weaker modules and the high stress on relatively stronger modules.
Case - 2: impedance variation:
Second life battery impedance may vary as a result of degradation. Battery impedance has been estimated online using the method described above. In order to show how the control system reacts to the impedance variation, a small external impedance of 0.033Ω has been put in series with module - 2 through an ON/OFF contactor. The nominal impedance of the module is found to be around 0.02Ω. Therefore, a 1.5 times increase of impedance has been presented. FIGS. 21 and 22 show experimental results of how the control system responds to an impedance variation during discharging and charging condition respectively. It is clear that module - 2 current decreases when impedance increases both in discharging and charging mode because it indicates a poor condition (for example low SOH) of that module. This strategy keeps a poor or weak battery module operational.
Validation of module bypassing/reliability
FIGS. 23 and 24 show the experimental results of module bypassing during charging and discharging respectively. In this test, module - 3 has been bypassed. It is noticed that module bypassing has no effect on grid side current. This result shows the proposed converter operating with 2-out-of-3 module redundancy.
Discussion of experimental results
It can be seen from the experimental results that the theoretical/calculated values are nearly identical (within ± 10%) both in terms of steady-state values and transient values. The proposed sharing strategy utilises the modules according to their individual optimum points and is also capable of responding according to the battery parameter variation. Since the different battery modules are controlled in different ways and possible redundancy is considered, each converter module is designed in the modular fashion
considering the maximum margin.
A multi-modular cascaded DC to DC converter based topology, a sharing strategy and module based distributed control scheme to integrate a hybrid second life battery storage system to the grid system has been reported, examined and experimentally validated for grid support applications. The proposed control uses an adaptive sharing strategy to distribute the required grid power to the heterogeneous battery modules in an optimum manner depending on their characteristics such as SOC and SOH. The power/current sharing strategy is capable of detecting the battery parameter variation such as internal impedance variation and/or possible battery failure online. The experimental results show the effectiveness of the proposed topology and control scheme. This application is mainly focused on low voltage grid systems. However, a similar control strategy is equally valid in medium/high voltage grid applications. The proposed hardware and control solution can easily be extended to a hybrid solution of new and second life batteries and even to second life batteries with other types of inverter connected energy storage such as fuel cells, or to electrical generation means .
NOMENCLATURE
Module battery voltage of i module
Maximum module voltage of ith module
Minimum module voltage of ich module
Module battery current of ith module
Dc-link capacitor voltage of itn module
Dc-link capacitor voltage reference of module
J-batt,! Small signal value of i module current
Small signal value of ± module dc-link voltage
Total DC-link capacitor voltage
Total DC-link capacitor voltage reference Battery voltage ripple of ith module Battery current ripple of ith module Idc DC-link current
di Instantaneous duty cycle of itn converter module
Di Average duty cycle of ith converter module d-max , i Maximum duty cycle of ith converter module s Module dc-link capacitors
P* Reference Power
L Boost inductance of module
s Line side filter inductance
C Line side filter capacitor
