ELECTRICAL ENERGY STORAGE DEVICE

Abstract

An electrical energy storage device includes a plurality of energy cell slots for receiving energy cells; and a controller; wherein the controller is arranged to estimate a characteristic of a cell in each slot; and wherein the controller is arranged to apply charge and discharge currents to each cell slot dependent upon at least one estimated characteristic currently associated with that slot. The controller may be a single controller that controls all slots or it may be implemented as multiple controllers each controlling more than one cell slot or a controller for each slot. The characteristic may be one or more of: a power capability, a storage capacity, a cell impedance, an energy cell type and an energy cell chemistry.

Claims

1. An electrical energy storage device comprising: a plurality of energy cell slots for receiving energy cells; and a controller; wherein the controller is arranged to estimate a characteristic of a cell in each slot; and wherein the controller is arranged to apply charge and discharge currents to each cell slot dependent upon at least one estimated characteristic currently associated with that slot.

2. The electrical energy storage device as claimed in claim 1, wherein the at least one characteristic comprises one or more of: a power capability, a storage capacity, a cell impedance, an energy cell type and an energy cell chemistry.

3. The electrical energy storage device as claimed in claim 2, wherein the at least one characteristic comprises a storage capacity and wherein the storage capacity estimate is based on a coulomb counting method.

4. The electrical energy storage device as claimed in claim 1, further comprising a bi-directional DC to DC converter for each cell slot, said DC to DC converters being controlled by said controller.

5. The electrical energy storage device as claimed in claim 1, wherein the controller is arranged to estimate the characteristic associated with each slot periodically.

6. The electrical energy storage device as claimed in claim 5, wherein the controller is arranged to estimate the characteristic associated with each slot after every charge phase.

7. The electrical energy storage device as claimed in claim 5, wherein the controller is arranged to estimate the characteristic associated with each slot after every discharge phase.

8. The electrical energy storage device as claimed in claim 1, wherein the controller is arranged to perform a full charge/discharge cycle upon startup.

9. The electrical energy storage device as claimed in claim 1, wherein the controller implements a function which depends on the at least one characteristic estimate associated with each slot to determine the amount of charge/discharge current to sink/source to said slot.

10. The electrical energy storage device as claimed in claim 9, wherein the controller stores a history of characteristic estimates for each slot and wherein said function depends on said history.

11. The electrical energy storage device as claimed in claim 9, wherein the controller updates the function every time the at least one characteristic is estimated.

12. The electrical energy storage device as claimed in claim 9, wherein the at least one characteristic comprises capacity and wherein the function scales the charge/discharge current linearly with the estimated capacity.

13. The electrical energy storage device as claimed in claim 1, wherein a plurality of said slots each have an energy cell connected thereto.

14. The electrical energy storage device as claimed in claim 13, wherein said energy cells comprise rechargeable battery cells.

15. The electrical energy storage device as claimed in claim 14, wherein said energy cells comprise supercapacitors.

16. The electrical energy storage device as claimed in claim 13, wherein the plurality of cells comprises cells of different storage capacities.

17. The electrical energy storage device as claimed in claim 13, wherein the plurality of cells comprises cells of different cell chemistries.

18. The electrical energy storage device as claimed in claim 13, wherein the plurality of cells comprises cells of different power capabilities.

19. The electrical energy storage device as claimed in claim 1, wherein the controller is arranged to identify, based on said at least one estimated characteristic that a cell is no longer suitable for use in the system.

20. A method of storing and using electrical energy in a plurality of energy cells comprising: estimating at least one characteristic of each of said plurality of cells; charging said plurality of cells in a charging step that comprises: simultaneously charging said plurality of cells by providing a charge current to each cell that is dependent upon at least one estimated characteristic of said cell; and discharging said plurality of cells in a discharging step that comprises: simultaneously discharging said plurality of cells by drawing a discharge current from each cell that is dependent upon at least one estimated characteristic of said cell.

21. The method as claimed in claim 20, wherein the at least one characteristic comprises one or more of: a power capability, a storage capacity, a cell impedance, an energy cell type and an energy cell chemistry.

