WO2015092441A2

WO2015092441A2 - Electricity generation

Info

Publication number: WO2015092441A2
Application number: PCT/GB2014/053815
Authority: WO
Inventors: Malcolm MCCULLOGH; Peter Michael ARMSTRONG; Ren KANG; Richard Wong
Original assignee: Isis Innovation Limited
Priority date: 2013-12-19
Filing date: 2014-12-19
Publication date: 2015-06-25
Also published as: GB201322600D0; WO2015092441A3

Abstract

A system comprising a reconfigurable array of electricity generating cells for charging a battery; an array switching means for dynamically reconfiguring the array; a means for DC-DC conversion of an array output voltage; and a means for altering the DC-DC conversion; wherein both the array switching means for array configuration and the means for altering the DC-DC conversion are modified in dependence on a battery terminal voltage.

Description

Electricity generation

The present invention relates to a dynamically reconfigurable array of electricity generating cells for charging a battery. The present invention further relates to a controller for electricity generating cells for charging a battery.

Reconfigurable PV array

According to one aspect of the invention, there is provided a system comprising: a reconfigurable array of electricity generating cells for charging a battery;

an array switching means for dynamically reconfiguring the array;

a means for DC-DC conversion of an array output voltage; and

a means for altering the DC-DC conversion;

wherein both the array switching means for array configuration and the means for altering the DC-DC conversion are modified in dependence on a battery terminal voltage (preferably such modifying being performed by means for modifying such as a controller).

By reconfiguring the array (such as a photovoltaic array), the array output voltage that provides the maximum available power can be matched to a battery terminal voltage. The efficiency of power exploitation of the array can thus be increased. By including a means for DC-DC conversion, charge can be extracted from the array at a voltage that provides the maximum available power, and yet a battery terminal voltage can be matched. By reconfiguring the array and at the same time including a means for DC-DC conversion, efficiency of the power exploitation of a reconfigurable array can be increased further.

As used herein, the term 'altering' preferably includes controlling a DC-DC converter to change the output voltage and/or bypassing the DC-DC converter altogether.

Preferably the system further comprises a controller adapted to determine an array configuration and a DC-DC conversion in dependence on a measured battery terminal voltage and modify the array switching means for array reconfiguration and the means for altering the DC-DC conversion in dependence on the measured battery terminal voltage in order to implement the determined array configuration and DC-DC conversion.

Preferably the means for altering the DC-DC conversion comprises a DC-DC conversion bypass switch for bypassing the means for DC-DC conversion in order to provide electricity from the electricity generating cells directly to the battery.

Preferably the or a controller is adapted to determine a DC-DC conversion bypass switch configuration in dependence on the or a DC-DC conversion; and modify the means for altering the DC-DC conversion in dependence on the or a DC-DC conversion in order to implement the determined DC-DC conversion bypass switch configuration. The controller may compare an efficiency penalty associated with DC-DC conversion and an efficiency penalty associated with providing electricity from the electricity generating cells directly to the battery in order to determine a DC-DC conversion bypass switch configuration. Preferably the controller is adapted to determine the array configuration and the DC-DC conversion and/or the or a DC-DC conversion bypass switch configuration further in dependence on a level of irradiance, a distribution of irradiance and/or an ambient temperature. Preferably a measured open circuit voltage and/or a measured short circuit current of the array is used to quantify the level of irradiance, the distribution of irradiance and/or the ambient temperature.

The controller may determine an array configuration and a DC-DC conversion and/or the or a DC-DC conversion bypass switch configuration by a look-up routine. The controller may determine an array configuration and a DC-DC conversion by a machine learning routine. The machine learning routine may be a fuzzy logic machine learning routine. The machine learning routine may be trained with a power output to the battery as feedback. The controller may be adapted to determine the array configuration and the DC- DC conversion by a configuration perturbation routine comprising steps of changing the array configuration and/or the DC-DC conversion, and comparing a power output to the battery before and after the change. The configuration perturbation routine may comprise an array configuration and/or DC-DC conversion sweep.

Preferably the array switching means for array reconfiguration is a remotely controllable switch.

Preferably the system further comprises a battery for being charged by the reconfigurable array of electricity generating cells.

Preferably the system further comprises a battery comprising a plurality of battery cells and a battery switching means for dynamically reconfiguring the plurality of battery cells, wherein the array switching means and the battery switching means are dynamically modified to selectively connect a sub-array of electricity generating cells to a sub-group of battery cells. Redundant PV cells charge parts of battery

According to another aspect of the invention, there is provided a system comprising an array switching means for dynamically reconfiguring a reconfigurable array of electricity generating cells for charging a battery with a plurality of battery cells; and a battery switching means for dynamically reconfiguring the plurality of battery cells; wherein the array switching means and the battery switching means are dynamically modified to selectively connect a sub-array of electricity generating cells to a sub-group of battery cells.

Preferably the or a controller is adapted to determine an array configuration comprising a sub-array and a battery configuration comprising a sub-group; and control the array switching means and the battery switching means in order to implement the determined array configuration and battery configuration.

Preferably the battery configuration is determined in dependence on the determined array configuration. Preferably the controller is adapted to provide charge from a first sub-array to a first sub-group of battery cells at a first voltage, and provide charge from a second sub-array to a second sub-group of battery cells at a second voltage. The second sub-array may comprise remainder cells to the first sub-array. The second sub-array may comprise underperforming cells. The underperforming cells may be damaged, shaded, and/or deteriorated cells.

Preferably the battery switching means is a remotely controllable array of switching elements connected to nodes between battery cells.

Preferably the or a controller is adapted to determine a sub-group of battery cells in dependence on a battery load; and control the battery switching means in order to provide charge from the sub-group of battery cells to the battery load.

Preferably the or a controller is adapted to provide charge to and/or draw charge from first a first sub-set of battery cells and then a second sub-set of battery cells in order to balance the charge across the battery cells. The system may further comprise a reconfigurable array of electricity generating cells for charging a battery. The system may further comprise a battery with a plurality of battery cells for being charged by the reconfigurable array of electricity generating cells. Preferably the battery switching means is dynamically adapted to provide charge to and/or draw charge from the battery by first providing a selective connection to a first sub-set of battery cells, and then providing a selective connection to a second sub-set of battery cells in order to balance the charge across the battery cells.

