ELECTRIC MOBILE REFRIGERATION UNIT

Abstract

Refrigeration units for cooling the interior of a trailer or vehicle, related software, systems and methods for deploying and managing such units. The refrigeration unit comprises a refrigeration system configured to mount to the trailer or vehicle. The refrigeration unit further comprises a battery rack configured to receive at least one of a plurality of rechargeable batteries so as to allow it to be swapped into and out of the rack to provide adaptive battery capacity. A power management system is configured to receive DC power from the plural batteries and deliver power to a compressor of the refrigeration system. A controller is configured to control the refrigeration system to cool the interior to a predetermined temperature.

Claims

1. A refrigeration unit for cooling an interior of a trailer or vehicle, wherein the refrigeration unit is configured to receive a plurality of rechargeable batteries, the refrigeration unit comprising: a refrigeration system configured to mount to the trailer or vehicle, wherein the refrigeration system includes: an evaporator; a condenser; an expansion valve; a controller; and a compressor configured to cause refrigerant to circulate and remove heat from the interior via the evaporator and emit heat to a surrounding environment via the condenser; a battery rack configured to receive at least one of the rechargeable batteries so as to allow it to be selectively swapped into and out of the rack by an operator, wherein when the at least one rechargeable battery is present in the rack it provides additional battery capacity to others of the plurality of rechargeable batteries to provide adaptive battery capacity; a power management system configured to receive direct current (DC) power from the plurality of rechargeable batteries and deliver power to the compressor; and a controller configured to control the refrigeration system to cool the interior to a predetermined temperature.

2. The refrigeration unit of claim 1, wherein at least one battery is fixed in the battery rack and at least one battery is swappable.

3. The refrigeration unit of claim 1, wherein the selectively swappable battery is configured to be electrically coupled to a connector of the power management system when it is received in the rack.

4. The refrigeration unit of claim 1, wherein each of the plurality of rechargeable batteries includes a battery management system that comprises at least one contactor configured to be selectively electrically connectable to a DC bus of the power management system to deliver power, or draw power to recharge the battery.

5. The refrigeration unit of claim 4, wherein each battery management system is configured to sense a voltage level of its rechargeable battery and communicate with the battery management systems of the other rechargeable batteries to manage the connection of its rechargeable battery to the DC bus via its contactor such that the rechargeable batteries with dissimilar voltages are not connected to the DC bus at the same time.

6. The refrigeration unit of claim 5, wherein either a) the battery management systems are configured to selectively switch between rechargeable batteries in turn via their contactors so a single battery is connected to the DC bus at any time, or b) the battery management systems are configured to selectively switch in one or more additional rechargeable batteries once the voltages of one or more rechargeable batteries already connected to the DC bus have equalized with the additional rechargeable batteries such that the additional rechargeable batteries and already connected rechargeable batteries are then connected in parallel via the DC bus.

7. (canceled)

8. The refrigeration unit of claim 1, further comprising at least one solar panel configured to mount to the trailer or vehicle, wherein the power management system is configured to receive DC power from the solar panel to power the compressor, to charge at least one of the rechargeable batteries, or a combination thereof.

9.-10. (canceled)

11. The refrigeration unit of claim 8, further comprising a connector configured to connect to a local or national electricity grid to selectively export electric power from the solar panel or one or more of the rechargeable batteries to the connected electricity grid, to selectively charge the one or more rechargeable batteries from the connected electricity grid, or a combination thereof.

12. (canceled)

13. The refrigeration unit of claim 1, wherein the power management system is configured to export energy to a tractor unit, to receive energy from the tractor unit, or a combination thereof.

14. The refrigeration unit of claim 1, further comprising a communication means configured to stream usage data to a remote software platform, wherein the usage data comprise one or more selected from a group consisting of: start time of journey; duration of journey; temperature set point; weather during journey; number, times and/or duration of delivery drops; payload mass and/or type; Location data; energy usage; and actual temperature profile.

15. The refrigeration unit of claim 1, wherein the power management system is configured to receive control signals from a remote software platform to charge the rechargeable batteries, to export power from the batteries to the electricity grid when connected thereto to export power from the solar panel to the electricity grid when connected thereto, or combinations thereof.