22. The method as claimed in claim 21, wherein the at least one characteristic comprises a storage capacity and wherein the storage capacity estimate is based on a coulomb counting method.

23. The method as claimed in claim 21, wherein said charge current and said discharge current are provided through a bi-directional DC to DC converter for each cell slot, said DC to DC converters being controlled by a controller.

24. The method as claimed in claim 21, wherein the characteristic associated with each slot is estimated periodically.

25. The method as claimed in claim 24, wherein the characteristic associated with each slot is estimated after every charge phase.

26. The method as claimed in claim 24, wherein the characteristic associated with each slot is estimated after every discharge phase.

27. The method as claimed in claim 21, comprising performing a full charge/discharge cycle upon startup.

28. The method as claimed in claim 21, comprising implementing a function which depends on the at least one characteristic estimate associated with each slot to determine the amount of charge/discharge current to sink/source to said slot.

29. The A method as claimed in claim 28, comprising storing a history of characteristic estimates for each slot and wherein said function depends on said history.

30. The method as claimed in claim 28, comprising updating the function every time the at least one characteristic is estimated.

31. The method as claimed in claim 28, wherein the at least one characteristic comprises capacity and wherein the function scales the charge/discharge current linearly with the estimated capacity.

32. An electrical energy storage device comprising: a plurality of slots for receiving electrical energy storage devices, further comprising: a controller for each cell slot; wherein the controllers are arranged to determine the type of electrical energy storage device in each slot.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Preferred embodiments of the invention will now be described by way of example only and with reference to the accompanying drawings in which:

[0053] FIG. 1 schematically illustrates a system for using reclaimed rechargeable cells in a scalable energy storage system;

[0054] FIG. 2 shows a DC-DC converter circuit for a power module of each cell in the system;

[0055] FIG. 3 shows a simplified schematic of the control system associated with each power module;

[0056] FIG. 4 illustrates a battery management system algorithm for charging and discharging cells within the system;

[0057] FIG. 5 shows the results of a simulation of multiple cell charging and discharging leading to cell synchronization;

[0058] FIG. 6 illustrates a combination of batteries with supercapacitors; and

[0059] FIG. 7 shows a rechargeable battery cell suitable for a series configuration.

DETAILED DESCRIPTION

[0060] A system for using reclaimed rechargeable battery cells is shown and described below. The electrical design of the system will be discussed with reference to FIGS. 1-3.

[0061] In order to maximize the remaining energy storage capacity in recovered Li-ion cells (or indeed other rechargeable cell chemistries) of varying degrees of degradation, the State of Charge (SOC) of each cell is monitored and controlled individually. This challenge is addressed by interfacing each cell with an individual power module. In the embodiments described here, the power module contains a small switch mode power supply (SMPS) which regulates the power in and out of the cell, a microcontroller which implements the control and BMS algorithms, and an output voltage bus that can be connected in parallel with other power modules to increase the energy storage capabilities of the entire system.

[0062] FIG. 1 shows a schematic representation of the system 100. Three power modules 10a, 10b, 10c are shown in parallel (although in principle any number of modules may be used). Power module 10a has a cell 20 (e.g. a reclaimed Li-ion cell) and a power stage 30 that includes a bi-directional DC-DC converter 32. Module connectors 42, 44 from each power module 10a, 10b, 10c are connected in parallel and connected to input/output port 50.

[0063] Port 50 provides a connection to the system 100 for connecting the system to devices (i.e. for using and thus discharging the cells 20) and for charging the cells 20, e.g. from a solar PV panel.

[0064] All three power modules 10a, 10b, 10c have the same components as those illustrated for module 10a.

Module Design

[0065] Each power module 10a, 10b, 10c may contain a micro-controller which runs the BMS and controls the power flow in and out of the cell 20. Alternatively a single microcontroller may simultaneously control all power modules 10a, 10b, 10c. In one embodiment a bi-directional half bridge dc-dc converter was used as the DC-DC converter 32. This is illustrated in FIG. 2. The converter 32 measures the input and output voltages, as well as the inductor current and battery temperature. All of these sensor inputs are also used by the BMS. The output voltage can be controlled by adjusting the duty cycles of MOSFETs Q.sub.H and Q.sub.L.