Balancing by multiple voltage distribution

According to a further aspect of the invention, there is provided a system comprising a battery switching means for dynamically reconfiguring a plurality of battery cells in a battery, wherein the battery switching means is dynamically adapted to provide charge to and/or draw charge from the battery by first providing a selective connection to a first sub-set of battery cells, and then providing a selective connection to a second sub-set of battery cells in order to balance the charge across the battery cells. Preferably the or a controller is adapted to determine an over- or undercharged sub-set of battery cells and control the battery switching means in order to correct the over- or undercharging. The battery switching means may comprise an array of switching elements connected to nodes between battery cells for providing charge to and/or drawing charge from a sub-set of battery cells. The system may further comprise a battery with a plurality of battery cells.

Preferably the battery switching means is dynamically adapted to provide voltage from a sub-cluster of the battery cells directly to a transistor as a gate drive source.

High voltage battery part for gate drive source

According to a yet further aspect of the invention, there is provided a system comprising a battery switching means for dynamically reconfiguring the plurality of battery cells in a battery, wherein the battery switching means is dynamically adapted to provide voltage from a sub-cluster of the battery cells directly to a transistor as a gate drive source (preferably such adapting being performed by means for adapting such as a controller).

As used herein, the term 'directly' preferably indicates in the absence of a gate drive circuit such as a boost circuit.

Preferably the or a controller is adapted to determine a required gate drive source voltage and control the battery switching means in order to provide at least the required gate drive source voltage from the sub-cluster.

The system may further comprise a battery with a plurality of battery cells. Reconfigurable PV array inverter

According to a yet further aspect of the invention, there is provided a system comprising: a reconfigurable array of electricity generating cells for providing electricity to a network;

an array switching means for dynamically reconfiguring the array;

a means for DC-AC conversion of an array output voltage; and

a means for altering the DC-AC conversion;

wherein both the array switching means for array configuration and the means for altering the DC-AC conversion are modified in dependence on an array output voltage (preferably such modifying being performed by means for modifying such as a controller).

Reconfigurable converter

According to a yet further aspect of the invention, there is provided a system comprising:

a reconfigurable converter for DC-DC conversion;

a converter switching means for dynamically reconfiguring the converter;

wherein the converter switching means for converter reconfiguration is modified in dependence on a converter load.

According to a yet further aspect of the invention, there is provided a method of providing electricity from a reconfigurable array of electricity generating cells to a battery for charging comprising:

determining a battery terminal voltage;

dynamically reconfiguring the array in dependence on the battery terminal voltage; and

adapting a means for DC-DC conversion of an array output voltage in dependence on a battery terminal voltage.

According to a yet further aspect of the invention, there is provided a method of providing electricity from a reconfigurable array of electricity generating cells to a battery with a plurality of battery cells for charging comprising:

dynamically reconfiguring a reconfigurable array of electricity generating cells for charging a battery; and

dynamically reconfiguring the plurality of battery cells; and

selectively connecting a sub-array of electricity generating cells to a sub- group of battery cells. According to a yet further aspect of the invention, there is provided a method of providing charge to and/or drawing charge from a battery with a plurality of battery cells comprising:

first dynamically reconfiguring the battery to provide a selective connection to a first sub-set of battery cells; and

then dynamically reconfiguring the battery to provide a selective connection to a second sub-set of battery cells, such that the charge across the battery cells becomes balanced.

According to a yet further aspect of the invention, there is provided a method of providing charge from a battery with a plurality of battery cells to a transistor comprising:

dynamically reconfiguring the plurality of battery cells to form a sub-cluster of battery cells; and

providing voltage from the sub-cluster of battery cells directly to a transistor as a gate drive source.

According to a yet further aspect of the invention, there is provided a method of providing electricity from a reconfigurable array of electricity generating cells to a network comprising:

determining an array output voltage;

dynamically reconfiguring the array in dependence on the array output voltage; and

adapting a means for DC-AC conversion of an array output voltage in dependence on an array output voltage.

According to a yet further aspect of the invention, there is provided a method of DC-DC conversion comprising:

determining a DC-DC converter load;

dynamically reconfiguring a DC-DC converter in dependence on the DC- DC converter load.

Features and aspects of the present invention include Reconfigure photovoltaic (PV) array to optimise energy transfer to battery under load

• A reconfigurable PV array connected to a DC battery system where energy transfer from the PV cells is maximised by optimal selection of system configuration.

• A reconfigurable array where the desired configuration is implemented using an assortment of switching elements connected between PV cells.

• A reconfigurable array where the switching elements can be: a relay, a bipolar transistor, opto-isolated transistor, MOSFET(s), thyristor(s), IGBT or any device capable of controlling the power that flows across it.

• A reconfigurable PV array where the switching elements comprise of a P channel and N channel MOSFET to allow small voltage control signals to control switching across the P Channel MOSFET.

• A reconfigurable PV energy storage system where an algorithmic routine maximises power extracted from the system by perturbing the system configuration and observing the resulting change in extracted power.

• A reconfigurable PV energy storage system where an algorithmic routine maximises power extracted from the system by measuring the open circuit voltage (or short circuit current) of the PV array to infer irradiation level and battery terminal voltage so that the optimal configuration can be determined from a look-up table.

• A reconfigurable PV energy storage system where an algorithmic routine maximises power extracted from the system by combining a look-up routine and configuration perturb and observe scheme.

• An algorithmic approach involving a current sweep method.

• An algorithmic approach involving fuzzy logic routine or any other machine learning approach.

• An algorithmic approach involving incremental conductance combined with any other approach.

Bypass power electronics

• A reconfigurable PV system, where the power electronics can be bypassed to reduce losses associated with components such as, MOSFETS, diodes, inductors, capacitors etc.

• A reconfigurable PV system where the power electronics comprise of any of: Buck converter, boost converter, buck-boost converter, flyback converter, resonant converter (soft switching variant of previous topologies), linear regulator or any DC-DC conversion circuitry capable of transferring energy from one voltage level to another.

Different voltage levels within battery pack

• A reconfigurable PV energy storage system where remainder cells that arise between configurations can be connected to different voltage levels within a battery pack.