16. (canceled)

17. A computer-implement method for managing power requirements of mobile refrigeration units, the refrigeration units configured to attach to a trailer or vehicle to cool an interior space thereof during a journey and being powered by, at least in part, one or more rechargeable batteries and optionally one or more solar panels, the method comprising: receiving an itinerary for a journey of the refrigeration unit and a set point temperature to be achieved and maintained by the refrigeration unit for cooling the interior space for that journey; modeling energy requirements to achieve the set point temperature for the journey, and determining a number of rechargeable batteries required and/or a battery charge level for each rechargeable battery to provide the modeled energy requirements; and in accordance with the determination: 1) outputting an indication of which rechargeable batteries are to be swapped into and/or out of the refrigeration unit; and/or 2) outputting a control signal to cause the required rechargeable batteries to be charged to the required battery charge level.

18. The computer-implement method of claim 17, wherein inputs to the model for the energy requirements include one or more parameters selected from a group consisting of: desired temperature set point; expected weather conditions during the journey; start time of journey; duration of journey; number, times and/or duration of delivery drops; and payload mass and/or type, wherein the determination takes into account a further input of an initial state of charge of the rechargeable batteries.

19. The computer-implemented method of claim 18, further comprising receiving usage data from at least one refrigeration unit during a journey, the usage data including said one or more parameters and data indicating the actual temperature achieved by the refrigeration unit and energy consumption of the refrigeration unit, which data is used to model performance of an individual refrigeration unit.

20. The computer-implemented method of claim 19, further comprising either a) using a digital twin for modeling performance of an individual refrigeration unit; or b) using plural digital twins for modeling the energy required and the energy available across plural refrigeration units in a fleet of trailers or vehicles, and optimizing the charging and swapping of batteries across the fleet.

21. (canceled)

22. The computer-implemented method of claim 20, further comprising outputting control signals to cause export of surplus electrical energy from a battery or solar panel to a national electricity grid, or to another battery charging on a local electricity grid.

23. (canceled)

24. A system for charging rechargeable batteries for powering mobile refrigeration units, the refrigeration units configured to be coupled to a trailer or vehicle to cool an interior space thereof during a journey, the system comprising: a swapping station comprising charging bays configured to receive plural respective rechargeable batteries removed from refrigeration units for charging; a mains electricity connector configured to receive and to optionally export power to a national electricity grid; charging control circuitry configured to selectively charge connected rechargeable batteries from mains electricity; and a processor configured to execute the computer-implemented method of claim 17, to determine a number of batteries and/or battery charge level for each battery to supply that energy, wherein in the processor is further configured to display, in accordance with the determination, which batteries are to be swapped into the refrigeration unit, to display a schedule of when to swap the batteries, to activate the charging control circuitry to charge the battery level to the required battery charge level, or a combination thereof.

25.-30. (canceled)

31. The refrigeration unit of claim 1, wherein a battery is configured to be electrically coupled to a connector of the power management system when the battery is received in the battery rack, and the power management system is configured to receive DC power from the rechargeable batteries present in the unit and manage delivery of power to the compressor, wherein each of the rechargeable batteries is selectively electrically connectable via a contactor to the power management system controlled by a battery management system for delivering power or drawing power for recharging the battery such that batteries with dissimilar voltages are not connected in parallel.

32. The refrigeration unit of claim 1, wherein the controller is further arranged to monitor a state of charge of the rechargeable batteries present in the unit and to display battery status information to an operator.

33. The refrigeration unit of claim 1, further comprising a housing that contains the refrigeration system, wherein at least one rechargeable battery is located within the housing.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Embodiments of the present disclosure will now be described by way of example with reference to the accompanying drawings, in which:

[0073] FIG. 1 shows a perspective view of an example of a TRU according to an embodiment of the present disclosure attached to a trailer;

[0074] FIG. 2 shows a view of the main unit of the TRU of FIG. 1;

[0075] FIG. 3 shows a schematic view of the TRU of FIG. 1 part of an overall system for providing and managing a fleet of TRUs;

[0076] FIG. 4 shows a possible temperature profile for a TRU in operation;

[0077] FIG. 5 shows an example of the energy prediction system used together with an example of a TRU with a modular compressor system in accordance with an embodiment of the present disclosure;

[0078] FIGS. 6a, 6b and 6c shows an example of a charging station as part of an overall system for providing managing a fleet of TRUs in accordance with an embodiment of the present disclosure;

[0079] FIG. 7 shows an example of a method of using the TRU of FIG. 1; and

[0080] FIG. 8 shows a schematic view of a further example of a TRU according to an embodiment of the disclosure.

