WO2021123440A1

WO2021123440A1 - Power routing system

Info

Publication number: WO2021123440A1
Application number: PCT/EP2020/087494
Authority: WO
Inventors: Matthew Peter LOWE; Chung Mo; Andre Justin VAN'T HOFF; James CAMPBELL-FITZPATRICK; Nicholas John CARPENTER
Original assignee: Delta Motorsport Limited
Priority date: 2019-12-20
Filing date: 2020-12-21
Publication date: 2021-06-24
Also published as: GB201918939D0

Abstract

A power routing system configured to route electrical current from a power supply to an energy storage device having a plurality of energy storage substrings, the power routing system comprising: a power input connection for connection to a power supply; a plurality of power output connections for connection to respective energy storage substrings of the plurality of energy storage substrings; a plurality of switching means disposed between the power input connection and the plurality of power output connections; and a controller. The controller is configured to: i) receive a plurality of signals indicative of a level of charge in each of the plurality of energy storage substrings; and ii) control the plurality of switching means, based on the received plurality of signals, to selectively provide current from the power input connection to each of the plurality of energy storage substrings in turn, such that a difference in the level of charge between each of the plurality of energy storage substrings is below a threshold value.

Description

Power routing system

Field of Invention

The present invention relates to power routing systems for use when charging rechargeable energy storage systems and methods of operating power routing systems, for example power routing systems for charging electric vehicle batteries.

Background

The growth in demand for low carbon vehicles has seen a steady rise in demand for Battery Electric Vehicles (BEVs) and Plug in Hybrid Electric Vehicles (PHEVs). In order to achieve equivalent or greater performance to Internal Combustion Engine vehicles the power of electric vehicles (e.g. the capacity and output voltage of electric vehicle energy storage systems) has been increasing. To increase electrical power requires either an increase in current, an increase in battery voltage, or both.

Increasing the current to achieve greater power requires larger conductors in the current path connecting the Energy Storage System (ESS) to the traction system. The conductors are usually made of copper and an increase in their size adds significant cost and weight to the vehicle, while the associated I²R power losses generate heat that must be dissipated by the vehicle's design. Therefore the effect of increasing power by increasing current is to reduce system efficiency and increase costs.

The preferred alternative is to increase battery voltage to increase power whilst maintaining conductor size and system efficiency. It is expected that vehicle EES voltages will increase from around 250-450 to around 750-800V in the coming years, and possibly increase to even higher voltages in the future. The industry refers to nominal 400V and 800V charging which can span the range of around 280-420V and around 600-900V respectively.

However, currently available DC electric vehicle supply equipment (EVSE, also referred to as charging stations) are typically limited to 500V DC output. Moreover, it is expected that the upgrades to the public charging infrastructure are likely to lag behind the pace of development of EV powertrains both now and in the future as both traction and charge power requirements increase leading to higher system voltage. Accordingly, there is a need to provide means for charging a higher voltage ESS using a lower voltage charging station.

The traditional solution to boosting the DC voltage is to use a DCDC Converter. This typically requires the incoming lower voltage DC supply to be converted to an AC wave form and then input to the primary windings of a transformer where the AC voltage is boosted at the secondary winding. The boosted AC voltage is then input to a full bridge rectifier and then smoothed using LC filters to output a ripple free DC charge supply. Alternatively, active rectification can be used. However, there are inefficiencies during the conversion process that can result in significant heat generation.

An alternative solution to boosting DC voltage, known from the field of solar charging of vehicle batteries, is to configure an EES such that a power supply can be connected to one or more individual cells in a battery independently, as described in patent application US3008/0143292 Al. This allows a solar panel, whose output voltage is less than the total battery voltage, to charge individual cells/groups of cells in isolation. In this case, the solar panel is electrically connected to each of the cells/groups of cells in sequence, thereby fully charging each of the cells/groups of cells one by one. However, such a charging regime introduces a voltage difference between individual cells. This voltage imbalance is detrimental to performance of the battery pack, and limits the overall output voltage available from the battery pack.

A further alternative solution is described in US 2018/0272883 Al, in which sections of a battery are connected in series during discharge of the battery, but in parallel during charging of the battery. While this allows a lower voltage power source to charge a higher output voltage battery, it does not allow vehicle systems to operate using the full battery voltage while the segments are connected in parallel. Thus systems requiring full battery voltage cannot operate during charging.

Whilst the above discusses the need to charge a vehicular EES from a DC charging station, it will be appreciated that there are other contexts in which it is desirable to charge higher voltage energy storage systems from lower voltage power supplies. There is thus a need for an efficient and low cost method of charging an ESS from a lower voltage DC power supply.

Summary of Invention

In order to at least partially address the issues noted above, the present invention provides a power routing system configured to route electrical current from a power supply to an energy storage device having a plurality of energy storage substrings (that is, portions/segments of a series string of energy storage cells/modules/devices), a method of operating a power routing system, a vehicle comprising a power routing system, and a rechargeable energy storage system comprising a power routing system.

In one aspect of the invention, a power routing system (or charge arbitration device) comprises: a power input connection for connection to a power supply (for example a connection for EVSE); a plurality of power output connections for connection to respective energy storage substrings of the plurality of energy storage substrings (in some embodiments, more than two such substrings); a plurality of switching means (such as transistor switches) disposed between the power input connection and the plurality of power output connections; and a controller. The controller is configured to: i) receive a plurality of signals (e.g. from a battery management system, a vehicle bus, and/or one or more sensors) indicative of a level of charge or energy (also referred to as the "state of charge") in each of the plurality of energy storage substrings; and ii) control the plurality of switching means, based on the received plurality of signals, to selectively provide current from the power input connection to each of the plurality of energy storage substrings (i.e. provide current to each energy storage substring individually/separately/one at a time), such that a difference in the level of charge between each of the plurality of energy storage substrings remains below a threshold value/difference.

For the purposes of this disclosure, "level of charge" can refer to either: an absolute amount of charge stored in a respective substring; or an amount of charge/energy stored in a respective substring as a fraction of a maximum amount of charge/energy capable of being stored in the respective substring. For instance, if the plurality of substrings all have the same or similar capacity/output voltage (e.g. the substrings are capable of storing the same/similar maximum amount of charge), then the controller may ensure that a difference in the absolute amount of charge stored in each substring is kept below the threshold value. Alternatively, if the capacity/output voltage of the plurality of substrings is different, the controller optionally ensures that a difference in an amount of charge as a fraction of the maximum possible stored charge between each of the plurality of energy storage substrings remains below a threshold value instead.

The energy storage substrings are supplied with current individually either in a predetermined order/sequence, in a dynamically determined order (e.g. the energy storage substring currently having the lowest level of charge is selected next), or in a combination of the two. For example, respective amounts of charge can be applied to each of the energy storage substrings in turn according to a predefined/dynamically defined sequence.

Beneficially, this arrangement not only permits the charging of a higher voltage/capacity energy storage system using a lower voltage power supply, but does so in a manner that improves battery performance (for example improving range in the context of BEVs and PHEVs) in a fast and effective manner. This means of charging ensures that the level of charge in each substring is brought to/remains at a similar level (i.e. within the level of charge threshold difference) throughout the charging process. The output of the energy storage system is typically limited by the lowest state of charge substring. Accordingly by progressively charging each substring in turn, and keeping the level of charge in different substrings similar, the arrangement effectively improves battery performance (e.g. BEV/PHEV range) quickly as compared to other methods for charging subsections of an energy storage device. Additionally, if the charging cycle is ended before the energy storage system has reached its maximum charge, the available output performance/ range is improved, as the lowest state of charge substring has a similar state of charge to the substring with the highest level of charge.

Moreover, by selectively charging each substring individually such that the voltage in each substring can be kept at similar levels, undesired in-rush currents between other connected components can be reduced during switch over events (i.e. when disconnecting one substring from the power input connection and connecting another substring to the power input connection). In a preferred embodiment, the controller is configured to: identify, prior to step ii) and based on the received signals, which of the plurality of energy storage substrings has a lowest level of charge; and control the plurality of switching means to selectively provide current from the power input connection to the identified energy storage substring prior to selectively providing current from the power input connection to any other of the plurality of energy storage substrings. Thus the energy storage substring with the lowest charge initially is provided with charging current first. By selectively charging the substring having the lowest charge/voltage level first, the power routing system improves the balance of the energy storage system and quickly improves output performance (e.g. BEV/PHEV range). Optionally the controller is also configured to identify a first difference in level of charge between the identified energy storage substring and another energy storage substring; and if the first difference in level of charge is greater than the threshold value: control the plurality of switching means to selectively provide current from the power input connection to the identified energy storage substring until the first difference in level of charge is less than the threshold value, prior to selectively providing current from the power input connection to any other of the plurality of energy storage substrings. In other words, if the energy storage modules are out of balance/have a difference in levels of charge exceeding the threshold value initially, the lowest level of charge energy storage substring is charged first so as to reduce the difference to less than the threshold value. Advantageously, in the case that substrings are unbalanced, i.e. have levels of charge that are different (e.g. exceed the threshold level of charge difference), this allows the level of charge in the substrings to be brought close together (e.g. within the level of charge threshold difference) and then remain close together throughout charging, again improving energy storage output performance quickly and effectively.