• A reconfigurable PV energy storage system where remainder cells that arise between configurations can be connected to different voltage levels within a battery pack using some form of cell tap (feed) selector.

• A reconfigurable energy storage system where the cell tap (feed) selector comprises an array of switching elements connected to nodes between cells within a battery pack.

• A reconfigurable energy storage system where the cell tap (feed) selector switching elements can be: a relay, a bipolar transistor, opto-isolated transistor, MOSFET(s), thyristor(s), IGBT or any device capable of controlling the power that flows across it.

• A reconfigurable energy storage system where the cell tap (feed) selector switching elements comprise of a P channel and N channel MOSFET to allow small voltage control signals to control switching across the P Channel MOSFET.

• A reconfigurable PV energy storage system where multiple voltage levels are provided from the battery system via a cell tap selector for the purposes of servicing loads with differing voltage requirements.

• An energy storage system where an algorithm controls cell tap (feed) selectors such that energy sources and energy sinks are extracted from the appropriate level within a battery pack.

• An energy storage system where an algorithm controls cell tap (feed) selectors such that cells remain balanced at all times.

• An energy storage system where an algorithm controls cell tap (feed) selectors such that cells remain balanced at all times and where a passive battery balancing system can be brought in where controllable energy sources and energy sinks are insufficient/excessive for the purposes of maintaining cell balance.

Other

• A fixed or reconfigurable PV array where a higher voltage source, selected from within a reconfigurable battery pack, is used as a source for the gate drive signal.

• A reconfigurable PV array where any one or more active switching elements within the switching matrix are used to provide isolation of shaded regions of the PV array.

The invention extends to methods and/or systems substantially as herein described with reference to the accompanying drawings.

The invention also provides a computer program and a computer program product for carrying out any of the methods described herein and/or for embodying any of the apparatus features described herein, and a computer readable medium having stored thereon a program for carrying out any of the methods described herein and/or for embodying any of the apparatus features described herein.

The invention also provides a signal embodying a computer program for carrying out any of the methods described herein and/or for embodying any of the apparatus features described herein, a method of transmitting such a signal, and a computer product having an operating system which supports a computer program for carrying out any of the methods described herein and/or for embodying any of the apparatus features described herein.

Any apparatus feature as described herein may also be provided as a method feature, and vice versa. As used herein, means plus function features may be expressed alternatively in terms of their corresponding structure, such as a suitably programmed processor and associated memory.

Any feature in one aspect of the invention may be applied to other aspects of the invention, in any appropriate combination. In particular, method aspects may be applied to apparatus aspects, and vice versa. Furthermore, any, some and/or all features in one aspect can be applied to any, some and/or all features in any other aspect, in any appropriate combination.

It should also be appreciated that particular combinations of the various features described and defined in any aspects of the invention can be implemented and/or supplied and/or used independently.

Furthermore, features implemented in hardware may generally be implemented in software, and vice versa. Any reference to software and hardware features herein should be construed accordingly.

These and other aspects of the present invention will become apparent from the following exemplary embodiments that are described with reference to the following figures in which:

Figure 1 shows photovoltaic device voltage-current curves and voltage-power curves for different levels of solar irradiation;

Figure 2 shows voltage-current curves and voltage-power curves for alternative configurations of a reconfigurable photovoltaic array;

Figure 3 shows a reconfigurable photovoltaic system;

Figure 4 shows a further reconfigurable photovoltaic system;

Figure 5 shows a battery system;

Figure 6 shows a further battery system;

Figure 7 shows a system with a gate drive source cell tap selector;

Figure 8 shows a switching array for a reconfigurable photovoltaic array;

Figure 9 shows a switching element;

Figure 10 shows the switching array of Figure 8 in a first configuration;

Figure 1 1 shows the switching array of Figure 8 in a second configuration;

Figure 12 shows the switching array of Figure 8 in a third configuration;

Figure 13 shows another switching array topology for a reconfigurable photovoltaic array;

Figure 14 shows the switching array of Figure 13 in a first configuration;

Figure 15 shows the switching array of Figure 13 in a second configuration;

Figure 16 shows a voltage-current curve and voltage-power curve for a first system configuration; Figure 17 shows a voltage-current curve and voltage-power curve for a second system configuration;

Figure 18 shows a voltage-current curve and voltage-power curve for a third system configuration;

Figure 19 shows a flow chart for configuration control by a configuration perturbation routine;

Figure 20 shows a flow chart for configuration control by a parameter-based lookup routine;

Figure 21 shows a decision tree for a look-up routine;

Figure 22 shows a flow chart for configuration control by a combined parameter- based look-up routine and configuration perturbation routine;

Figure 23 shows a reference map of optimum operating configurations;

Figure 24 shows a battery management system;

Figure 25 shows a first configuration for a solar photovoltaic installation;

Figure 26 shows a second configuration for a solar photovoltaic installation;

Figure 27 shows a conventional resonant converter; and

Figure 28 shows a reconfigurable resonant converter.

A photovoltaic (PV) device, typically comprising a number of photovoltaic cells, can be characterised by its voltage-current curve or voltage-power curve such as are shown in Figure 1 . Ideally such a PV array is operated such that maximum power is extracted at all times. To maximise the output from a PV array for a given irradiation, power electronics ensure that energy extraction occurs at the voltage associated with maximum power, at the so-called knee 10 of the power- voltage curve.

The energy available from PV systems is intermittent and variable in nature and consequently photovoltaics are commonly used in tandem with an energy storage system such as a battery, in particular for mobile and isolated (non-grid- connected) applications. As a battery is subjected to a load, it can discharge (if the load draws more energy than the amount delivered from the PV array). As a consequence, the battery terminal voltage drops.

Power electronics are traditionally used to extract maximum power from the PV array, and modulate that power to transfer it from the PV array to the battery. In particular DC-DC voltage converters that modulate the power can lose efficiency as the difference between the input and output voltage increases. Therefore maintaining the PV array voltage at a level close to the battery terminal voltage is desirable.