vs RMS grid voltage
Is RMS grid current
Qi Capacity of ith battery module
SOCi State-of-charge of ich battery module OCVi Open circuit voltage of ith battery module ocvmin Minimum open circuit voltage of ith module ocvmax Maximum open circuit voltage of izh module
Zi Impedance of 1th battery module nom, i Nominal Impedance of ith battery module i Charging weightage for 1th module currents i,i Discharging weightage for ith module
currents
Kv Proportional gain for module voltage
controller
Tv Integral gain for module voltage
controller
Td Current controller delay
Ts Sampling time
Proportional gain for module current controller
Gain crossover frequency for ith converter module
2 ,i Rise time for ith converter module
Hi Efficiency of ith converter module few Closed loop bandwidth (in Hz) Rotating frame line side currents
Rotating frame line side voltages
G Converter gain for i th dc-dc module
BW Bandwidth
REFERENCES
[1] Manz, D.; Piwko, R.; Miller, N. , "Look before You Leap:
The Role of Energy Storage in the Grid, " Power and Energy- Magazine, IEEE, vol.10, no. , pp.75, 84, July 2012.
[2] Nourai, A., "Large-scale electricity storage technologies for energy management," Power Engineering Society Summer Meeting, 2002 IEEE, vol.1, no., pp.310, 315 vol.1, 25-25 July 2002.
[3] Velasco de la Fuente, D.; Rodriguez, C.L.T.; Garcera, G.;
Figueres, E.; Gonzalez, R.O., "Photovoltaic Power System With Battery Backup With Grid-Connection and Islanded Operation Capabilities," Industrial Electronics, IEEE Transactions on , vol.60, no.4, pp.1571, 1581, April 2013.
[4] Seul-Ki Kim; Jin-Hong Jeon; Chang-Hee Cho; Jong-Bo Ann;
Sae-Hyuk Kwon, "Dynamic Modeling and Control of a Grid- Connected Hybrid Generation System With Versatile Power Transfer," Industrial Electronics, IEEE Transactions on , vol.55, no.4, pp.1677, 1688, April 2008.
[5] Vazquez, S.; Lukic, S.M.; Galvan, E.; Franquelo, L.G.;
Carrasco, J.M., "Energy Storage Systems for Transport and Grid Applications," Industrial Electronics, IEEE Transactions on , vol.57, no.12, pp .3881 , 3895, Dec. 2010.
[6] Chatzakis, J.; Kalaitzakis, K. ; Voulgaris, N.C.; Manias, S.N., "Designing a new generalized battery management system," Industrial Electronics, IEEE Transactions on , vol.50, no.5, pp.990, 999, Oct. 2003.
[7] Manenti, A.; Abba, A.; Merati, A.; Savaresi,, S.M.;
Geraci, A., "A New BMS Architecture Based on Cell Redundancy," Industrial Electronics, IEEE Transactions on , vol.58, no.9, p .4314 , 4322 , Sept. 2011.
[8] Serban, I.; Marinescu, C. , "A look at the role and main topologies of battery energy storage systems for integration in autonomous microgrids," Optimization of Electrical and Electronic Equipment (OPTIM) , 2010 12th International Conference on , vol., no., pp .1186, 1191 , 20-22 May 2010.
[9] Trintis, I.; Munk-Nielsen, S . ; Teodorescu, R., "Single stage grid converters for battery energy storage," Power Electronics, Machines and Drives (PEMD 2010), 5th IET International Conference on, vol., no., pp.1,6, 19-21 April 2010.
[10] Lee, Yuang-Shung; Ming-Wang Cheng, "Intelligent control battery equalization for series connected lithium-ion battery strings," Industrial Electronics, IEEE Transactions on, vol.52, no.5, pp.1297, 1307, Oct. 2005.
[11] Baughman, A.C.; Ferdowsi, M. , "Double-Tiered Switched- Capacitor Battery Charge Equalization Technique," Industrial Electronics, IEEE Transactions on , vol.55, no.6, pp.2277, 2285, June 2008.
[12] Hong-sun Park; Chong-Eun Kim; Chol-Ho Kim; Gun-Woo Moon;
Joong-Hui Lee, "A Modularized Charge Equalizer for an HEV Lithium-Ion Battery String," Industrial Electronics, IEEE Transactions on , vol.56, no.5, pp.1464, 1476, May 2009.
[13] Hung, S.T.; Hopkins, Douglas C; Mosling, C.R., "Extension of battery life via charge equalization control," Industrial Electronics, IEEE Transactions on, vol.40, no.l, pp.96, 104, Feb 1993.
[14] Available [Online] :