In some embodiments, the controller is configured to control the plurality of switching means to selectively provide current from the power input connection to each of the plurality of energy storage substrings in turn such that: a predetermined amount of charge/energy is provided to one or more (e.g. each) of the plurality of energy storage substrings (e.g. increments of 1%, 5% 10%, etc. of the maximum level of charge of the substring/energy storage system); the level of charge in one or more (e.g. each) of the plurality of energy storage substrings reaches a predetermined target (e.g. charging until the level of charge of the substring reaches a specific fraction of the maximum level of charge of the substring); and/or the level of charge in one or more (e.g. each) of the plurality of energy storage substrings reaches a dynamically determined target, wherein the dynamically determined target is based on (e.g. equal to) a level of charge of an energy storage substring having a highest level of charge.

In some embodiments, the controller is configured to: control the plurality of switching means to selectively provide current from the power input connection to the identified energy storage substring prior to selectively providing current from the power input connection to any other of the plurality of energy storage substrings until the level of charge in a first energy storage substring (e.g. the energy storage substring that presently has the lowest level of charge) reaches the dynamically determined target. Thus respective amounts of charge can be applied to each of the energy storage substrings in turn according to a dynamically defined sequence. Advantageously, by selecting the energy storage substring presently having the lowest level of charge for providing with charging current next, the system dynamically ensures that battery performance/vehicle range is improved/optimised as the charging routine progresses.

In one embodiment, the controller is configured to: subsequent to selectively providing current to a first energy storage substring and prior to selectively providing current to a second energy storage substring: selectively control current flow caused by a difference in stored charge in the first energy storage substring and the second energy storage substring, by using a pulse width modulated signal to operate the switching means (i.e. by using a pulse width modulated signal to operate the switching means) so as to provide an effective resistive load between the power input connection and the second energy storage substring. For example, the controller can use "soft switching" techniques to control in-rush currents between components in the wider system in which the power routing device is deployed caused by the substrings having a different level of charge when one substring is being disconnected from the power input connection and the next substring is being connected. Accordingly this further reduces the effects of in-rush currents.

Preferably the controller is configured to repeat step ii), that is repeatedly selectively charge the individual substrings, for example until the charging cycle is ended (e.g. by a user) or until all substrings have reached a target (e.g. maximum) stored energy level. For example, respective amounts of charge can be repeatedly applied to each of the energy storage substrings in turn according to a predefined/dynamically defined sequence. Optionally, the controller is further configured to control the switching means to alternately provide current from the power input connection to respective energy storage substrings of the plurality of energy storage substrings at a rate equal to or exceeding a threshold frequency. Put differently, the controller is configured to switch between charging individual substrings at a frequency at or above a threshold frequency (e.g. 1 Hz). Advantageously, this prevents an EVSE from detecting a discontinuity in current flow greater than a predetermined tolerance, and thus allows current to be provided to respective substrings in turn without ramping down the charging current demanded form the EVSE. Thus the overall time taken for charging to take place is reduced.

In another aspect of the invention, there is provided rechargeable energy storage system comprising the power routing system and the energy storage device as mentioned above.

In a further aspect of the invention there is provided a vehicle comprising either the power routing system or the rechargeable energy storage system as mentioned above.

In an additional aspect of the invention, there is provided a method for charging an energy storage device having a plurality of energy storage substrings comprising: i) receiving a plurality of signals indicative of a level of charge in each of the plurality of energy storage substrings; and ii) selectively providing current from a power input connection to each of the plurality of energy storage substrings, such that a difference in the level of charge between each of the plurality of energy storage substrings remains below a threshold value.

In one embodiment, the method comprises: identifying, prior to ii) and based on the received signals, which of the plurality of energy storage substrings has a lowest level of charge; selectively providing current from the power input connection to the identified energy storage substring prior to selectively providing current from the power input connection to any other of the plurality of energy storage substrings. Optionally, the method further comprises prior to step ii): identifying a first difference in level of charge between the identified energy storage substring and another energy storage substring; and if the first difference in level of charge is greater than the threshold value: selectively providing current from the power input connection to the identified energy storage substring until the first difference in level of charge is less than the threshold value, prior to selectively providing current from the power input connection to any other of the plurality of energy storage substrings.

Preferably the method comprises, at step ii), selectively providing current from the power input connection to each of the plurality of energy storage substrings in turn such that: a predetermined amount of charge/energy is provided to each of the plurality of energy storage substrings; the level of charge in each of the plurality of energy storage substrings reaches a predetermined target; and/or the level of charge in each of the plurality of energy storage substrings reaches a dynamically determined target, wherein the dynamically determined target is based on a level of charge of an energy storage substring having a highest level of charge.

In some embodiments the method comprises, at step ii): selectively providing current from the power input connection to a first energy storage substring prior to selectively providing current from the power input connection to any other of the plurality of energy storage substrings until the level of charge in the first energy storage substring reaches the dynamically determined target.

In some embodiments, the method comprises: subsequent to selectively providing current to a first energy storage substring and prior to selectively providing current to a second energy storage substring: selectively controlling current flow caused by a difference in stored charge in the first energy storage substring and the second energy storage substring, by using a pulse width modulated signal to operate switching means so as to provide an effective resistive load between the power input connection and the second energy storage substring.

Preferably the method comprises repeating step ii). Optionally, the method further comprises alternately providing current from the power input connection to respective energy storage substrings of the plurality of energy storage substrings at a rate equal to or exceeding a threshold frequency. Brief Description of Drawings

Embodiments of the present invention will now be described, by way of example only, with reference to the following figures in which: Figure 1 shows a schematic circuit diagram of a rechargeable energy storage system in accordance with the present invention;

Figures 2A and 2B show schematic representations of an exemplary vehicle including a rechargeable energy storage system in accordance with the present invention;

Figure 3 shows a flow diagram illustrating a method of operating a power routing system in accordance with the present invention; Figures 4A, 4B, 4C, 4D and 4E show graphs representing exemplary charging routines;

Figure 5 shows a schematic circuit diagram of a power routing system in accordance with the present invention;

Figure 6 shows an exemplary process for performing charging steps;

Figure 7 shows graphs representing a further exemplary charging routine; Figure 8 shows a further exemplary process for performing charging steps.

Detailed Description

The specific embodiments discussed below are described in the context of charging automotive energy storage systems, however it will be appreciated that the claimed invention is equally applicable to charging energy storage systems in other contexts (for example charging energy storage systems in other types of vehicle, domestic energy storage systems and industrial energy storage systems).

Figure 1 shows a schematic circuit diagram of a rechargeable energy storage system (RESS) 100 in accordance with the present invention. The RESS 100 comprises an energy storage device 102, the energy storage device 102 is a series string of energy storage elements/modules (for example a series string of individual cells 106 as shown in figure 1, or alternatively a series string of groups/modules, each module/group comprising a plurality of cells connected in series or parallel). The energy storage device comprises at least two energy storage substrings 104a, 104b (portions/segments of the string of series connected energy storage elements/modules).

Each energy storage substring 104a, 104b comprises one or more electrical energy storage elements/modules. As shown in figure 1, each energy storage substring 104a, 104b comprises a battery of series connected electrochemical cells 106, however in other embodiments, the energy storage device may comprise one or more capacitors, or other electrical energy storage means. Where two or more energy storage means are included in a substring 104a, 104b, the energy storage means may be connected in series or parallel within the substring 104a, 104b. Two energy storage substrings 104a, 104b are shown connected in series in figure 1 - alternatively more than two energy storage substrings can be provided in series. Preferably each energy storage substring 104a, 104b is configured to make up 100/N% of the total stored energy with the RESS 100. Accordingly, each energy storage substring 104a, 104b preferably has the same nominal output voltage/ capacity, though it will be appreciated that the system can be adapted to utilise substrings 104a, 104b having different nominal output voltages/capacities.

The energy storage device 102 comprises at least three connection points 108a, 108b, 108c facilitating independent connection to each of the energy storage substrings. The number of connections points 108a, 108b, 108c is equal to the number of energy storage substrings 104a, 104b plus one.