Dynamic reconfiguration of a PV array allows the array voltage associated with maximum power to follow the battery terminal voltage so that DC-DC conversion losses can be avoided or reduced. Figure 2 shows voltage-current curves and voltage-power curves for alternative configurations of the PV array. An array configuration can be selected for a given battery terminal voltage in order to both provide maximum power, and also to minimise the difference between the input and output voltage at the power electronics. By this means a reconfigurable PV array can improve the efficiency of power take-off and optimise the output of a solar panel with respect to a DC-battery load. The reconfigurable PV array can be implemented using an array of transistors. An example of a switching scheme for a reconfigurable PV array is described in more detail below. A variety of control schemes are also described in more detail below.

As the array of PV cells can be reconfigured, the DC-DC converter can in some situations be bypassed in order to reduce the losses associated with the power electronics.

The reconfigurable PV array can in some situations include PV cells that are not in use. Such remainder cells, which would otherwise be redundant, can be utilised at different power feed positions within a battery pack. Redundant groups of PV cells that are created during reconfiguration are made use of by injecting them at lower voltage levels within the battery pack. The input feed points are rotated to maintain the cells in balance at all times. Multiple voltage levels, which would otherwise be delivered from DC-DC converters connected to the primary battery output terminals, are instead delivered directly from the battery pack at multiple power tap positions. This can save energy that would normally be lost through DC-DC conversion from the primary pack terminals. Battery cell balance can be maintained by rotation of extraction tap points. The system makes use of the existing balancing scheme that is typically found in lithium ion battery packs, where so-called 'balancing' wires are in place to facilitate charging and pack maintenance. Additionally loads and feeds can be shifted around the battery pack so that battery cells can be balanced without the losses associated with a traditional, resistive passive scheme.

By integrating the battery pack with the battery management system it becomes possible to source the gate drive voltage from a higher tap point, and thereby reduce the overheads associated with boot-strapping the gate of the switching transistor.

A reconfigurable PV system 300 is now described in more detail with reference to Figures 3 and 4.

A reconfigurable PV array 302 is formed of individual PV cells, in the illustrated example 130 cells. The reconfigurable PV array 302 is connected to a battery 306, such as a lithium ion battery pack. The battery 306 has multiple battery cells typically in series (though parallel connections between the cells are possible). The configuration of the reconfigurable PV array 302 is selected by a controller 308, typically by a maximum power point tracking (MPPT) control algorithm 316. The MPPT control algorithm 316 monitors by way of a power feedback 318 the output power from the PV array 302 (by measuring PV array output current and voltage).

The power electronics 304 includes a DC-DC converter 312, for example with buck, boost, buck-boost or flyback topology. The maximum power point tracking (MPPT) algorithm 316 also controls the duty cycle applied to the switching device within the DC-DC converter 312 in order to maintain a DC-DC conversion level and thus optimise the power supplied to the battery 306.

The PV panel is a reconfigurable array 302 that can be reconfigured continuously (dynamically and in real time) to ensure that the maximum power point voltage is maintained close to the terminal voltage of the battery 306. This improves the efficiency of the DC-DC converter 312 and, in some modes of operation, allows the DC-DC converter 312 with the associated efficiency penalty to be bypassed altogether. A bypass switch 314 (such as a MOSFET switch) can enable bypassing of the DC-DC converter 312. For some panel configurations 'redundant' cells are left over that cannot share the common output terminal associated with the main PV array. For example in a panel with 130 cells when a 12x10 array is selected there are 10 'remainder' cells. Such residual PV cells are utilised separately by connecting them at an appropriate voltage level within the battery system 410 using a battery cell feed selector 406.

Figure 4 shows a system 400 with remainder cells 402 utilised in addition to the main PV array 404. A battery cell feed selector 406 selects the appropriate voltage level via switching devices connected to battery cell nodes within the battery system 410. Examples of suitable battery systems 410 are described in more detail with reference to Figures 5 and 6. A battery management system 310 controls a cell feed selector 406 that changes the position of the connection to ensure that individual battery cells 412 within the battery system 410 remain nominally balanced with respect to one another. A battery cell tap selector 408 can also be controlled by the battery management system 310 to balance the battery system 410.

The remainder cells may also arise not from a particular configuration selection, but due to underperformance of individual PV cells. For example, if a part of the PV panel is shaded, certain PV cells can underperform and be grouped together in a separate array. In order to do so the switching elements within the reconfigurable PV array can be used as active blocking diodes. Feedback of current through the switching element can inform a controller when a shaded region of the panel is underperforming. An active diode function can be implemented to isolate that region of the panel without the losses normally associated with the forward diode voltage that occurs in conventional arrays.

For some applications multiple voltage levels are required from a single battery pack. In order to provide multiple voltage levels from a single battery pack conventionally a voltage regulator (switching or linear DC-DC conversion) converts the battery terminal voltage to the desired voltage. Each voltage level incurs a loss due to the DC-DC conversion. The losses of such DC-DC conversion are proportional to the difference between the battery terminal voltage and the required voltage level.

Instead of using a battery output voltage regulator, in the system 400 battery cells can be grouped into sub-groups at different voltages. These different voltage levels can be provided directly or with lower losses. Battery output cell tap selectors 408 are used to provide multiple load voltages.

Figure 5 illustrates a battery system 500 with switches 506 for battery input cell feed selectors connected to cell nodes within the battery system 410. A battery input cell feed selector 406 can implement suitable connections of a power input 504, marked rPV+ and rPV-, by means of battery input switches 506 in order to form sub-groups of battery cells and to connect those sub-groups to the power source 504. In the example the power source 504 is a PV panel, but other power sources can be used.

Figure 6 illustrates a battery system 600 with switches 506 for battery input cell feed selectors (marked in solid lines) for a power source 504, as shown in Figure 5. Additionally switches 604 606 for battery output cell tap selectors (marked in broken lines) for two different outputs 600 602, a 5V output 600 and a 12V output 602, are provided (marked in different broken lines). A cell tap selector for the 5V output 600 can implement suitable connections by means of battery output switches 604 in order to form sub-groups of battery cells and to connect those sub-groups to a load requiring 5V. Similarly, another cell tap selector for the 12V output 602 can implement suitable connections by means of battery output switches 606 in order to form sub-groups of battery cells and to connect those sub-groups to a load requiring 12V.