DETAILED DESCRIPTION

[0081] FIG. 1 shows a perspective view of an example of a mobile refrigeration unit, more specifically in this example a Transport Refrigeration Unit 10, attached to a semi-trailer 12 of the sort that can be attached to and pulled by a tractor unit (not shown) to transport goods loaded to the interior of the trailer, where the TRU 10 implements a system for refrigerating the interior of the trailer. It will be appreciated that the TRU may equally be attached to other vehicles types, such as rigid body trucks, vans and lorries. As will be described further below, the TRU forms part of an overall system 5 for providing and managing a fleet of TRUs for respective trailers or other vehicles.

[0082] The TRU 10 comprises a main refrigeration unit 14, shown in more detail in FIG. 2, attached to the near end of the trailer 12 (with the doors allowing access to the interior of the trailer being at the far end), as per known arrangements. The main unit 14 comprises the four primary components for the refrigeration cycle in a vapour compression refrigeration system 29: evaporator 30, compressor 32, condenser 34, and expansion valve 36. When the compressor 32 is driven, these combine to chill air in the interior of the trailer 12 to cool the contents. The refrigeration system may alternatively be mounted in a skid on the front of the trailer, under the trailer chassis, as an over-cab mounted skid on a rigid sided vehicle, or any other mounting arrangement.

[0083] The TRU 10 also comprises one or more solar panels 16 attached to the roof of the trailer 12. The solar panels 16 may be low profile, semi-flexible, 20% efficient, polycrystalline panels for instance. These may be mounted to the roof of the trailer, but can be mounted at any convenient point. The TRU 10 also comprises a battery rack 20 which receives one or more removable batteries 22 attached to the trailer 12 in an accessible position. This may for instance be attached to one of the I-beams running the length of a standard trailer. More than one rack may be provided, e.g. on both sides of the trailer. The main unit 14 also may have one or more fixed battery 50. The fixed battery 50 and removable batteries 22, together with the solar panels 16, provide power the TRU 10. These batteries may be battery packs, each comprising multiple individual battery cells monitored and managed by a battery management system in the battery.

[0084] The fixed battery or batteries are typically embedded in the TRU in a form in which they are not intended to be removed during operational life of the TRU. In other words, the TRU would need to be dismantled and/or specialist tools and expertise would be needed to access, detach, remove and replace the fixed battery. Hence, these batteries would not normally be removed (except for instance in exceptional cases where, for instance, a battery failed.) On the other hand, the swappable batteries are intended to be regularly and simply accessed, detached, removed and swapped during the life of the TRU according to the requirements of the operator, as discussed below. Thus, the TRU may be put into operation with any number of swappable batteries present in the rack, e.g. with some slots occupied and others vacant.

[0085] These and other elements of the system 5 and TRU 10 are shown in more detail in the schematic view of FIG. 3, showing in particular elements of the refrigeration system/cycle and elements of the electrical system of the TRU. This shows the aforementioned main elements of the refrigeration system 29 namely the evaporator 30, compressor 32, condenser 34 and expansion valve 36, in this example an electronic expansion valve (EEV) driven by an EEV driver 26a, although a thermal expansion valve may be used. The refrigerant 33, here in an example a low GWP refrigerant, enters the compressor 32 at low temperature and low pressure in a gaseous state. Here, compression takes place to raise the temperature and refrigerant pressure. The refrigerant leaves the compressor 32 and enters the condenser 34. Since this process requires work, the compressor 32 must be driven, here by an electric motor 32a. The compressor 32 may be scroll, screw, centrifugal or reciprocating types. In the present example, the compressor 32 is a reciprocating type driven by an internal permanent magnet AC motor.

[0086] Plural compressors may be used. These may be arranged in a modular way, which may have advantages due to the fact that the power required to power a compressor at fixed speed changes throughout a pull-down. The compressors can be configured to vary the cooling power demand of the refrigeration system. The compressor or compressors may be used with optional liquid injection or economizer if required.

[0087] The condenser 34 acts as a heat exchanger. Heat is transferred from the refrigerant to a flow of fluid—here ambient air driven across the heat exchanger surface area by fans 34a, and so lost to the environment.