Optionally, the energy storage system 102 also comprises a battery management system (BMS) 109 or other suitable electronics module/modules, configured to monitor the energy storage means within the substrings 104a, 104b (e.g. monitor each of the cells 106). The BMS 109 may be of a known type (battery management systems are known for use in lithium-ion battery packs and other energy storage technologies), which may comprise of several electrical/electronics devices and is primarily responsible for monitoring the voltage of each series cell or groups of cells in a battery module, temperatures of cells or critical components and broadcasting safety limits based on internal calculations/models. It may also have control over isolating contactors within the RESS and a method for measuring electrical isolation between the high voltage and low voltage systems. For example, the BMS 109 may monitor the voltage, current, and/or temperature of each of the cells 106 (and/or of each of the substrings 104a, 104b and/or of the energy storage system 102 as a whole) over time.

The energy storage device 102 further comprises an output connection 110 for connection to the desired load. In the case of an automotive vehicle output connection 110 may be connected to one or more electric motors and/or other vehicular electronic systems. As shown in figure 1, the output connection 110 enables connection to the full output voltage of the energy storage system 102 (that is the total voltage of both the substrings 104a, 104b combined). However further output connections (not shown) allowing connection to less than the total number of substrings 104a, 104b (e.g. to one substring 104a, 104b) may also be provided. For example, in the context of a BEV or PHEV, an electric drivetrain may be connected across all of the substrings 104a, 104b via one output connection 110 to access the full output voltage of the energy storage system 102 (e.g. 800V), whereas other ancillary systems (e.g. DC/DC converters, HVAC systems etc.) are connected across one substring 104a, 104b if they require a lower voltage power source (e.g. 400V).

The RESS 100 further comprises a power routing system 120 (which can also be referred to as a charge arbitration system), configured to route charging current from a power supply (such as EVSE) to different energy storage substrings 104a, 104b selectively. As shown in figure 1, the power routing system 120 is provided with the energy storage device 102 as a single unit. Alternatively the power routing system can be provided as separate unit or units for connecting to the energy storage device 102.

The power routing system comprises a power input connection 122 for connection to a suitable power supply. For example, when RESS 100 is employed in a battery electric vehicle or plug in hybrid electric vehicle, the power supply may be a charging station conforming to a charging standard, for example the Combined Charging System (CCS) or CHAdeMO standards. In this case the power input connection 122 is chosen for compatibility with the relevant charging station.

The power routing system comprises a plurality of power output connections 124a, 124b, 124c facilitating connection to each of the energy storage substrings 104a, 104b. Preferably a respective power output connection 124a, 124b, 124c is provided for connecting to each of the connection points 108a, 108b, 108c in the energy storage device 102.

The power routing system comprises a plurality of switching means 126a, 126b. The switching means are located such that current from the power supply (via the power input connection 122) can be supplied to one or more of the energy storage substrings 104a, 104b selectively. Preferably each switching means 126a, 126b comprises a transistor switch.

As shown in figure 1, there are provided two switching means 126a, 126b (though further switching means can be provided in the event that more than two energy storage substrings 104a 104b are present in the energy storage device 102). Each switching means 126a, 126b is connected to: a central power output connection 124b; a respective end power output connection 108a, 108b; and a respective terminal of the power input connection 122. The switching means 126a, 126b are in communication with a controller 128, wherein the controller 128 is configured to control operation of the switching means 126a, 126b as is described in more detail below. The controller 128 is preferably configured to communicate with the BMS 109 (if provided).

As shown in figure 1, the power routing system 120 (when configured for use with two energy storage substrings 104a, 104b), the power input connection 122, the switching means 126a, 126b, and the power output connections 124a, 124b, and 124c can be arranged as a H-bridge circuit. This is advantageous, since suitable H-bridge circuits are widely available commercially for a variety of purposes. Thus the arrangement of figure 1 can be manufactured at relatively low cost.

A plurality of diodes 130a, 130b, 130c, 130d are preferably provided in combination with the switching means 126a and 126b in order to supply current to one or more of the energy storage substrings 104a, 104b selectively.

In use, when the power input connection 122 is connected to a power supply (e.g. an electric vehicle charging station), the power routing system 120 is configured to selectively provide current from the power supply to each energy storage substring 104a, 104b independently. For example, by opening switch 126a and closing switch 126b, current is permitted to flow through energy storage substring 104a via power output connections 124a and 124b, whilst substantially no current is provided to energy storage substring 104b. Similarly, by closing switch 126a and opening switch 126b, current is permitted to flow through energy storage substring 104b via power output connections 124b and 124c, whilst substantially no current is provided to energy storage substring 104a.

Advantageously, this arrangement allows for a given power supply/charging station having a certain output voltage to charge a battery with a higher nominal output voltage, without the use of a DC/DC converter. This in turn allows for higher voltage energy storage means to be used, for example in BEVs and PHEVs, whilst enabling compatibility with existing, lower voltage, charging infrastructure.

As an example, an electric vehicle charging station having an output of 400V can be used to charge an 800V vehicle battery by implementing the present invention. In particular the electric vehicle 800V vehicle battery may be configured as two series connected 400V energy storage substrings, connected to a power routing system as described above. Similarly, an electric vehicle charging station having an output of 400V can be used to charge vehicle battery with a 1200V or 1600V output, by providing three or four series connected 400V substrings respectively connected to a suitable power routing system.

Figure 2A shows an exemplary vehicle 200 including the RESS 100 as described above in relation to figure 1.

Figure 2B shows a top-down schematic representation of the exemplary vehicle 200, when connected to a charging station (e.g. EVSE) 210. The vehicle 200 includes electrical systems powered by the RESS (via a high voltage bus 205), for example an electric drivetrain 204. Various electronic components 208 are connected to one or more vehicle busses 206 (e.g. a CAN bus, Ethernet, flexray and/or optical busses) for exchange of information, for example a DCDC converter, an HVAC heater and/or compressor, and/or an on-board charger (AC/DC). During charging, the RESS 100 can be electrically connected to a charging station 210 for charging a battery or other energy storage system having a greater nominal output voltage that the output voltage of the charging station 210. The vehicle 200 as shown is provided with two powered wheels 212, 214, however it will be appreciated that the present invention may be implemented for a vehicle having any number of powered wheels, such as four powered wheels. Optionally the RESS 100 is configured to be charged by inductive charging (e.g. via an interface with an inductive charger included in electronic components 208) while the vehicle 200 is moving.

Figure 3 shows a flow diagram illustrating a method 300 of operating the power routing system 120 in accordance with the present invention.

Method 300 begins at step S301, when a power supply is connected to the power input connection 122. In a preferred embodiment, the method 300 is started in response to the controller 128 detecting that the power routing system 120 has been connected to the power supply, for example via signals received from sensors (not shown) indicative of connection to the power supply, or via a signal received by the controller 128 from the power supply (e.g. an electric vehicle charging station/EVSE).

At step S302, the controller 128 receives signals indicative of a level of charge (also referred to as a state of charge or SoC) in each of the energy storage substrings 104a, 104b. In the preferred embodiment, the signals are generated by BMS 109, which monitors quantities such as the voltage, temperature and current for each substring 104a, 104b over time, and estimates or calculates a level of charge (or otherwise infers a level of charge) for each of the substrings 104a, 104b based on the monitored quantities. The calculated/inferred level of charge may further be based on a predetermined model, for example a model describing the properties of the substrings during charging. In other embodiments, signals may be produced by appropriately placed sensors (e.g. voltage measurement devices) in the power routing system and/or the energy storage system 102 (sensors not shown). Alternatively, the controller 128 is provided with a connection to a vehicle CAN (or Ethernet, flexray or optical) bus, and the signals are provided via the bus. In a less preferred embodiment, the signals may simply be voltages measured across the substrings 104a, 104b, which are used as a rough proxy for the level of charge.

At step S304, the controller 128 then determines which of the energy storage substrings 104a, 104b to start charging first. Optionally, in this arbitration step the controller determines which substring 104a, 104b has the lowest level of charge based on the signals indicative of the level of charge in each substring 104a, 104b, and identifies the substring 104a, 104b having the lowest level of charge for charging first.

Advantageously, by selectively charging the energy storage substring 104a, 104b with the lowest level of charge initially, the method allows for quickly achieving large gains in output performance of the energy storage system 102. The output performance of a battery of series connected energy storage substrings is typically limited by the substring having the lowest level of charge. By allowing individual energy storage substrings 104a, 104b to be charged individually, and by selecting the substring having the lowest level of charge for charging first, rapid gains in over battery performance can be achieved. In the context of BEVs and PHEVs, quickly increasing the overall performance of the energy storage system 102 in this manner means quickly increasing the range of the vehicle (i.e. the total distance the vehicle can travel using energy from the energy storage system). Thus the claimed methodology results in making large gains to vehicle range in a short amount of time. For example in the case of the RESS 100 comprising two substrings 104a, 104b, range can be added to the vehicle at an effective rate of twice the rate of the charging current (e.g. if range can generally be added to the vehicle at a rate of 100 miles per unit time period on a lOOkW charger, then during the initial phase when the substring with the lowest level of charge is connected to the EVSE, range is effectively added to the vehicle at a rate of 200 miles per unit time until the two half packs reach an equal level of charge).