[0066] FIG. 2 shows a schematic of the power stage 30. The output of each bi-directional half bridge dc-dc converter 32 is a bi-directional power port which can be connected in parallel with other converters 32 and be connected to a charging source (not shown). The charging source can be, but is not limited to, a grid connected power supply or a solar PV panel. In the case where the output is connected to a solar PV panel, the maximum power point (MPP) of the panel will change with temperature and solar irradiance conditions. Therefore, the micro-controller will implement a perturb and observe maximum power point tracking (MPPT) algorithm to track the maximum power of the panel. The power stage 30, will charge the cell 20 when the voltage at the input port 50 is above a predetermined threshold.

[0067] Each converter 32 acts independently of the others to share the load between cells in proportion to their capacity.

Control

[0068] FIG. 3 shows a diagram of the control logic implemented in the micro-controller. There are three main operating modes of the converter. The discharge mode, Mode 1, provides a nominal 12 V to the output of the DC-DC converter 32. The charging modes, Mode 2 and Mode 3, are activated once the output of the converter 32 is connected to a voltage source between 14 V and 20 V.

[0069] Mode 1 is the discharge mode. In Mode 1, the cell 20 is being discharged into a load connected to the output terminals 42, 44 of the power module 10 (more generally connected to the port 50 of system 100). In this mode, the control flow switches S1 and S2 are in the up position as shown in FIG. 3. The inner current control loop with controller C.sub.i(z) and the outer voltage control loop with controller C.sub.v(z) work together to maintain a voltage V.sub.ref* at the output terminals. Current sharing of the load is achieved using voltage droop control. Voltage droop control works by adjusting the voltage reference V.sub.ref by subtracting a value proportional to the converter current. This new reference is V.sub.ref*, shown in FIG. 3. The proportionality constant which is multiplied by the converter current, K.sub.b, is inversely proportional to the capacity of the cell connected to the converter, 30, and is determined by the BMS. Thus, power modules 10a, 10b, 10c which have larger capacity cells 20 will provide more current to the load than modules 10a, 10b, 10c with smaller capacity cells 20.

[0070] While the converter 32 is operating, the BMS 60 monitors the cell 20, ensuring that it is operating within its safety limits. The BMS 60 also performs a capacity estimation (to be described further below) to determine the parameter K.sub.b.

[0071] Mode 2 is a first charging mode for charging with constant current. Mode 2 is activated when the power modules' outputs 42, 44 are connected to a charging source with a voltage between 14 V and 20 V. In this mode, the control flow switch 51 is in the down position of FIG. 3, and the voltage controller, C.sub.v(z), is off. The current reference I.sub.ref for the converter 32 is provided by the BMS 60 which is implementing a perturb and observe MPPT algorithm. The current reference I.sub.ref will be proportional to the capacity of the cell 20 and will vary according to the MPPT algorithm. If a new cell 20 is attached, the current reference I.sub.ref will be set to its minimum value. This minimum value is a preset value stored within the BMS. In the case where the converter 32 is connected to a grid-connected voltage source, the MPPT algorithm will request the maximum charging current for the cell 20 that is being charged.

[0072] During Mode 2, the BMS 60 monitors the cell voltage V.sub.bat and switches to Mode 3 when the upper voltage limit of the cell is reached. The upper voltage limit of the cells may be determined by detecting the chemistry of the cell, e.g. by examining the cell voltage-SOC curve for the cell (which is different for different chemistries). This is further discussed below.

[0073] Mode 3 is a second charging mode for charging with constant voltage. In Mode 3 the control flow switch 51 is in the up position of FIG. 3 and the control flow switch S2 is in the down position of FIG. 3. The BMS 60 provides a voltage reference V.sub.cc which is compared to the battery voltage V.sub.bat. The voltage controller C.sub.v(z) now controls the battery voltage V.sub.bat instead of the output voltage V.sub.out. The BMS 60 will determine when the battery is fully charged by monitoring I.sub.L (the inductor current from the converter 32) and comparing it to a preset cut-off current. It will also determine if there is enough power from a charging source such as a solar PV panel by ensuring that V.sub.out remains above 14 V while I.sub.L is still charging the battery.