[0088] When the refrigerant 33 enters the expansion valve 36, it expands and releases pressure. Consequently, the temperature drops. Because of these changes, the refrigerant leaves the valve as a liquid vapor mixture. The expansion valve serves to maintain a pressure differential between low- and high-pressure sides, as well as controlling the amount of liquid refrigerant entering the evaporator 30.

[0089] At the stage of entering the evaporator 30, the refrigerant is at a lower temperature than its surroundings. Therefore, it evaporates and absorbs latent heat of vaporization from the air inside the trailer 12 which is circulated by fans 30a to cool the contents. Heat extraction from the air to the refrigerant happens at low pressure and temperature. Compressor 32 suction effect helps maintain the low pressure.

[0090] A liquid suction heat exchanger 37 may be installed between the condenser 34 and the expansion valve before entering the evaporator 30. This helps subcooling liquid before entering the EEV and superheating the gas before entering the compressor, which provides better control of the EEV and avoids liquid droplets entering the compressor. Optionally, an accumulator (not shown) is provided upstream of the compressor 32 to prevent liquid refrigerant from flooding back to the compressor 32. A liquid receiver 40 is also provided after the condenser 34 which acts as a storage vessel designed to hold excess refrigerant not in circulation. A pressure reduction valve (not shown) coupled to the liquid receiver 40 safely relieves pressure in case of over-pressure. Various sensors 44 monitor temperature and pressure at various points in the cycle both of the refrigerant and ambient air. Further sensors may monitor the state of the various electrical elements.

[0091] The TRU 10 uses either an electric method to defrost the refrigerated compartment evaporator, or a reverse vapour compression cycle method, or a hot gas defrost method.

[0092] An electrical system 45 of power electronics is provided, the primary purpose of which is to supply electric power to drive the compressor and the fans. The fixed batteries 50 and the swappable batteries 22 are connected to a bus 52. In the present example, the bus is provided within a DC power distribution unit in the TRU main body, which may further comprise fuses, contactors, and CAN I/O module for communications with the controller. In the present example, the batteries are 48 VDC 10 kWh capacity and the TRU may have 4 fixed battery modules and the battery racking system 20 may accommodate up to 6 batteries. However, it will be appreciated that different voltages, capacities or numbers of fixed and/or swappable batteries and or their positioning may be adopted. In some examples, the batteries may be fixed or all batteries may be swappable. The number of batteries can be adjusted on a per journey basis, as energy demand can change between customers, season and application.

[0093] Within the electrical system 45, the batteries are connected via the bus 52 to various power controllers 70,64,66 to manage delivery of power from the various power sources to the batteries and from the batteries and other sources to the power consuming devices (as described further below). These power controllers are generally referred to as the power management system 60 herein. It will be appreciated that in other examples, these functions may be consolidated in a power management system, e.g. into a so-called hybrid solar inverter of the sort known from the solar industry, rather than being provided by separate power controllers/components in the electrical system 45.

[0094] Each battery pack is connected to the DC bus via a contactor 51 (or a contactor for each of the positive and negative terminal) in a Battery Management System (BMS) fitted within the battery pack. A contactor is a heavy duty version of a relay (e.g. solid state or electromagnetic) used to switch power to/from the battery packs. This is controllable to individually connect the battery to the bus for charging/discharging. The system controller might provide high-level instructions (typically from the HMI or a signal received via the cloud platform) such as start refrigeration system, or start recharging via solar, in which case the contactors via the BMS are closed to enable battery packs.

[0095] Individual control of each battery pack is also enabled via the BMS. The BMS monitors the voltage, calculates State of Charge, State of Health and many other parameters. For instance, a battery may produce a voltage range of between 46V (when entirely discharged, i.e. 0% State of Charge) and 58V (when the battery is fully charged, i.e. 100% SoC). The BMS then communicates with the BMS of the other battery packs via the CAN bus in the PDU. Depending on the voltages of the other modules, and whether their contactors are engaged, a BMS determines if it should also engage its contactor. In particular, it will be appreciated that if batteries with dissimilar voltages are connected in parallel, there is a tendency if the difference is too great for one battery to feed energy into the other battery, as current flows from high potential to low. This can shorten battery life, blow fuses and lead to other undesirable effects. The BMSs cooperate to alleviate this by avoiding connecting batteries to the DC bus with dissimilar voltages, e.g. more than 5% higher or lower than each other.