Alternatively, the substring for starting charging first is selected based on other factors at arbitration/selection step S304.

At step S306, the controller 128 controls the switching means 126a, 126b so as to selectively route current from the power supply to the energy storage substring selected/identified for charging first. For example, in the context of BEVs/PHEVs, once the identified substring 104a, 104b has been connected to an EVSE via the switching means 126a, 126b, the EVSE matches the voltage of the connected substring 104a. The BMS 109 is preferably configured to transmit a signal indicative of the maximum current supported by the connected substring 104a, 104b to the EVSE, either directly or via the controller 128. The EVSE then raises its voltage appropriately to drive the amount of current into the vehicle in a known manner. The EVSE will then provide a current equal to, or lower than the maximum current indicated by the BMS 109 in a known manner.

In a preferred embodiment, the controller 128 is configured to monitor the level of charge provided to/stored in each of the substrings 104a, 104b throughout the charging process. For example, the controller 128 can interrogate BMS 109, suitable sensors and/or vehicle busses 206 to obtain one or more of the voltage across, the temperature of, and the current through each of the substrings 104a, 104b. In the preferred embodiment, the BMS 109 and/or the controller 128 calculates an amount of charge or energy provided to the substring 104a, 104b based on the magnitude of the current provided to the substring 104a, 104b by the power supply and the amount of time during which the current has been provided.

The strategy employed for charging the energy storage system 102 preferably aims to transfer a certain quantity of energy to each substring 104a, 104b. The controller 128 is preferably configured to direct current to the identified energy storage substring 104a, 104b until a target amount of charge (or energy) has been provided to energy storage substring 104a, 104b (or alternatively until a target level of charge/energy in the energy storage substring 104a, 104b has been reached). Preferably the BMS 109 monitors the current though each substring 104a, 104b and the amount of time current is provided to each substring 104a, 104b, and from this the BMS 109 and/or the controller 128 can determine the amount of charge that is stored in and/or has been provided to each substring 104a, 104b. This "coulomb counting" technique is advantageous in that it allows for effective measurement of charge/energy provided to each substring, even if the substrings have unequal capacities. Alternatively (or in addition), the BMS 109 and/or controller 128 monitors the voltage across each substring 104a, 104b as a proxy for the level of charge in respective substrings 104a, 104b.

In one example, the controller 128 is configured to provide a predetermined amount of charge/energy to the connected substring 104a, 104b. For example, the BMS 109 may request a certain current from the power supply (e.g. EVSE) for charging a connected substring 104a, 104b, and the controller 128 is configured to direct that current to the connected substring for a predetermined time, corresponding to a predetermined amount of charge/energy being provided to the connected substring 104a, 104b. The amount of time for which current is provided is optionally determined based on: a) the current demanded by the BMS 109; and b) a slew rate of the power supply (e.g. EVSE), put differently the rate at which the power supply is able to ramp up and ramp down current. For example, because of the intrinsic slew rate of an EVSE, switching between substrings 104a, 104b, may add to the overall time taken to fully charge the energy storage system 102. Thus the time between switching events may be a trade-off between reducing the over time to fully charge the energy storage system 102 (i.e. reduced the number of switching events) and keeping the substrings 104a, 104b at a similar voltage (i.e. increasing the number of switching events). The exact rate of switching will depend on the specific application of the RESS 100.

In another example, the target level of charge may be an absolute predetermined value, for example corresponding to a fixed percentage of the total maximum level of charge of the energy storage substring 104a, 104b, or the total maximum level of charge of the energy storage system 102. For example, the charge in the substring 104a, 104b being charged may be increased by an amount corresponding to 1%, 5%, 10%, etc. of the total maximum level of charge of the energy storage system 102.

In one embodiment, the amount of charge (or energy) provided to each substring is dynamically determined. In particular, controller 128 is configured to first provide a first amount of charge/energy to the substring 104a, 104b having the lowest level of charge (e.g. the lowest amount of charge/energy as a fraction of the maximum amount of charge/energy that can be stored in the substring 104a, 104b), wherein the first amount of charge/energy brings the level of charge in the connected substring to the same or similar level as in the one or more other substrings 104a, 104b. For example, the controller 128 is configured to provide charge to each of the substrings 104a, 104b such that their level of charge is aligned with the level of charge of the substring 104a, 104b presently having the higher level of charge. Thereafter, a predetermined amount of charge/energy is preferably provided to each substring in turn, thereby keeping the level of charge similar in each substring. See also figure 4B below. Further detail regarding the amount of charge/energy provided to each of the substrings 104a, 104b when connected to the power supply by the controller 128 is provided below with reference to figures 4A to 4E.

As noted above, the amount of charge provided to each substring 104a, 104b can be determined in a number of different ways (i.e. whether a predetermined amount is provided, whether a predetermined/dynamically determined target is reached, or whether a combination of the above is used). In the case that the capacity/nominal output voltage of each substring 104a, 104b is the same or similar, the controller 128 is optionally configured to monitor the difference in amount of stored charge between each of the substrings 104a, 104b (e.g. using information provided by BMS 109) and compare the monitored difference to a threshold difference. If the monitored difference is already greater than the threshold difference (e.g. because at the start of a charging procedure one substring 104a, 104b has been depleted to a greater extent that another substring 104a, 104b), then the amount of charge provided to each of the substrings 104a, 104b is selected to bring the monitored difference to less than the threshold difference. Similarly, as charging progresses, the controller 128 is preferably configured to provide amounts of charge to each substring 104a, 104b such that at any point in time the monitored difference is less than the threshold difference. It will be appreciated that the value of the threshold difference chosen may depend on a number of different factors dependent on the specific application of the system 100.

In the case that the capacity/nominal output voltage of each substring 104a, 104b is different, rather than monitor the difference in absolute amount of charge between each of the substrings 104a, 104b, the controller 128 optionally monitors the charge/energy stored by each substring as a proportion of its maximum stored charge/energy (for example the average level of charge per cell 106 for each substring 104a, 104b). The controller then compares a difference in amount of stored charge/energy as a proportion of the maximum stored charge/energy between substrings (e.g. the difference in average level of charge per cell between substrings 104a, 104b) to a threshold difference. As charging progresses, the controller 128 is preferably configured to provide amounts of charge to each substring 104a, 104b such that at any point in time the difference (e.g. the difference in average level of charge per cell between substrings 104a, 104b) is less than the threshold difference. Put differently, in the event that the output voltage/capacity of different substrings 104a, 104b is different, current is selectively provided to the energy storage substringsl04a, 104b, such that a difference in a fractional amount of charge (i.e. amount of stored charge as a fraction of the maximum stored charge per substring) between each of the plurality of energy storage substrings remains below a threshold difference. Again, the value of the threshold difference chosen may depend on a number of different factors dependent on the specific application of the system 100.

After the identified energy storage substring 104a, 104b has reached the target level of charge/has been provided with the target amount of charge, the controller 128 controls the switching means 126a, 126 to stop charging the identified energy storage substring 104a, 104b, and selectively charge a different energy storage substring 104a, 104b. Optionally the controller 128 and/or BMS 109 is configured to ramp the current demanded from the power supply down before the controller 128 selectively connects a different substring 104a, 104b to the power supply.

In one example, at step S307 the method 300 returns to step S302 and it is determined which substring 104a, 104b now has the lowest level of charge, and the newly identified substring 104a, 104b is selectively charged as described above. This is particularly advantageous in the context of using more than two substrings 104a, 104b, as it provides large improvements to battery performance in a short space of time by bringing the level of charge in each of the substrings 104a, 104b to a similar level quickly.

Alternatively, the method 300 can optionally omit step S307, and perform step S308, in which the controller selects a next substring 104a, 104b to charge by other means, for example in accordance with a predetermined order. This may be appropriate when using two substrings 104a, 104b, wherein after initially charging the substring 104a, 104b with the lowest charge, it can be inferred that the other substring 104a, 104b of the pair now has the lowest level of charge, without actively determining which substring 104a, 104b has a lowest level of charge.