http: //evworld. com/news . cfm?newsid=29192
[15] Viswanathan, V.V.; Kintner-Meyer, M. ; , "Second Use of Transportation Batteries: Maximizing the Value of Batteries for Transportation and Grid Services," Vehicular Technology, IEEE Transactions on , vol.60, no.7, pp .2963-2970 , Sept. 2011. [16] Arup and Cenex, "Investigation into the Scope for the Transport Sector to Switch to Electric Vehicles and Plugin Hybrid Vehicles", Department of Transport, October 2008.
[17] Mukherjee, N. ; Strickland, D.; Cross, A.; Hung, W., "Reliability estimation of second life battery system power electronic topologies for grid frequency response applications," Power Electronics, Machines and Drives (PEMD 2012), 6th IET International Conference on , vol., no., pp.1,6, 27-29 March 2012.
[18] Einhorn, M. ; Guertlschmid, .; Blochberger, T.; Kumpusch,
R. ; Permann, R.; Conte, F.V.; Krai, C; Fleig, J., "A Current Equalization Method for Serially Connected Battery Cells Using a Single Power Converter for Each Cell," Vehicular Technology, IEEE Transactions on , vol.60, no.9, pp. 227, 4237, Nov. 2011.
[19] Ball, G.J.; Corey, G.; Norris, B.L., "Government, industry, and utility development and evaluation of a modular utility battery energy storage system," Energy Conversion, IEEE Transactions on , vol.10, no.3, pp.549, 554, Sep 1995.
[20] Bragard, M. ; Soltau, . ; Thomas, S . ; De Doncker, R.W., "The Balance of Renewable Sources and User Demands in Grids: Power Electronics for Modular Battery Energy Storage Systems," Power Electronics, IEEE Transactions on , vol.25, no.12, pp.3049, 3056, Dec. 2010.
[21] Trintis, I.; Munk-Nielsen, S.; Teodorescu, R. ; , "Cascaded H-bridge with bidirectional boost converters for energy storage," Power Electronics and Applications (EPE 2011), Proceedings of the 2011-14th European Conference on , pp.1-9, Aug. 30 2011-Sept. 1 2011.
[22] Walker, G.R.; Sernia, P.C., "Cascaded DC-DC converter connection of photovoltaic modules," Power Electronics, IEEE Transactions on, vol.19, no.4, pp.1130, 1139, July 2004.
[23] Vighetti, S . ; Ferrieux, J.; Lembeye, Y., "Optimization and Design of a Cascaded DC/DC Converter Devoted to Grid- Connected Photovoltaic Systems," Power Electronics, IEEE
Transactions on, vol.27, no.4, pp.2018, 2027, April 2012.
[24] Bratcu, A. I.; Munteanu, I.; Bacha, S.; Picault, D.;
Raison, B. , "Cascaded DC-DC Converter Photovoltaic Systems: Power Optimization Issues," Industrial Electronics, IEEE
Transactions on, vol.58, no.2, pp.403, 411, Feb. 2011.
[25] Vazquez, M.J.V.; Marquez, J.M.A. ; Manzano, F.S., "A
Methodology for Optimizing Stand-Alone PV-System Size Using
Parallel-Connected DC/DC Converters," Industrial Electronics, IEEE Transactions on , vol.55, no.7, pp.2664, 2673, July 2008.
[26] Maharjan, L . ; Inoue, S.; Akagi, H . ; Asakura, J., "State- of-Charge (SOC) -Balancing Control of a Battery Energy
Storage System Based on a Cascade PWM Converter, " Power Electronics, IEEE Transactions on , vol.24, no.6, pp.1628, 1636, June 2009.
[27] Jurgen Remmlinger, Michael Buchholz, Markus Meiler, Peter
Bernreuter, Klaus Dietmayer, "State-of-health monitoring of lithium-ion batteries in electric vehicles by on-board internal resistance estimation", Journal of Power Sources,
Volume 196, Issue 12, 15 June 2011.
[28] Munoz-Condes, P.; Gomez-Parra, M. ; Sancho, C; San
Andres, M.A.G.; Gonzalez-Fernandez, F.J.; Carpio, J.;
Guirado, R., "On Condition Maintenance Based on the Impedance Measurement for Traction Batteries: Development and Industrial Implementation," Industrial Electronics, IEEE
Transactions on , vol.60, no.7, pp .2750 , 2759, July 2013.
[29] Einhorn, M. ; Conte, F.V.; Krai, C; Fleig, J., "A Method for Online Capacity Estimation of Lithium Ion Battery Cells Using the State of Charge and the Transferred Charge,"
Industry Applications, IEEE Transactions on , vol.48, no.2, pp.736, 741, March-April 2012.