Figure 7 shows a system 700 where a gate drive source cell tap selector 702 provides a higher gate drive voltage for a MOSFET. Conventionally, in order to provide a high enough voltage to operate a MOSFET switch, a dedicated gate drive circuit such as a boost circuit (e.g. a boot strap circuit) is included. Instead of using a gate drive circuit, the gate drive source cell tap selector 702 can select the gate drive source from the battery system 410 such that its voltage is sufficiently high. Similar to the other selectors described above the gate drive source cell tap selector 702 can implement suitable connections in the battery system 410 by means of appropriate battery output switches in order to form sub- groups of battery cells and connect those sub-groups to the MOSFET gate. By using the gate drive source cell tap selector 702 losses associated with a gate drive circuit can be avoided.

In the example illustrated in Figure 7 the MOSFET is associated with the DC-DC converter 312 in the power electronics 304. The MOSFET may alternatively be elsewhere in the system, for example in switching elements within the PV array or the DC-DC bypass switch 314. A sufficiently high gate drive voltage is provided by selection of an appropriate sub-group of battery cells (with a sufficient number of individual battery cells in series and hence a sufficiently high potential) by the gate drive source cell tap selector 702. It may be that a power input to the battery is supplied at a voltage below the battery terminal voltage, for example using remainder cells as described above.

An example of a switching array for a reconfigurable PV array is now described in more detail with reference to Figures 8 to 12.

Figure 8 shows a switching array for a reconfigurable PV array with 130 individual cells. The switching array is made up of a number of switching elements, all shown to be open in Figure 8. The switching elements are for example relays, contactors, MOSFETs, bipolar transistors, thyristors or any other device capable of connecting or interrupting the flow of power across it. In Figure 9 an example of a switching element is shown with a pair of MOSFETs, one N channel and one P channel. The reconfigurable PV array with 130 individual cells can assume the following configurations:

• 13x10

• 12x10 + 1 x10

• 10x13 A connection scheme for switching elements to provide the 13x10 configuration is shown in Figure 10, with shading indicating closed switches. 13 cells are connected in series, and 10 such series strings are connected in parallel. A connection scheme for switching elements to provide the 12x10 + 1 x10 configuration is shown in Figure 1 1 , with shading indicating closed switches. 12 cells are connected in series, and 10 such series strings are connected in parallel. A further sub-array of 10 remainder cells is shown connected in parallel, but can just as well be connected in series.

A connection scheme for switching elements to provide the 10x13 configuration is shown in Figure 12, with shading indicating closed switches. 10 cells are connected in series, and 13 such series strings are connected in parallel. Each of the three configurations can be connected to a battery via a DC-DC converter with maximum power point tracking, or directly via a DC-DC bypass switch, depending on proximity between maximum power point voltage and battery terminal voltage. Since the PV cells can be either connected directly to the battery or via a DC-DC converter for each of the three PV array configurations, there are 6 possible system configurations that can be implemented.

Figures 13 to 15 show another switching array topology for a reconfigurable photovoltaic array. Figure 14 shows a connection scheme for switching elements to provide the 13x10 configuration, with shading indicating closed switches. 13 cells are connected in series, and 10 such series strings are connected in parallel. Figure 15 shows a connection scheme for switching elements to provide the 10x13 configuration, with shading indicating closed switches. 10 cells are connected in series, and 13 such series strings are connected in parallel.

Operation of the reconfigurable PV array with a sequence of different system configurations in response to a load drawing power from a battery is now described in more detail with reference to Figures 16 to 18. An example where an electric vehicle draws power from a 48V battery system attached to a reconfigurable PV array is described. Initially the battery is fully charged and, as shown in Figure 16, the battery terminal voltage is at 49 V which is close to the knee 10 of the l-V curve associated with the 13x10 configuration. If the PV array is connected directly to the battery, the PV terminal voltage matches the battery terminal voltage since the impedance of the battery is low compared to the impedance of the PV array. In the initial condition the loss associated with the deviation from maximum power (for direct connection bypassing DC-DC converter) is less than the loss associated with the DC-DC buck converter (for connection via DC-DC converter), so the controller causes the DC-DC buck converter to be bypassed and the 13x10 configuration is directly connected to the battery.

As the vehicle is used, the battery discharges since the propulsion system draws more energy than the amount delivered from the PV array. As a consequence, the battery terminal voltage drops to 46V, as shown in Figure 17. At this point, the controller determines that the 10x13 configuration, directly connected, is still the most efficient system configuration.

As the battery is discharged further, the terminal voltage drops to 42 V as shown in Figure 18. At this voltage the efficiency penalty associated with the deviation from maximum power is greater than the efficiency penalty associated with the DC-DC buck converter. Consequently the buck circuit DC-DC converter is enabled by the controller. As the battery discharges further, the 12x10 configuration becomes more favourable, and is eventually adapted by the controller.

Controllers for controlling the system configuration for the reconfigurable PV array are now described in more detail with reference to Figures 19 to 21 . The controller determines the system configuration that yields maximum power from the reconfigurable PV array.

A first example of a controller perturbs the system configuration (including the PV array and the power electronics) and observes the power output. Figure 19 shows a configuration perturbation routine. In the illustrated example, the perturbation routine initialises the PV array with the configuration that provides the highest voltage; in the example of the 130 cell array described above, this is the 13x10 configuration bypassing the DC-DC converter. In the next step the array configuration is changed, for example by switching to the 12x10 + 1 x10 array configuration, or by connecting by way of the DC-DC converter and changing the duty cycle of the DC-DC converter (in order to change the conversion level). In the new configuration, the power output (the power going into the battery) is determined and compared with the power output in the previous setting. If the power output is lower in the second configuration, then the previous configuration is adopted. If the power output is higher in the second configuration, then a further step of changing the array configuration is tested, and the routine continues until the optimal configuration is found.