It should be noted that the amount of charge to be provided to a particular substring 104a, 104b when connected to the power supply may be different to that to be provided to a different substring 104a, 104b. For example, one of the substrings 104a, 104b may be connected to an ancillary system that continues to draw current from said substring 104a, 104b whilst charging is taking place. Thus the controller 128 can be configured to provide a greater quantity of charge to said substring 104a, 104b than to other substrings 104a, 104b in order to compensate for the charge lost to the ancillary system (e.g. the controller 128 is configured to provide charge until the same target level of charge is reached for each substring 104a, 104b, which may mean providing different absolute amounts of charge to each substring 104a, 104b). In such circumstances the controller 128 is preferably configured to direct current to said substring 104a, 104b connected to the ancillary system for a longer period of time or at a higher rate than to other substrings 104a, 104b. Advantageously this ensures that the level of charge in the two of more substrings 104a, 104b can be kept at similar levels throughout the charging process.

Accordingly the controller 128 alternates between charging each of the provided energy storage substrings 104a, 104b. The time between alternating is in part determined by the time taken to provide the desired amount of charge to a particular substring 104a, 104b and/or charge the particular substring 104a, 104b to a particular level of charge. As noted above, by sequentially charging the substrings 104a, 104b in this manner, the level of charge across each of the substrings 104a, 104b can be kept similar (i.e. within the threshold difference) throughout the charging process. This is advantageous for several reasons.

Firstly, by reducing the difference in level of charge between the two (or more) substrings 104a, 104b, "in-rush currents" can be reduced. In-rush currents are currents between various components in the wider system in which the RESS 100 is implemented, induced by a difference in the level of charge in different parts of an energy storage system (e.g. battery). For example, in the context of a charging a BEV/PHEV, in-rush currents are mainly between a DC link capacitance of the EVSE and the substring that it subsequently connects to - for example, if the EVSE were at a potential of 300V and the substring it connected to were at 290 V, then the 10V difference may induce in-rush currents. Another element which can be associated with in-rush currents when charging BEVs/PHEVs are Y- capacitors (devices fitted for electrical noise mitigation) in any of the systems connected to the RESS such as Inverters (for motors), DCDC converters, HVAC heaters or compressors and any On-Board Charger (ACDC). The present invention allows for a reduction in in-rush currents without the need for additional components/higher-current rated components.

Additionally electromagnetic noise associated with high in-rush currents is also reduced. By reducing the in-rush currents upon switch over the magnetic field created by current flowing in a conductor is limited. This improves electromagnetic interference emissions which can be detrimental to other systems. Where the application is a road vehicle, there are standards which must be met for the maximum amount of electromagnetic interference that a vehicle can emit. Being able to tune the set point for energy transfer can be advantageous during vehicle qualification.

As a further benefit, if the charging routine is ended before all energy storage substrings 104a, 104b have been fully charged, battery performance is improved (and in the context of BEVs/PHEVs, range is improved), since the performance is limited by the substring 104a, 104b having the lowest level of charge - in the case of the present system, the substring 104a, 104b having the lowest level of charge will still have a level of charge similar to that of the other substrings 104a, 104b in the RESS 100. Thus the use of the power routing device 100 is able to add range quickly and efficiently in the context of BEVs and PHEVs.

Furthermore, charging of individual substrings 104a, 104b as described above means that the overall voltage balance of the RESS 100 can be improved more quickly than by using conventional charging methodologies. For instance, traditional charging methodologies charge an entire battery to a high charge level (e.g. greater than 80% charge), then conduct a separate step of extensive balancing within the battery before charging can be recommenced. In contrast, the present methodology intrinsically performs a certain degree of balancing throughout the charging process, by ensuring that the different substrings 104a, 104b have similar levels of charge at all times.

During switchover between charging of different energy storage substrings 104a, 104b, the controller 128 is optionally configured to communicate with an attached power supply (e.g. EVSE) so as to demand a reduced, but non-zero current (e.g. 1 amp). Advantageously this ensures a charging session is continued in the event that power supply/EVSE is configured to end if the demanded current drops to zero (such as EVSE operating according to the CHAdeMO standard). Accordingly this further improves the compatibility of the RESS 100 with existing charging infrastructure.

Alternatively, rather than demanding a reduced, non-zero current when switching between substrings (that is, causing the EVSE to ramp down current before a first substring is disconnected, and ramp current up again once a new substring has been connected), the controller 128 is configured to operate switching means 126a, 126b at a rate equal to or exceeding a threshold frequency. This allows the charging current demanded to remain at a constant level when switching between substrings 104a, 104b, thereby reducing overall charging time. This is described in more detail in relation to figures 7 and 8 below.

Preferably, the power routing system 120 further comprises means for controlling in-rush current through the power routing system 120 during switch over between substrings (that is when disconnecting a substring from power input connection 122 that has just had a charging current applied, and connecting a next substring to be charged to the power input connection 122). This means provides a load path between the EVSE (or other charging means) and substrings 104a, 104b, to enable a DC link voltage of the EVSE (or other charging means) connected to the power input connection 122 (which is initially at the present voltage of the substring 104a, 104b that has just been charged) to be reduced in a controlled (i.e. current limited) manner to the present voltage of the next substring 104a, 104b to be charged. This means may take the form of a dedicated load (e.g. a resistor, not shown) that is selectively connected between the power input connection 122 and respective substrings 104a, 104b such that in-rush currents are reduced. More preferably, the means for controlling current is implemented by "soft switching" the switching means 126a, 126b; in other words pulsing the switches 126a, 126b in a fashion that allows for the current through the device to be controlled (i.e. pulse width modulation of the current). This "soft switching" effectively gradually changes the resistance of the switching means 126a, 126b, such that the EVSE (or other charging means) is effectively briefly connected to the next substring to be charged via a resistive load (the resistive load in this case being the plurality of switching means 126a, 126b). The use of such a "soft switching" technique advantageously reduces the number of components required in the power routing device 120, since it utilises the existing switching means 126a, 126b. An additional benefit of providing a load path between the input and output during switchover (via connecting a dedicated load or by "soft switching" transistor switching means 126a, 126b as described above), is that such current limiting provisions allow for the use of substrings 104a, 104b having dissimilar output voltages/ capacity.

In the absence of the charging routine being interrupted (for example by a user disconnecting a power supply from the RESS 100), the process continues for as long as charging is required. For example an electric automotive charging station is typically configured to continue supplying current to a connected RESS for as long as current is demanded by the RESS. In the present case, the controller 128 and/or BMS 109 is configured to continue demanding current from the power supply until all substrings 104a, 104b have reached a desired level of charge (e.g. 100% charge). In one embodiment, the BMS 109 is configured to reduce the current demand for a connected substring 104a, 104b when the connected substring 104a, 104b is approaching its maximum level of charge. Preferably the BMS 109 (and/or controller 128) is also configured to:

• reduce the current demand to zero (i.e. stop charging the connected substring 104a, 104b) when at least one of the cells 106 in the connected substring 104a, 104b reaches a maximum level of charge, and perform balancing between the cells in the connected substring 104a, 104b (i.e. let charge redistribute throughout the connected substring 104a, 104b) or via the use of balancing resistors (not shown);

• increase the current demand after balancing until one or more of the cells 106 reaches a maximum level of charge;

• repeat this process until the charging process ends. Optionally the charging process ends when one or more of the cells 106 in the connected substring 104a, 104b have reached their maximum level of charge. Alternatively, the charging process ends when different criteria are met (e.g. when a target level of charge in one or more of the cells 106/one or more of the substrings 104a, 104b/the energy storage system 102 is reached, or when a charge current limit reaches a certain threshold value). In the case of the present invention, the voltage measured by a charging station at any one time would be that across one of the energy storage substrings 104a, 104b (less an amount due to resistive losses in cabling, switches and other components), depending which was currently connected to the power input connection 122 via the switching means 126a, 126b.

Figures 4A-4E show representations of exemplary charging routines implemented using an RESS 100 comprising two energy storage substrings 104a, 104b as described above. In such a system the energy storage substrings 104a, 104b may be referred to as "half-packs" though it is noted that the two substrings/half packs need not necessarily have equal capacities. In each graph 400, 420, 430, 440, 450, the percentage of the energy stored in the energy storage system 102 is plotted on the y-axis, and charging time is plotted against the x-axis. Whilst the charging regimes shown in figures 4A-4E are described separately for ease of understanding, in some embodiments, the different regimes may be combined and/or alternated between during the charging process.

Figure 4A shows a graph 400 representing an exemplary charging routine. Line 402 shows how the energy stored in the substring having the lowest level of charge initially e.g. energy storage substring 104a increases in steps over time. Line 404 shows how the energy stored in the other substring, e.g. energy storage substring 104b, increases in steps over time. Line 406 shows how the total energy storage by the energy storage system 102 increases over time.