Claims

1. Energy transfer apparatus for conveying power to and/or from an electrical grid, the apparatus comprising:
a plurality of power transfer modules, each power transfer module comprising a power transfer unit for storing DC power and/or for supplying DC power;
power conversion means arranged to:
convert between AC power in the electrical grid and DC power in the plurality of power transfer modules; and
convert between a local DC voltage across each power transfer unit and a respective power transfer module voltage on a grid-side of the respective power transfer module; and a controller communicably connected to the power
conversion means and arranged to:
determine an operation state for each power transfer unit; and
dynamically adjust a distribution of power conveyed between the electrical grid and the plurality of power transfer modules based on the operation states for each of the power transfer units.
2. The energy transfer apparatus according to claim 1, wherein the power conversion means comprises a DC to DC converter in each of the plurality of power transfer modules arranged to convert between the local DC voltage across the power transfer unit and a DC power transfer module voltage on a grid-side of the DC to DC converter.
3. The energy transfer apparatus according to claim 2, wherein :
the power conversion means comprises a power converter for converting between AC power in the electrical grid and DC power in the apparatus, and
the plurality of power transfer modules are connected in series with a power transfer module side of the power
converter, so that a total DC voltage across the power transfer module side of the power converter is a sum of the DC power transfer module voltages.
4. The energy transfer apparatus according to claim 3, wherein the controller is arranged to dynamically adjust the distribution of power conveyed between the electrical grid and the plurality of power transfer modules to maintain the total DC voltage across the power transfer module side of the power converter at a constant level or within a predetermined range.
5. The energy transfer apparatus according to claim 2, wherein:
the power conversion means comprises a power converter for converting between AC power in the electrical grid and DC power in the apparatus, and
the plurality of power transfer modules are connected in parallel with a power transfer module side of the power converter, so that the DC power transfer module voltages are the same as a DC voltage across the power transfer module side of the power converter.
6. The energy transfer apparatus according to claim 2, wherein the power conversion means comprises a power converter in each of the plurality of power transfer modules arranged to convert between the DC power transfer module voltage on the grid-side of the DC to DC converter and an AC power transfer module voltage on a grid-side of the power transfer module.
7. The energy transfer apparatus according to claim 1, wherein the power conversion means comprises a power converter in each of the plurality of power transfer modules arranged to convert between the local DC voltage across the power transfer unit and an AC power transfer module voltage on a grid-side of the power transfer module.
8. The energy transfer apparatus according to claim 6 or claim 7, wherein the plurality of power transfer modules are connected in series with the electrical grid, so that a total AC voltage across the electrical grid is a sum of the AC power transfer module voltages.
9. The energy transfer apparatus according to claim 6 or claim 7, wherein the plurality of power transfer modules are connected in parallel with the electrical grid, so that the AC power transfer module voltages are the same as an AC voltage across the electrical grid.
10. The energy transfer apparatus according to any one of claims 2 to 6, wherein each DC to DC converter is
controllable by a respective DC reference voltage, and wherein the controller is arranged to set each respective DC reference voltage based on the operation state for its respective power transfer unit.
11. The energy transfer apparatus according to any one of claims 2 to 6, wherein the DC to DC converters of the plurality of power transfer modules form a cascaded H-bridge DC to DC converter.
12. The energy transfer apparatus according to any one of the previous claims, wherein the controller is arranged to obtain a value representative of each of one or more
characteristic parameters of each power transfer unit in order to determine the operation state for that power transfer unit.
13. The energy transfer apparatus according to claim 12, wherein the controller is arranged to: calculate a weighting factor for each power transfer unit from the obtained values representative of the characteristic parameters of that power transfer unit, and
apportion a DC current corresponding to DC power in the apparatus according to the weighting factor for each power transfer unit.
14. The energy transfer apparatus according to claim 12 or claim 13, wherein the power transfer unit in one or more of the plurality of power transfer modules comprises one or more battery cells, and the characteristic parameters of the one or more battery cells include one or more of:
a voltage of the one or more battery cells,
a capacity of the one or more battery cells,