In a variant of the configuration perturbation routine, the controller performs a sweep method where the system periodically sweeps across the full range of system configurations (including duty cycles for the DC-DC converter and PV array configurations) to determine the characteristics of the system (PV array and power electronics). As with the configuration perturbation routine the controller changes the panel configuration and observes the power output. Having tested all configurations, the optimum configuration can then be adopted. The sweep method tests all possible configurations, whereas the perturbation routine only tests until a (local) optimum is found. The sweep method is generally slower and less responsive than the perturbation routine. Another approach is to refer to a look-up table based on given parameters. An example of a look-up routine is shown in Figure 20. The terminal voltage of the battery and the solar intensity are determined, and based on these two parameters the optimal configuration is looked up in a reference table. A reference map of optimum operating configurations (as shown in more detail in Figure 23) or a look up table is derived from simulation or prior experiment.

In order to determine the solar intensity the open circuit voltage (typically measured at PV array level, but may alternatively be measured at PV cell level) is determined. The relationship between open circuit voltage and irradiance (or solar intensity) is known for example from prior measurement or from the PV device supplier. This relationship can be used in the look-up routine to determine the approximate level of irradiance, based on the open circuit voltage, for any particular configuration. The open circuit voltage can also be used as an indicator of distribution of radiance (e.g. partial shading of the panel) and/or ambient temperature. The other parameter for the look-up routine, the terminal voltage of the battery, can simply be monitored by the controller.

Instead of determining the open circuit voltage as a measure of the solar intensity, the short circuit current can be measured to provide an indication of the solar intensity. The relationship between short circuit current and irradiance (or solar intensity) is known for example from prior measurement or from the PV device supplier. This relationship can be used in the look-up routine to determine the approximate level of irradiance, based on the short circuit current, for any particular configuration. Measurement of the short circuit current at PV array level (or alternatively at PV cell level) requires a shunt FET. Assessing irradiance for the purposes of a look up routine measurement of the short circuit current can provide superior accuracy, whereas measurement of the open circuit voltage can be simply implemented with existing hardware. The controller can also access information relating to the DC-DC converter in order to determine the efficiency of the power electronics. If the voltage conversion required by the DC-DC converter is relatively large, the efficiency penalty associated with DC-DC conversion is also relatively large. A calculation model of the DC-DC converter can be used to quantify the efficiency penalty of the DC-DC converter based on a given input voltage and a desired output voltage. Similarly, a further calculation model can be used to quantify the efficiency penalty for bypassing the DC-DC converter. Since the internal impedance of the battery is low compared with the PV array, direct connection between the PV and battery forces the PV array terminal voltage to match the battery terminal voltage; hence the PV array might be forced to operate at a voltage other than at the knee of the power-voltage curve; this can cause an efficiency penalty for bypassing the DC-DC converter. Comparison of the efficiency penalties for direct connection and for connection via the DC-DC converter allows determination of the optimum power electronics configuration. Figure 21 shows the look-up routine as a decision tree with system parameters (battery voltage, PV array open circuit voltage) as inputs. A first decision level determines a PV array configuration and a second decision level determines whether to bypass the DC-DC converter or to use DC-DC conversion. A further level can determine the conversion level of the DC-DC converter (that is, the duty cycle of the DC-DC converter) if DC-DC conversion is used.

Instead of referring to a look-up table, the given parameters can be used in a machine learning approach based for example on probabilistic logic, fuzzy logic or a neural network. Such machine learning algorithms can classify a given set of parameters (such as terminal voltage of the battery and the solar intensity or PV array open circuit voltage or short circuit current) in order to assign an optimum configuration. A machine learning control scheme thus determines the best configuration and duty cycle for a given open circuit voltage or short circuit current and battery terminal voltage.

For training the machine learning algorithm (including by fuzzy logic or an artificial neural network) the power output (the power going into the battery) is used as a feedback input. The feedback changes the system model (e.g. mean and standard deviation of a number of conditions, such as too low voltage, low voltage, high voltage, too high voltage) to improve the selection of the system configuration. If experimental data is available then supervised learning can be used; if experimental data is not available then unsupervised learning can be used.

A further parameter-based approach, referred to as an incremental conductance method, uses the derivative of the output power with respect to the output voltage (that is, the change per voltage increment of the power going into the battery) to predict an optimum system configuration. This approach could be used to determine the optimum operating point on the voltage-current curve within a particular configuration of the PV array.

Within any array configuration, any of the above routines, such as: perturb and observe, incremental conductance, duty cycle sweep, could be conducted in order to arrive at the array's maximum power point. Any array configuration could be arrived at via a perturbation implemented by the controller as it seeks the optimum array configuration and duty cycle associated with the system's operation. Figure 22 shows an elaboration of the look-up approach where first a parameter- based look-up routine is used to find a close-to-optimal configuration as described above, and then a local configuration perturb routine is run whereby the configuration is changed and the response in output power monitored. This controller can for example use look-up to determine an optimum PV array configuration, and then determine an optimum duty cycle (and hence the optimum DC-DC conversion level) of the DC-DC converter with a configuration perturbation routine.

Figure 23 shows in more detail a reference map of optimum operating configurations in dependence on the two parameters of terminal voltage of the battery and solar intensity. Four different zones are marked for different optimum operating configurations:

• 13x10 configuration with DC-DC conversion

• 13x10 configuration without DC-DC conversion (direct connection bypassing DC-DC converter)

• 10x13 configuration with DC-DC conversion

• 10x13 configuration without DC-DC conversion (direct connection bypassing DC-DC converter)

A battery management system is now described in more detail with reference to Figure 24.

Conventionally a passive balancing scheme, where excess charge is dissipated through resistors, is often used in battery systems (composed of a number of individual battery cells) in order to avoid damage to the battery cells. Instead of (or in addition to) passive balancing the battery management system actively maintains balance. This is achieved by an active balance controller of the battery management system that controls the cell feed selectors and cell tap selectors (as described above with reference to Figure 4), with respect to both the power input to the battery from the PV device, and the power drawn from the battery by the load. The active balance controller shifts the loads and PV power injection in order to address and correct an imbalance when individual battery cells approach overcharging or undercharging. By such active balance management losses associated with passive, resistive heating in conventional passive balancing can be avoided or reduced.