Figure 4A also shows a graph 410 illustrating the current demand supplied to each of the energy storage substrings 104a, 104b during the charging routine represented by graph 400. Line 412 shows the periods during which current is supplied (represented by a value of 1) and the periods in which current is not supplied (represented by a value of 0) to the substring having the lowest level of charge initially, e.g. energy storage substring 104a. Line 414 similarly shows the periods during which current is supplied and the periods in which current is not supplied to the other substring, e.g. energy storage substring 104b.

In the case shown in figure 4A, each substring is alternately charged, such that a predetermined amount of charge (in this case 5% of the maximum charge of the energy storage system 102) is provided to each substring in turn. As will be appreciated, if the charging routine is interrupted before the energy storage system achieves 100% charge, then the two substrings 104a, 104b will have similar charge levels, thus reducing the impact of detrimental effects due to an imbalance of charge levels in the energy storage substrings 104a, 104b, and in the context of BEVs and PHEVs, improving/maximising the range for a given total charge level of the energy storage system 102.

Figure 4B shows a graph 420 representing an exemplary charging routine in which the substrings 104a, 104b are out of balance (that is have unequal levels of charge) when charging is commenced. Line 422 shows how the energy stored in a first substring, e.g. energy storage substring 104a, increases in steps over time. Line 424 shows how the energy stored in the other substring, e.g. energy storage substring 104b, increases in steps over time. Line 426 shows how the total energy storage by the energy storage system 102 increases over time. Line 428 represents how the effective range of the vehicle increases over time. In this scenario, at t=0 with a vehicle 200 (that is RESS 100) starts with 80% State of Charge (SoC) and equally charged substrings 104a, 104b. The vehicle 200 is driven between t=0 and t=10 where the electrical load on the second substring 104b is higher than that of the first substring 104a, and thus by t=10 the SoC of substring 2 is 5% lower than that of the first substring 104a. The available range of the vehicle is dictated by the lowest SoC of either of the substrings 104a, 104b. At t=10 the vehicle is plugged in to the Electric Vehicle Supply Equipment (EVSE) and charging is commenced. The controller 128 of the power routing system 120 selects the substring with the lowest SoC and charges this substring continuously until t= 15 where the SoC of the second substring 104b matches the first substring 104a. During t=10 to t= 15 the SoC of the first substring 104a is constant at 30% but since the vehicle range is dictated by the lowest SoC of either substring 104a, 104b, range is added to the vehicle at twice the effective charging rate, shown by the steeper line 428. Once the two substrings 104a, 104b are at equal SoCs the controller 128 sequentially charges each substring in 1% steps between t= 15 to t=40 at which point the EVSE is removed and charging stops. It can be seen here that since the first substring 104a is now 1% lower energy than the second substring 104b that the available range, line 428, is lower than the total stored energy, line 426. In this example, the use of the power routing system 120 means that the two substrings 104a, 104b are first brought to a similar level of charge, then maintained at a similar level of charge throughout the charging process. Figure 4C shows a graph 430 illustrating a charging regime where each half pack is charged to the same SoC during each step. This method can be thought of as a "one step at a time" strategy. Line 432 shows how the energy stored in a first substring, e.g. energy storage substring 104a, increases in steps over time. Line 434 shows how the energy stored in the other substring, e.g. energy storage substring 104b, increases in steps over time. Line 436 shows how the total energy storage by the energy storage system 102 increases over time. Line 438 represents how the effective range of the vehicle increases over time. If the energy storage system 102 is charging from 0% SoC, two substring 104a, 104b charge events are required to realise any vehicle range. Each substring 104a, 104b is connected to the EVSE in turn, and the same amount of energy transferred in each step. The control system checks the state of charge at each step and charges the substring 104a, 104b of lowest SoC to ensure maximum available range.

Figure 4D shows a graph 440 a regime which reduces the number of switch-over events in order to reduce the overall charging time. This can be thought of as a "one step ahead of the last" strategy. Line 442 shows how the energy stored in a first substring, e.g. energy storage substring 104a, increases in steps over time. Line 444 shows how the energy stored in the other substring, e.g. energy storage substring 104b, increases in steps over time. Line 446 shows how the total energy storage by the energy storage system 102 increases over time. Line 448 represents how the effective range of the vehicle increases over time. Again, if the energy storage system 102 is charging from 0% SoC, two substring 104a, 104b charge events are required to realise any vehicle range. In this regime, after the first charge event, the second event transfers twice the energy resulting in the second substring 104b having twice the SoC of the first substring 104a. The system then switches to the first substring 104a and the process repeats. The available range of the vehicle is the same as the "1 step at a time" regime but the number of switch-over events is reduced.

Figure 4E shows a graph 450 a regime in which the amount of charge provided to each substring 104a, 104b during each charging step is configurable. Line 452 shows how the energy stored in a first substring, e.g. energy storage substring 104a, increases in steps over time. Line 454 shows how the energy stored in the other substring, e.g. energy storage substring 104b, increases in steps over time. Line 456 shows how the total energy storage by the energy storage system 102 increases over time. Line 458 represents how the effective range of the vehicle increases over time. As shown in figure 4E, charging is performed in steps of 1% of the total SoC of the energy storage system 102 (contrasted with the 5% steps shown in figure 4B for example), however the steps are configurable and may alternatively be 5% of the total SoC of the energy storage system 102, or some other predetermined amount. Advantageously, this configurability allows for flexibility in the integration of the power routing system 120. For example:

• Where the voltage of the energy storage system 102 is low but the capacity high, such as a 48V energy store being charged from a 24V supply, the charging can be done in large steps (i.e. a large percentage of the total SoC of the energy storage system 102 since the voltage difference between substrings will remain low, even if the stored energy is quite different.

• Where the energy storage system 102 is both high voltage and high capacity, keeping the voltage difference between the half packs low (i.e. using steps corresponding to a relatively low percentage of the total SoC of the energy storage system 102) is beneficial in relation to control in-rush currents (see above).

• Where the rushing of currents is not a significant system integration issue or can be controlled by other means (see above), the step size can be customised to optimise available vehicle range, charging time or half-pack (i.e. substring) balancing.

Figure 5 shows a schematic circuit diagram of a further example of a power routing system 520. The power routing system 520 is similar to power routing system 120 described above in relation to figure 1. The power routing system 520 includes a power input connection 522, a plurality of power output connections 524a, 524b, 524c, a plurality of switching means 526a, 526b, a controller 528, and a plurality of diodes 530a, 530b, 530c, 530d, whose function/operation is as described above with reference to the power routing device 120 of figure 1.

The example of figure 5 is specific to an automotive RESS, and during charging, controller 528 is in communication with the vehicle CAN bus and the charging station (EVSE). The power routing system 520 includes an additional switch 532 in series with a load 534 (as shown a resistor) connected in parallel with diode 530a. When it is desired to charge an upper energy storage substring - i.e. switch 526a is open (in a state of high resistance) and switch 526b is closed (in a state of low resistance) - the controller 528 is configured to selectively close switch 532 for a certain amount of time, thus bypassing diode 530a and allowing current to flow through the load 534. Advantageously this allows the EVSE to "see" the connected upper energy storage substring (that is, detect a voltage across the connected upper energy storage substring) that would otherwise be prevented by the presence of diode 530a. Put differently, this enables flow of current required to pre-charge the EVSE capacitance. This provision ensures compatibility between the power routing system 520 and EVSE that requires detection of a voltage across a connected energy storage device before it provides current. Once the EVSE starts providing current, switch 532 can be opened whilst the upper energy storage substring is charged. Beneficially, use of the load 534 (resistor) limits the current through switch 532 allowing a lower capability switch to be used as compared to the switches 526a 526b which are configured to handle higher currents.

A current sensor 536 is optionally incorporated into the power routing device 520 to enable the current provided by the EVSE to be measured and relayed by the controller 528 to the EVSE where required. The current sensor measures any current flow from the EVSE irrespective of the substring connected. This can be relayed to the EVSE if required by the standard utilised by the EVSE, to a vehicle communications system, or used by controller 528 as a check to confirm another measured value from the EVSE and/or vehicle system.

Figure 6 illustrates an exemplary process 600 for performing the charging steps described above in relation to figures 3-4E. In particular, figure 6 shows an exemplary software process for use with a vehicular RESS 100 having two substrings ("half-packs") being charged by an EVSE. The sequence shown is specifically for the CHAdeMO charging standard, which has a requirement to maintain a minimum of a 1A charge current IMin, demand throughout the charging session in order to prevent charge session termination during switch over.