a state-of-charge (SOC) of the one or more battery cells, a state-of-health (SOH) of the one or more battery cells, an impedance of the one or more battery cells, and a temperature of the one or more battery cells.
15. The energy transfer apparatus according to any one of the previous claims, wherein each of the plurality of power transfer modules includes a bypass switch controllable by the controller to disconnect the power transfer module so that power is not conveyed between the power transfer unit of that power transfer module and the electrical grid.
16. The energy transfer apparatus according to any one of the previous claims, wherein the controller is switchable between a first state where power is supplied to the
electrical grid from the power transfer units and a second state where power is received from the electrical grid and stored in the power transfer units.
17. The energy transfer apparatus according to claim 2, wherein the DC to DC converters are selectively operable in any one of a boost mode, a buck mode, and a buck-boost mode.
18. The energy transfer apparatus according to any one of the previous claims, wherein the power transfer unit in one or more of the plurality of power transfer modules comprises one or more battery cells.
19. The energy transfer apparatus according to claim 18, wherein the one or more battery cells include one or more transportation batteries.
20. A method of distributing power within an energy transfer apparatus for conveying power to and/or from an electrical grid, the method comprising:
determining an operation state for a power transfer unit in each of a plurality of power transfer modules of the energy transfer apparatus, the power transfer units being arranged to store DC power and/or to supply DC power; and
dynamically apportioning the power conveyed between the electrical grid and the plurality of power transfer modules based on the operation states for each of the power transfer units .
21. The method according to claim 20, wherein the method comprises converting between a local DC voltage across the power transfer unit and a respective DC power transfer module voltage in each of the plurality of power transfer modules.
22. The method according to claim 21, wherein:
the plurality of power transfer modules are connected in series with a power transfer module side of a power converter for converting between AC power in the electrical grid and DC power in the apparatus, so that a total DC voltage across the power transfer module side of the power converter is a sum of the DC power transfer module voltages; and
dynamically apportioning the power conveyed between the electrical grid and the plurality of power transfer modules maintains the total DC voltage across the power transfer module side of the power converter at a constant level or within a predetermined range.
23. The method according to any one of claims 20 to 22 for conveying power from the energy transfer apparatus to the electrical grid, wherein dynamically apportioning the power conveyed between the electrical grid and the plurality of power transfer modules includes setting the discharge rates of the power transfer units in a manner whereby they would all fully discharge simultaneously.
24. The method according to any one of claims 20 to 23, wherein determining an operation state for each power transfer unit includes obtaining a value representative of each of one or more characteristic parameters of that power transfer unit.
25. The method according to claim 24, wherein one or more of the power transfer units comprise one or more battery cells, and the characteristic parameters of the one or more battery cells include one or more of:
a voltage of the one or more battery cells,
a capacity of the one or more battery cells,
a state-of-charge (SOC) of the one or more battery cells, a state-of-health (SOH) of the one or more battery cells, an impedance of the one or more battery cells, and a temperature of the one or more battery cells.
26. The method according to any one of claims 20 to 25, including calculating a DC current for each power transfer unit based on the operation state of the power transfer unit.
27. The method according to any one of claims 20 to 26, wherein dynamically apportioning the power conveyed between the electrical grid and the plurality of power transfer modules includes :
setting a reference voltage for a DC to DC converter in each of the plurality of power transfer modules, whereby the DC to DC converter converts between the local DC voltage across the power transfer unit and a respective DC power transfer module voltage on a grid-side of the DC to DC converter .
28. The method according to any one of claims 20 to 27, wherein determining an operation state for each power transfer unit is performed rapidly enough for dynamic apportioning to be performed in real time.
29. The method according to any one of claims 20 to 27 wherein the method comprises repeatedly, and optionally periodically, determining an operation state for each power transfer unit.
PCT/GB2014/051215 2013-05-07 2014-04-17 Energy transfer apparatus and distribution control method therefor WO2014181081A1 (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
GBGB1308189.8A GB201308189D0 (en) 2013-05-07 2013-05-07 Energy transfer apparatus and distribution control method therefor
GB1308189.8 2013-05-07
GB1315787.0 2013-09-05
GBGB1315787.0A GB201315787D0 (en) 2013-05-07 2013-09-05 Energy transfer apparatus and distribution control method therefor