Figure 24 shows an example of an active balance controller for the battery management system. In this example balance is sought first through load and/or PV control before resorting to a passive resistive scheme as a final resort. Selected battery cell nodes are monitored for cell voltage and cell temperature as indicators of under- or overcharging. If a battery cell approaches under- or overcharging then cell feed points and/or cell tap points are changed. Only if this fails to correct the under- or overcharging is passive balancing implemented. Alternatively to a system where a re-configurable PV array is directly connected to the terminals of a battery as described above, a re-configurable array can be connected to an AC network such as the grid or a local, off-grid electrical network. In a typical conventional solar photovoltaic installation inverters convert the (variable) DC PV output into an AC utility frequency to provide power from a solar photovoltaic unit to the grid. Inverters used in solar PV installations are a well-known technology; however they generally suffer from a decreased efficiency when operating away from their optimal efficiency point, for example with efficiency decreasing significantly when operated as powers below 20% of the rated PV installation output power. In addition to the power electronics of the inverter, many solar PV installations require that the inverter be connected to the grid via a transformer. Such a transformer can introduce further losses that can decrease the overall efficiency of the system.

Figures 25 and 26 show a two-array example of an application of a reconfigurable array to improve the yield of a conventional solar installation. Figure 25 shows a configuration for normal operation where there is a large irradiance on the solar installation. In this configuration both Inverters 1 and 2 are in operation and provide power to the grid. Figure 26 shows a configuration for operation where the irradiance is decreased. In this configuration switches Sp and Sn can switch such that Array 1 and Array 2 now become one large array. Due to the decrease in irradiance, a single inverter can be used to process all of the power, as in the example illustrated in Figure 26 Inverter 1 . When only Inverter 1 is operating, Inverter 2 can be switched off, and Transformer 2 disconnected from the grid. This provides the following efficiency advantages:

· Inverter 1 now operates at a higher efficiency operating point

The grid connected transformer, Transformer 2 is no longer producing any losses

Conversely, this system has a few disadvantages, namely:

It may be more difficult to track the MPPT in the new larger array

· The extra wiring and switches Sp and Sn can increase the cost of the system.

This reconfigurable strategy is applicable to any number of arrays and inverters. Additional switches can be included such that switching load sharing can be alternated on Inverters one or two.

Now an approach to improve the efficiency of resonant converters is described. A reconfigurable converter can improve the efficiency of resonant converters by extending the load range in which they operate with zero voltage switching (ZVS), as in the case of MOSFETs, or zero current switching (ZCS), as in the case of bipolar junction transistors. The extension of the ZVS and ZCS load range is done by reconfiguring the converter during normal operation.

Figure 27 shows a conventional full bridge resonant converter. The resonant tank consists of an arrangement of capacitors and inductors. This arrangement defines the resonant frequency of the converter. Typically resonant converters operate most efficiency at, or near, their resonant frequency. As the load R_L varies, the frequency at which the switches S1 , S2, S3 and S4 of the full bridge are driven changes. As a consequence, the operating point is no longer at its optimal point, and the efficiency drops.

A reconfigurable converter that addresses these issues is shown in Figure 28. Compared to Figure 27, switches S5 and S6 are added, as well as another inductor, LR2. During normal operation, switches S1 , S2, S3 and S4 are gating, and switches S5 and S6 are off. The first resonant tank is formed with the transformer, LR1 and CR. When the load, R_L, changes to a point where the first resonant tank is longer operating in an efficient manner, the converter can switch to the second resonant tank formed by switches S1 , S2, S5 and S6. In this second resonant tank, the transformer, LR2 and CR are used. A different resonant frequency is defined with these components. The first and second resonant tanks can be designed such that their optimal operating points are different. In this way, the efficiency of the converter can be increased over a wide load range. The converter can be adapted in dependence on the converter load, the converter input power, the converter output voltage or current, and/or the converter input voltage or current.

The technique is not limited to the resonant converter topology, but can be applied to different power converter topologies in order to increase converter efficiency.

The described reconfigurable converter, especially the resonant reconfigurable converter, requires bidirectional blocking switches for switches S3, S4, S5 and S6. Therefore, conventional MOSFETs are insufficient. While only two configurations are described above and shown in Figure 28, any number of the number of operation-optimising configurations can be implemented.

It will be understood that the present invention has been described above purely by way of example, and modifications of detail can be made within the scope of the invention.

Each feature disclosed in the description, and (where appropriate) the claims and drawings may be provided independently or in any appropriate combination.

Reference numerals appearing in the claims are by way of illustration only and shall have no limiting effect on the scope of the claims.

Claims

1 . A system comprising:

a reconfigurable array of electricity generating cells for charging a battery; an array switching means for dynamically reconfiguring the array;

a means for DC-DC conversion of an array output voltage; and

a means for altering the DC-DC conversion;

wherein both the array switching means for array configuration and the means for altering the DC-DC conversion are modified in dependence on a battery terminal voltage.

2. A system according to Claim 1 , further comprising a controller adapted to: determine an array configuration and a DC-DC conversion in dependence on a measured battery terminal voltage; and

modify the array switching means for array reconfiguration and the means for altering the DC-DC conversion in dependence on the measured battery terminal voltage in order to implement the determined array configuration and DC-DC conversion.

3. A system according to Claim 1 or 2, wherein the means for altering the DC- DC conversion comprises a DC-DC conversion bypass switch for bypassing the means for DC-DC conversion in order to provide electricity from the electricity generating cells directly to the battery.

4. A system according to Claim 3, wherein the or a controller is adapted to: determine a DC-DC conversion bypass switch configuration in dependence on the or a DC-DC conversion; and

modify the means for altering the DC-DC conversion in dependence on the or a DC-DC conversion in order to implement the determined DC- DC conversion bypass switch configuration.

5. A system according to Claim 2 or 4, wherein the controller is adapted to determine the array configuration and the DC-DC conversion and/or the or a DC-DC conversion bypass switch configuration further in dependence on a level of irradiance, a distribution of irradiance and/or an ambient temperature.

6. A system according to Claim 5, wherein a measured open circuit voltage or a measured short circuit current of the array is used to quantify the level of irradiance, the distribution of irradiance and/or the ambient temperature.

7. A system according to any of Claims 2 or 4 to 6, wherein the controller determines an array configuration and a DC-DC conversion and/or the or a DC-DC conversion bypass switch configuration by a look-up routine.

8. A system according to any of Claims 2 or 4 to 7, wherein the controller determines an array configuration and a DC-DC conversion by a machine learning routine.