Figure 6 shows the steps taken 602, the charging current provided by the EVSE 604 and which substring ("half-pack") is connected 606 with respect to time over the course of a single charging cycle (i.e. a cycle during which charge is applied to one of the substrings before charge is applied to a different one of the substrings).

Half-pack 1 has the lowest state of charge at the start of the process. In step 1 (e.g. during time period tl), it is assessed whether the state of charge of the energy storage system 102 is at a maximum value (e.g. by measuring/inferring the state of charge as described above, or receiving information from the CAN or other vehicle bus). If the state of charge is less than the maximum, the charging process proceeds to step 2.

At step 2 (e.g. during time period t2), the substring having the lowest state of charge (in this case half-pack 1) is connected to the EVSE. Because the CHAdeMO charging standard has a requirement to maintain a minimum current IMin, there is a step change in the current from 0 to 1A when the power routing device 120 connects the CHAdeMO EVSE to half-pack 1. It will be appreciated that this requirement for IMin is standard-specific, and may not be required for other standards (e.g. CCS).

During step 1 and initially during step 2, the current supplied is shown as being 0 A. Alternatively, for certain charging standards the current supplied may not drop below IMin (as shown by dotted line 608).

In step 3, the BMS 109/controller 128 transmits a current demand to the EVSE, and the current is ramped up to ICharge during time period t3. The ramp rate of current shown in step 3 is in part dictated by the charging standard: different standards may have a different gradient to the slope and thus time t3 may be different. The magnitude of the charging current ICharge also effects the duration t3, since the ramp rate of current is fixed, the larger the current demand, the greater amount of time it takes to achieve ICharge.

At step 4, during time period t4, the demanded charging current (or, if lower, the maximum charging current the EVSE is capable of providing) ICharge is provided to the connected substring (in this case half-pack 1). Step 4 is where the majority of the energy is transferred. As described above, the system aims to transfer a predetermined quantity of energy to the connected substring during each cycle, therefore the higher the magnitude of ICharge the shorter the duration of t4. The duration of t4 and the charge current ICharge are chosen based on the amount of charge to be provided, and is based on any combination of the factors discussed above.

At step 5, during period t5, the current from the EVSE to the connected substring is ramped down to IMin. Again, the ramp rate of current shown in step 5 may depend on the maximum ramp rate of the standard being used and the magnitude of the charging current ICharge.

For the CHAdeMO standard, after steps 1-5 have been performed for the first time, the demanded charging current is preferably kept at IMin until the next substring has been connected (see dotted line 608) - keeping current demand at or above IMin throughout the rest of the charging procedure prevents early termination of the charging process due to characteristics of the CHAdeMO standard. After the final time step 5 has been performed, the current may be ramped down to 0A to end the charging session (see dotted line 610).

It will be appreciated that the requirement to maintain charging current at or above IMin may not be present for charging standards other than CHAdeMO. For example CCS will permit the current to be ramped down to 0 A when switching between substrings without ending the charging session (see dotted line 610).

At the end of step 5 the sequence starts again at step 1. In this case, during step 2, the other of the two substrings, half-pack 2, will be selected if it now has a level of charge lower than that of half-pack 1. Alternatively, if the two substrings were initially significantly out of balance with each other, then it may take more charge to bring the two halves in to balance - in this case half-pack 1 may be charged for longer at step 4 (i.e. time period t4 is longer in duration), or selected for several consecutive cycles before half-pack 2 is selected.

After the first arbitration and charging steps (i.e. after steps 1-5 have been completed for a first substring/half-pack), the next arbitration step (i.e. the next instance of step 1) may optionally be made whilst the EVSE is still connected to the first substring, as shown in dotted line 612 (again, this is dependent on the charging standard - for example the current cannot drop to zero according to the CHAdeMO standard, but can according to the CCS standard). Figure 7 shows a graph 700 representing a further exemplary charging routine. Line 702 shows how the energy stored in the substring having the lowest level of charge initially e.g. energy storage substring 104a increases in steps over time. Line 704 shows how the energy stored in the other substring, e.g. energy storage substring 104b, increases in steps over time. Line 706 shows how the total energy storage by the energy storage system 102 increases over time. Line 708 represents how the effective range of the vehicle increases over time.

Figure 7 also shows a graph 710 illustrating the current demand supplied to each of the energy storage substrings 104a, 104b during the charging routine represented by graph 700. Line 712 shows the periods during which current is supplied (represented by a value of 1) and the periods in which current is not supplied (represented by a value of 0) to the substring having the lowest level of charge initially, e.g. energy storage substring 104a. Line 7414 similarly shows the periods during which current is supplied and the periods in which current is not supplied to the other substring, e.g. energy storage substring 104b.

In the case shown in figure 7, each substring is alternately charged, similar to the situation set out in figures 4A and 4C. In this case however, the controller 128 is configured to operate switching means 126a, 126b so as to alternate between connecting respective substrings 104a, 104b to the power input connection 122 at or above a threshold frequency. By alternately and selectively connecting each substring 104a, 104b to suitable EVSE via power input connection 122 at or above the threshold system, the EVSE is unable detect a discontinuity in the current flow due to the switching between substrings 104a, 104b (or any discontinuity in the current flow due to the switching between substrings 104a, 104b detected by the EVSE is within an acceptable tolerance). It will be appreciated that the threshold frequency is system dependent - the threshold frequency will depend on various system characteristics (e.g. on the charging standard being used by an EVSE in the context of charging a BEV or PHEV), and on the context in which the energy storage system 102 is employed (e.g. its end application and the environment in which it is deployed). In one example, the threshold frequency is between 0.1 Hz and 100Hz, for example 0.1 Hz, 1 Hz, 10Hz or 100Hz.

Advantageously, switching between substrings 104a, 104b at a rate at or above the threshold frequency avoids the need to ramp down the current demanded from the EVSE before disconnecting one substring 104a, 104b, and subsequently ramp the current demand up again after another substring 104a, 104b has been connected, at rates determined by the charging standard being used. By avoiding these current ramp down/ramp up periods when switching between substrings 104a, 104b, the overall time taken to charge the energy storage system 102 can be reduced.

Similarly to figures 4A and 4C, figure 7 shows a predetermined amount of charge (in this case 0.025% of the maximum charge of the energy storage system 102) being provided to each substring in turn, and the substrings 104a, 104b are charged to the same level. Alternatively, charging can progress according to a "one step ahead of the last" strategy as described above in relation to figure 4D, a configurable amount of charge per cycle strategy as described above in relation to figure 4E, and/or include an initial balancing step as described above in relation to figure 4B.

Figure 8 illustrates an exemplary process 800 for performing the charging steps described above in relation to figure 7. In particular, figure 8 shows an exemplary software process for use with a vehicular RESS 100 having two equal capacity substrings (also referred to as "half-packs”) being charged by an EVSE, however it will be appreciated that more than two substrings may alternatively be provided. The sequence shown uses the CHAdeMO charging standard, though it will be appreciated similar techniques may be used for other charging standards.

Figure 8 shows the steps taken 802, the charging current provided by the EVSE 804 and which substring ("half-pack") is connected 806 with respect to time over the course of charging the energy storage system 102.

Initially the steps taken closely follow the steps shown in relation to figure 5. Half pack 1 has the lowest state of charge at the start of the process. In step la (e.g. during time period tla), it is assessed whether the state of charge of the energy storage system 102 is at a maximum value (e.g. by measuring/inferring the state of charge as described above, or receiving information from the CAN or other vehicle bus). If the state of charge is less than the maximum, the charging process proceeds to step 2a.

At step 2a (e.g. during time period t2a), the substring having the lowest state of charge (in this case half-pack 1) is connected to the EVSE. Because the CHAdeMO charging standard has a requirement to maintain a minimum current IMin, there is a step change in the current from 0 A to IMin when the power routing device 120 connects the CHAdeMO EVSE to half-pack 1. It will be appreciated that this requirement for IMin is standard-specific, and may not be required for other standards (e.g. CCS).

In step 3a, the BMS 109/controller 128 transmits a current demand to the EVSE, and the current is ramped up to ICharge during time period t3a. The ramp rate of current shown in step 3a is in part dictated by the charging standard: different standards may have a different gradient to the slope and thus time t3a may be different. The magnitude of the charging current ICharge also effects the duration t3a, since the ramp rate of current is fixed, the larger the current demand, the greater amount of time it takes to achieve ICharge.

At step 4a, the process differs from the process described in relation to figure 5. During period t4a, the demanded current ICharge is continuously supplied by the EVSE while the controller 128 causes the switching means to rapidly (i.e. at a rate equal to or exceeding the threshold frequency) alternate between connecting half pack 1 and half-pack 2 to the EVSE. Step 4a optionally continues (in the absence of user interruption) until the energy storage system 102 has reached a predetermined level of charge/amount of stored energy (for example until the energy storage system 102 is approaching a maximum level of charge/stored energy).