[0074] The algorithms designed for this BMS 60 serve two main purposes:

A) Condition monitoring for safe operation
B) Current control and balancing of individual cells

A. Condition Monitoring for Safe Operation

[0075] Each cell 20 is equipped with a temperature sensor, a voltage sensor and a current sensor. One example of upper and lower safety limits on those parameters is given in Table 1 below. These safety limits are based on a review of manufacturer specifications of Li-ion cells commonly used in electronic devices. Temperature limits are similar for most cell types. The lower temperature limit is more conservative for charging, since very low temperatures can trigger lithium plating and dendrite growth, which can lead to internal short circuits. Voltage limits depend on the cathode chemistry. For example LFP cells generally have a lower range of operating voltage than most other chemistries (3.6 V to 2.0 V).

[0076] In some embodiments, the cell type may be identified by an initial voltage measurement of the newly inserted cell and measurement of the voltage gradient during a subsequent charging step. For example LFP cells may be identified by detecting a sharp voltage gradient during charge when approaching their maximum voltage of 3.6 V. In this example, for all other chemistries, the most conservative voltage range of 4.2 V to 3.0 V is applied. The current is limited to 3.0 A, which is well within the operating range of 2500 mAh to 2900 mAh cells. The identified cell chemistry and associated voltage limits are stored in the memory of the controller for the corresponding cell slot.

TABLE-US-00001 TABLE 1 Safety Limits Parameter Upper Limit Lower Limit Temperature 60° C. charge: 0° C. discharge: −10° C. Voltage 3.6 V-4.2 V 2.0 V-3.0 V Current 3.0 A —

[0077] In some embodiments, newly inserted cells that have already reached their end of life may be identified during an initial voltage measurement and subsequent charging step. For example, if the cell voltage of the newly inserted cell is below a minimum preset value (e.g. 1.5 V) and the cell voltage rises rapidly when a charging current is applied and reaches an upper voltage limit of 3.6 V within a few seconds or minutes, the cell may be identified as unsuitable for operation and the user is informed by an indicator, e.g. an LED light. Similar procedures may be used to identify cells that have become unsuitable over time and need to be replaced.

[0078] In this particular embodiment, these safety limits are continuously monitored, by sampling at a frequency of 5 kHz. Breaching any safety limits triggers an immediate shut down of the power module, isolating the affected cell 20.

B. Algorithms for Current Control and Cell Balancing

[0079] As described above, the bi-directional DC-DC converters 32 allow independent current control on each cell 20. In order to optimally utilize their capacities, the current through each cell 20 is controlled such that all cells 20 discharge simultaneously. This means that a given load current is provided by individual cells 20 according to their capacities, i.e. higher capacity cells are subjected to higher currents than lower capacity cells. The system 100 is designed for used cells. For such cells the cell capacities are not initially known. More generally for used cells, the type (e.g. chemistry) and state of health of any given cell is not known. This problem is addressed with an algorithm that estimates cell capacities, for example (but not limited to) by means of a comparative/iterative Coulomb counting approach. The capacity of a cell 20 at a given discharge current can be calculated according to:

Q=∫.sub.t=0.sup.tI(t)dt

where I is current and t is discharge time.

[0080] For discrete time intervals k, this can be expressed as:

[00001]$Q=\underset{k=1}{\overset{N}{.Math.}}.Math..Math.I_k.Math.\Delta .Math..Math.t$

[0081] An estimate of cell capacity can thus be calculated from accurate current measurements performed at small time intervals. This cell capacity measurement can be employed to determine the parameter K.sub.b, used in the voltage droop controller as shown in FIG. 3.

[0082] Capacity measurements and current scaling are implemented for all cells 20 in the system 100 and the computations can be repeated with every charge and discharge cycle, as illustrated in the flowchart shown in FIG. 4.

[0083] As shown in FIG. 4, the algorithm is initiated with a first constant current constant voltage (CCCV) charge to balance the cells at a uniform state of charge (SOC) (Step 1 in FIG. 4). This initial charging step may be performed upon system start-up or after a reset operation, or after a new cell has been inserted or replaced. All cells 20 are charged with equal currents to their maximum voltages, which are held until a predefined time limit is exceeded. After that, the cells 20 are discharged with equal currents, until the cut-off voltages are reached (this is the first pass through Step 2 in FIG. 4). Measuring the time of this first discharge cycle allows calculating the cell capacities and provides an initial estimate of K.sub.b.