9. A system according to Claim 8, wherein the machine learning routine is a fuzzy logic machine learning routine.

10. A system according to Claim 8 or 9, wherein the machine learning routine is trained with a power output to the battery as feedback.

1 1 . A system according to any of Claim 2 or 4 to 10, wherein the controller is adapted to determine the array configuration and the DC-DC conversion by a configuration perturbation routine comprising steps of changing the array configuration and/or the DC-DC conversion, and comparing a power output to the battery before and after the change.

12. A system according to Claim 1 1 , wherein the configuration perturbation routine comprises an array configuration and/or DC-DC conversion sweep.

13. A system according to any preceding claim, wherein the array switching means for array reconfiguration is a remotely controllable switch.

14. A system according to any preceding claim, further comprising a battery for being charged by the reconfigurable array of electricity generating cells.

15. A system according to any preceding claim, further comprising a battery comprising a plurality of battery cells and a battery switching means for dynamically reconfiguring the plurality of battery cells, wherein the array switching means and the battery switching means are dynamically modified to selectively connect a sub-array of electricity generating cells to a subgroup of battery cells.

16. A system comprising:

an array switching means for dynamically reconfiguring a reconfigurable array of electricity generating cells for charging a battery with a plurality of battery cells; and

a battery switching means for dynamically reconfiguring the plurality of battery cells;

wherein the array switching means and the battery switching means are dynamically modified to selectively connect a sub-array of electricity generating cells to a sub-group of battery cells.

17. A system according to Claim 15 or 16, wherein the or a controller is adapted to:

determine an array configuration comprising a sub-array and a battery configuration comprising a sub-group; and

control the array switching means and the battery switching means in order to implement the determined array configuration and battery configuration.

18. A system according to any of Claims 15 to 17, wherein the battery configuration is determined in dependence on the determined array configuration.

19. A system according to Claim 17 or 18, wherein the controller is adapted to provide charge from a first sub-array to a first sub-group of battery cells at a first voltage, and provide charge from a second sub-array to a second subgroup of battery cells at a second voltage.

20. A system according to Claim 19, wherein the second sub-array comprises remainder cells to the first sub-array.

21 . A system according to Claim 19, wherein the second sub-array comprises underperforming cells.

22. A system according to Claim 21 , wherein the underperforming cells are damaged, shaded, and/or deteriorated cells.

23. A system according to any of Claims 15 to 22, wherein the battery switching means is a remotely controllable array of switching elements connected to nodes between battery cells.

24. A system according to any of Claims 15 to 23, wherein the or a controller is adapted to:

determine a sub-group of battery cells in dependence on a battery load; and

control the battery switching means in order to provide charge from the sub-group of battery cells to the battery load.

25. A system according to any of Claims 15 to 24, wherein the or a controller is adapted to provide charge to and/or draw charge from first a first sub-set of battery cells and then a second sub-set of battery cells in order to balance the charge across the battery cells.

26. A system according to any of Claims 15 to 25, further comprising a reconfigurable array of electricity generating cells for charging a battery.

27. A system according to any of Claims 15 to 26, further comprising a battery with a plurality of battery cells for being charged by the reconfigurable array of electricity generating cells.

28. A system according to any of Claims 15 to 27, wherein the battery switching means is dynamically adapted to provide charge to and/or draw charge from the battery by first providing a selective connection to a first sub-set of battery cells, and then providing a selective connection to a second sub-set of battery cells in order to balance the charge across the battery cells.

29. A system comprising:

a battery switching means for dynamically reconfiguring a plurality of battery cells in a battery;

wherein the battery switching means is dynamically adapted to provide charge to and/or draw charge from the battery by first providing a selective connection to a first sub-set of battery cells, and then providing a selective connection to a second sub-set of battery cells in order to balance the charge across the battery cells.

30. A system according to Claim 28 or 29, wherein the or a controller is adapted to:

determine an over- or undercharged sub-set of battery cells; and control the battery switching means in order to correct the over- or undercharging.

31 . A system according to any of Claim 28 to 30, wherein the battery switching means comprises an array of switching elements connected to nodes between battery cells for providing charge to and/or drawing charge from a sub-set of battery cells.

32. A system according to any of Claims 28 to 31 , further comprising a battery with a plurality of battery cells.

33. A system according to any of Claims 28 to 32, wherein the battery switching means is dynamically adapted to provide voltage from a sub-cluster of the battery cells directly to a transistor as a gate drive source.

34. A system comprising

a battery switching means for dynamically reconfiguring the plurality of battery cells in a battery;

wherein the battery switching means is dynamically adapted to provide voltage from a sub-cluster of the battery cells directly to a transistor as a gate drive source.

35. A system according to Claim 33 or 34, wherein the or a controller is adapted to:

determine a required gate drive source voltage; and

control the battery switching means in order to provide at least the required gate drive source voltage from the sub-cluster.

36. A system according to any of Claims 33 to 35, further comprising a battery with a plurality of battery cells.

37. A system comprising:

a reconfigurable array of electricity generating cells for providing electricity to a network;

an array switching means for dynamically reconfiguring the array;

a means for DC-AC conversion of an array output voltage; and

a means for altering the DC-AC conversion;

wherein both the array switching means for array configuration and the means for altering the DC-AC conversion are modified in dependence on an array output voltage.

38. A system comprising:

a reconfigurable converter for DC-DC conversion;

a converter switching means for dynamically reconfiguring the converter; wherein the converter switching means for converter reconfiguration is modified in dependence on a converter load.

39. A method of providing electricity from a reconfigurable array of electricity generating cells to a battery for charging comprising:

determining a battery terminal voltage;

40. A method of providing electricity from a reconfigurable array of electricity generating cells to a battery with a plurality of battery cells for charging comprising:

dynamically reconfiguring the plurality of battery cells; and

selectively connecting a sub-array of electricity generating cells to a subgroup of battery cells.

41 . A method of providing charge to and/or drawing charge from a battery with a plurality of battery cells comprising:

42. A method of providing charge from a battery with a plurality of battery cells to a transistor comprising:

43. A system substantially as described herein with reference to the accompanying drawings.

44. A method substantially as described herein with reference to the accompanying drawings.