At the end of step 4a, half-pack 2 is connected to the EVSE. At step 5a, during period t5a, the current from the EVSE to the connected substring (half-pack 2) is ramped down to IMin. Again, the ramp rate of current shown in step 5a may depend on the maximum ramp rate of the standard being used and the magnitude of the charging current ICharge. Standards other than CHAdeMO may not require ramping the current down to IMin (current could be ramped down to 0A using other standards, e.g. CCS). By connecting one substring 104a, 104b (in this case half-pack 1) to the EVSE during the initial ramp up of current during step 3a, and connecting the other of the two substrings 104a, 104b (in this case half-pack 2) during current ramp down at step 5a, substantially equal amounts of charge can be provided to each substring 104a, 104b during the ramp up/down procedures required by the charging standard. Optionally, the initial pulse(s) durations of step 4a (i.e. the amount of time each substring 104a, 104b is connected to the EVSE supplying ICharge) can be varied in length to accommodate for out of balance substrings or substrings having different capacity/output voltage. This allows the stored energy (or, in the case of substrings having unequal capacity/output voltages, the presently stored energy as a proportion of the maximum stored energy for respective substrings) in each of the two substrings to be brought to the same/a similar amount, before providing a symmetrical pulse regime (i.e. equal length pulses for each substring in sequence) for the bulk of the energy transfer from the EVSE to the energy storage system 102. Similarly the initial and final charging pulse(s) duration of step 4a can be varied in duration to accommodate for differences in the stored energy in each substring 104a, 104b caused by the energy transferred during the ramp up/ramp down of charging current.

For example, as shown in figure 7, the first two pulses provided to half-pack 2 are of longer duration (i.e. the amount of time half pack 2 is connected to the EVSE is longer for the first two instances in which it is connected to the EVSE), to bring the level of charge in half-pack 2 into alignment with the level of charge in half pack 1 following the transfer of charge to half-pack 1 during the current ramp up in step 3a. Similarly, the final two pulses provided to half pack 1 are also of longer duration to account for the charge that will be supplied to half-pack 2 during current ramp down in step 5a.

Although the above embodiments have been described in relation to charging energy storage systems for automotive vehicles in particular (for example for charging energy storage systems for powering electric automotive drivetrains), it will be appreciated that the present invention can equally be applied for the purpose of charging energy storage systems in other contexts. For example, the disclosed power routing systems 120, 520 can be used for charging series connected energy storage substrings in other vehicles (including energy storage systems on water and rail vehicles), in domestic and industrial energy storage units, and in energy storage units associated with renewable power generation (such as grid-connected energy storage systems associated with wind or solar power generation).

The above embodiments are provided as examples only. Further aspects of the invention will be understood from the appended claims.

Claims

1. A power routing system configured to route electrical current from a power supply to an energy storage device having a plurality of energy storage substrings, the power routing system comprising: a power input connection for connection to a power supply; a plurality of power output connections for connection to respective energy storage substrings of the plurality of energy storage substrings; a plurality of switching means disposed between the power input connection and the plurality of power output connections; and a controller; wherein the controller is configured to: i) receive a plurality of signals indicative of a level of charge in each of the plurality of energy storage substrings; and ii) control the plurality of switching means, based on the received plurality of signals, to selectively provide current from the power input connection to each of the plurality of energy storage substrings, such that a difference in the level of charge between each of the plurality of energy storage substrings remains below a threshold value.

2. The power routing system of claim 1, wherein the controller is configured to: identify, prior to step ii) and based on the received signals, which of the plurality of energy storage substrings has a lowest level of charge; and control the plurality of switching means to selectively provide current from the power input connection to the identified energy storage substring prior to selectively providing current from the power input connection to any other of the plurality of energy storage substrings.

3. The power routing system of claim 2, wherein the controller is configured to, prior to step ii): identify a first difference in level of charge between the identified energy storage substring and another energy storage substring; and if the first difference in level of charge is greater than the threshold value: control the plurality of switching means to selectively provide current from the power input connection to the identified energy storage substring until the first difference in level of charge is less than the threshold value, prior to selectively providing current from the power input connection to any other of the plurality of energy storage substrings.

4. The power routing system of any preceding claim, wherein at step ii) the controller is configured to control the plurality of switching means to selectively provide current from the power input connection to each of the plurality of energy storage substrings such that: a predetermined amount of charge is provided to one or more of the plurality of energy storage substrings; the level of charge in one or more of the plurality of energy storage substrings reaches a predetermined target; and/or the level of charge in one or more of the plurality of energy storage substrings reaches a dynamically determined target, wherein the dynamically determined target is based on a level of charge of an energy storage substring having a highest level of charge.

5. The power routing system of claim 4 wherein at step ii) the controller is configured to: control the plurality of switching means to selectively provide current from the power input connection to a first energy storage substring determined to have the lowest level of charge, prior to selectively providing current from the power input connection to any other of the plurality of energy storage substrings, until the level of charge in the first energy storage substring reaches the dynamically determined target.

6. The power routing device of any preceding claim, wherein the controller is configured to: subsequent to selectively providing current to a first energy storage substring and prior to selectively providing current to a second energy storage substring: selectively control current flow caused by a difference in stored charge in the first energy storage substring and the second energy storage substring, by using a pulse width modulated signal to operate the switching means so as to provide an effective resistive load between the power input connection and the second energy storage substring.

7. The power routing system of any preceding claim wherein the controller is configured to repeat step ii).

8. The power routing system of claim 7, wherein the controller is configured to control the switching means to alternately provide current from the power input connection to respective energy storage substrings of the plurality of energy storage substrings at a rate equal to or exceeding a threshold frequency.

9. The power routing system of any preceding claim, wherein level of charge is defined as either: an absolute amount of charge stored in a respective substring; or an amount of charge stored in a respective substring as a fraction of a maximum amount of charge of the respective substring.

10. A rechargeable energy storage system comprising: the power routing system of any of claims 1 to 9; and the energy storage device having the plurality of energy storage substrings.

11. A vehicle comprising the power routing system of any of claims 1 to 9, or the rechargeable energy storage system of claim 10.

12. A method for charging an energy storage device having a plurality of energy storage substrings comprising: i) receiving a plurality of signals indicative of a level of charge in each of the plurality of energy storage substrings; and ii) selectively providing current from a power input connection to each of the plurality of energy storage substrings, such that a difference in the level of charge between each of the plurality of energy storage substrings remains below a threshold value.

13. The method of claim 12, comprising: identifying, prior to ii) and based on the received signals, which of the plurality of energy storage substrings has a lowest level of charge; selectively providing current from the power input connection to the identified energy storage substring prior to selectively providing current from the power input connection to any other of the plurality of energy storage substrings.

14. The method of claim 13, comprising, prior to step ii): identifying a first difference in level of charge between the identified energy storage substring and another energy storage substring; and if the first difference in level of charge is greater than the threshold value: selectively providing current from the power input connection to the identified energy storage substring until the first difference in level of charge is less than the threshold value, prior to selectively providing current from the power input connection to any other of the plurality of energy storage substrings.

15. The method of any of claims 12 to 14, comprising, at step ii), selectively providing current from the power input connection to each of the plurality of energy storage substrings in turn such that: a predetermined amount of charge is provided to one or more of the plurality of energy storage substrings; the level of charge in one or more of the plurality of energy storage substrings reaches a predetermined target; and/or the level of charge in one or more of the plurality of energy storage substrings reaches a dynamically determined target, wherein the dynamically determined target is based on a level of charge of an energy storage substring having a highest level of charge.

16. The method of claim 15 comprising, at step ii): selectively providing current from the power input connection to a first energy storage substring prior to selectively providing current from the power input connection to any other of the plurality of energy storage substrings until the level of charge in the first energy storage substring reaches the dynamically determined target.

17. The method of any of claims 12 to 16 comprising: subsequent to selectively providing current to a first energy storage substring and prior to selectively providing current to a second energy storage substring: selectively controlling current flow caused by a difference in stored charge in the first energy storage substring and the second energy storage substring, by using a pulse width modulated signal to operate switching means so as to provide an effective resistive load between the power input connection and the second energy storage substring.

18. The method of any of claims 12 to 17 comprising repeating step ii).

19. The method of claim 18, comprising alternately providing current from the power input connection to respective energy storage substrings of the plurality of energy storage substrings at a rate equal to or exceeding a threshold frequency.

20. The method of any of claims 12 to 19, wherein level of charge is defined as either: an absolute amount of charge stored in a respective substring; or an amount of charge stored in a respective substring as a fraction of a maximum amount of charge of the respective substring.