[0084] For the CCCV charge cycle of Step 3 of FIG. 4, the K.sub.b value calculated in Step 2 is used to correct the charge current. In Step 3, the charge capacity is calculated by coulomb counting in the same manner as the discharge capacity. K.sub.b is updated at the end of the charge cycle.

[0085] Upon start-up of the device, a full charge-discharge-charge cycle is preferably conducted (i.e. Steps 1 to 3 of FIG. 4) in order to adjust cell currents and synchronize charge and discharge times. During subsequent ordinary operation the current correction factors K.sub.b are updated. K.sub.b may be updated for each cell by continuous comparison of the charged/discharged energy with that of the previous charge/discharge cycle. In this manner, K.sub.b continually reflects the changing capacity as the cell degrades over time. A history of K.sub.b values is preferably recorded to provide more information on the cell.

[0086] The above described algorithm was implemented in MATLAB Simulink. FIG. 5 demonstrates how the algorithm synchronises discharge and charge cycles of three cells 20 with different capacities by adjusting the current load on each cell 20 in proportion to their capacities. It can be seen that the cycle times of the three cells 20 converge after the first two discharge-charge cycles. The cell capacities used for this simulation were 1600 mAh for Cell 1, 2000 mAh for Cell 2 and 2400 mAh for Cell 3. The standard Li-ion battery model of Simscape SimPowerSystems was used to emulate the battery voltage in response to a current load. For the first cycle of the simulation, a discharge power of 30 W and a charge power of 20 W were divided equally among the three cells. The cell capacities were estimated during each successive charge and discharge cycle to vary the parameter K.sub.b. The simulation clearly demonstrates the feasibility of the BMS algorithm described above and illustrated in FIG. 4 for synchronizing cells of different capacities by active current control.

[0087] The system is capable of combining battery and supercapacitor cells in order to combine the high energy capabilities of batteries with the high power capabilities of supercapacitors. FIG. 6 shows a possible hardware layout of such a combination. The DC to DC converters at each slot enable any arbitrary combination of battery cells and supercapacitors of different voltage and capacity characteristics. The currents drawn from supercapacitors and batteries are scaled with a factor K, which is a function of the energy capabilities K.sub.Q and the power capabilities K.sub.p, of supercapacitors and battery cells, and of the rate of change of the load current. For example, if the system experiences a high power demand over a short time period (i.e. large values for

[00002]$\frac{{\mathrm{dI}}_{\mathrm{total}}}{\mathrm{dt}}),$

the current is primarily drawn from the supercapacitors, which are identified by high values for K.sub.p, that are stored in the BMS for the corresponding cell slot. If the system experiences a low power demand over extended time periods (i.e. small values for

[00003]$\frac{{\mathrm{dI}}_{\mathrm{total}}}{\mathrm{dt}}),$

the current is drawn from the battery cells, which are identified by high values for K.sub.Q. The values K.sub.Q and K.sub.p are updated as the battery and supercapacitor cells degrade over time, as described in the algorithm for cell management.

[0088] FIG. 7 shows a rechargeable cell 120 with a battery management system (BMS) 130 that is suitable for combining a number of cells in series while charging and discharging the cells at a rate according to their capacity in a similar manner as described above, implementing a modular multi-level control (MMC) circuit. The BMS 130 controls two switches 135, 136. When first switch 135 is on and second switch 136 is off, the cell 120 is connected into the circuit. When the first switch 135 is off and the second switch 136 is on, the cell 120 is disconnected from the circuit (bypassed). By carefully adjusting the switching rates so as to adjust the duty cycle for which the cell 120 is connected into the circuit, the cell's charge rate and discharge rate can be controlled. High switching rates and a filter may be used to smooth out the resultant voltage fluctuations. The circuit of FIG. 7 may be replicated, with the negative terminal 140 of one circuit connected to the positive terminal 150 of an adjacent circuit so as to connect multiple cells 120 in series.