Publications (1)

Publication Number Publication Date
WO2014181081A1 true WO2014181081A1 (en) 2014-11-13

Family

ID=48627393

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/GB2014/051215 WO2014181081A1 (en) 2013-05-07 2014-04-17 Energy transfer apparatus and distribution control method therefor

Country Status (2)

Country Link
GB (2) GB201308189D0 (en)
WO (1) WO2014181081A1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018037087A1 (en) * 2016-08-24 2018-03-01 General Electric Technology Gmbh Primary power grid frequency response characterization using phasor measurement unit data
CN109193776A (en) * 2018-10-08 2019-01-11 江苏佳讯纳通能源技术有限公司 A kind of power distribution method suitable for echelon battery energy storage
WO2019154519A1 (en) * 2018-02-12 2019-08-15 Siemens Aktiengesellschaft Operating an energy store

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012205480A (en) * 2011-03-28 2012-10-22 Mitsubishi Heavy Ind Ltd Power storage system
CN102842964A (en) * 2012-09-25 2012-12-26 上海电气集团股份有限公司 Energy storage access system

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012205480A (en) * 2011-03-28 2012-10-22 Mitsubishi Heavy Ind Ltd Power storage system
CN102842964A (en) * 2012-09-25 2012-12-26 上海电气集团股份有限公司 Energy storage access system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LAXMAN MAHARJAN ET AL: "State-of-Charge (SOC)-Balancing Control of a Battery Energy Storage System Based on a Cascade PWM Converter", IEEE TRANSACTIONS ON POWER ELECTRONICS, INSTITUTE OF ELECTRICAL AND ELECTRONICS ENGINEERS, USA, vol. 24, no. 6, 1 June 2009 (2009-06-01), pages 1628 - 1636, XP011262262, ISSN: 0885-8993 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018037087A1 (en) * 2016-08-24 2018-03-01 General Electric Technology Gmbh Primary power grid frequency response characterization using phasor measurement unit data
US10074983B2 (en) 2016-08-24 2018-09-11 General Electric Technology Gmbh Primary power grid frequency response characterization using phasor measurement unit data
US10476270B2 (en) 2016-08-24 2019-11-12 General Electric Technology Gmbh Primary power grid frequency response characterization using phasor measurement unit data
WO2019154519A1 (en) * 2018-02-12 2019-08-15 Siemens Aktiengesellschaft Operating an energy store
CN109193776A (en) * 2018-10-08 2019-01-11 江苏佳讯纳通能源技术有限公司 A kind of power distribution method suitable for echelon battery energy storage

Also Published As

Publication number Publication date
GB201315787D0 (en) 2013-10-23
GB201308189D0 (en) 2013-06-12

Similar Documents

Publication Publication Date Title
Pires et al. Power converter interfaces for electrochemical energy storage systems–A review
Akar et al. An energy management strategy for a concept battery/ultracapacitor electric vehicle with improved battery life
US20190245237A1 (en) Rechargeable multi-cell battery
US10347952B2 (en) Battery system
US10110006B2 (en) Management of battery capacity
Jin et al. Implementation of hierarchical control in DC microgrids
Rothgang et al. Modular battery design for reliable, flexible and multi-technology energy storage systems
Lahyani et al. Battery/supercapacitors combination in uninterruptible power supply (UPS)
Shang et al. A cell-to-cell battery equalizer with zero-current switching and zero-voltage gap based on quasi-resonant LC converter and boost converter
Pinto et al. Bidirectional battery charger with grid-to-vehicle, vehicle-to-grid and vehicle-to-home technologies
KR101369692B1 (en) Energy storage system and controlling method of the same
Zhang et al. A distributed architecture based on microbank modules with self-reconfiguration control to improve the energy efficiency in the battery energy storage system
US9300016B2 (en) Battery system and energy storage system
Maharjan et al. State-of-charge (SOC)-balancing control of a battery energy storage system based on a cascade PWM converter
Li et al. SOH balancing control method for the MMC battery energy storage system
KR101084215B1 (en) Energy storage system and method for controlling thereof
US8248033B2 (en) Secondary battery temperature-increasing control apparatus and vehicle including the same, and secondary battery temperature-increasing control method
Arancibia et al. Modeling of an electric vehicle charging station for fast DC charging
Speltino et al. Cell equalization in battery stacks through state of charge estimation polling
JP2015052590A (en) Battery pack, device having battery pack, and management method for battery pack
JP5676767B2 (en) Power conversion system for energy storage system and control method thereof
Tashakor et al. A bidirectional battery charger with modular integrated charge equalization circuit
US8742722B2 (en) Dynamic power management system and method
Ge et al. State-of-charge balancing control for a battery-energy-stored quasi-Z-source cascaded-multilevel-inverter-based photovoltaic power system
Zheng et al. A hybrid cascaded multilevel converter for battery energy management applied in electric vehicles

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14719818

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14719818

Country of ref document: EP

Kind code of ref document: A1