[0096] In a first example, each battery may be used in turn to provide power to the compressor to avoid the situation where battery modules of dissimilar SoC and thus voltages across their terminals will be connected together to the DC bus. For instance, the controller may select the battery module with the highest SoC to initially provide power to the compressor and other power drawing components on the DC bus, and then move to the next battery when the first battery is discharged.

[0097] In a second example, one or more batteries with a relatively high SoC are selected to initially provide power. As those batteries discharge power, their voltages drop until they reach a similar voltage to at least one other battery pack which initially had a lower SoC, at which point, that battery pack is connected to the bus via the contactor so that those battery packs provide power jointly. So for instance, a first battery may initially have a 100% SoC and a second battery has an initial charge of 50% SoC. The first battery is selected first to supply power, until its charge falls to about 50%, at which point, the second battery module can be connected to the DC bus to provide power in combination with the first battery, and so on for any other battery packs.

[0098] This second example may be generally illustrative, as it tends to distribute the load across all batteries, and thus extends their lifespan. However, there may be occasions where using some batteries in preference to others may be advantageous.

[0099] A similar technique operates when charging the battery packs via solar and/or from mains power. In other words, power to charge the batteries is selectively applied to the batteries by activating the contactors such that batteries with unequal voltages/SoC are not connected in parallel at the same time.

[0100] The trailer may have one or more solar panels to provide power when in transit and/or when stationary. Where solar power is available, this can be used to power the compressor motor (in conjunction with battery power if solar is insufficient). If there is excess solar energy, the excess can be used to charge the batteries by selectively connecting the batteries to the DC bus via the contactors. Otherwise, the batteries can be left disconnected. Often solar can only cover part of the energy required, in which case batteries or the on board chargers (OBC) 66 (if plugged in via the grid connector) will provide the remaining. If no solar or OBC is available batteries will provide the energy.

[0101] Each solar panel is connected to a MPPT (Maximum Power Point Tracker) charge controller with bock-boost converter, which in turn is connected to the battery packs via the DC bus. the efficiency of power transfer from the solar cell depends on the amount of sunlight falling on the solar panels, the temperature of the solar panel and the electrical characteristics of the load. As these conditions vary, the load characteristic that gives the highest power transfer efficiency changes. The efficiency of the system is optimized when the load characteristic changes to keep the power transfer at highest efficiency (the maximum power point). MPPT is the process of finding this point and keeping the load characteristic there. Electrical circuits can be designed to present arbitrary loads to the photovoltaic cells and then convert the voltage, current, or frequency to suit other devices or systems, and MPPT solves the problem of choosing the best load to be presented to the cells in order to get the most usable power out. The bock-boost converter then bocks or boosts the voltage level for charging the batteries. Thus, for the scenario where the panel voltage is lower than the battery voltage, it steps up the voltage to be suitable for the battery requirements so it can charge, and similarly where the voltage is too great, it steps it down.

[0102] The grid connector may be single phase, with a on-board charger (OBC) to provide power at the appropriate DC voltage level to the DC bus. Alternatively, the connector may be 3 phase and have at least one on-board charger (OBC) 66 provided for each phase or a 3 phase charger. This allows high current to be generated, as the DC bus is low voltage, to allow maximum power efficiency in charging. The charger is bi-directional, so as to be capable of the reverse process, i.e. converting PV or battery DC power to AC for sharing power with other TRUs via the local grid or exporting surplus power to the wider power grid (described in more detail below).

[0103] The power distribution unit 60 may also have contactors for selectively supplying current to the fans and other components.

[0104] The battery packs are connected to the DC bus in a post-PDU architecture, meaning that each battery pack has its own contactor(s). This electrical architecture allows integration and management of battery packs with different states of charge. It will be appreciated that this is a particular benefit in a system where multiple battery packs are detachable, swappable and scalable, and different battery packs may have very dissimilar states of charge at various points in their operation. This is a problem that does not arise in, say, EV battery management, where the battery modules in a battery packs are typically combined and hardwired together in parallel, such that the overall capacity is fixed and the battery module charge is always at a mutually similar level. While it may be known for an EV battery packs to have a contactor, this is typically used only for connecting the overall battery pack to a load (or possibly for fault isolation of battery modules in a battery pack), rather than controlling individual battery packs in a system as in the present novel architecture.

[0105] The compressor 32 in this example is powered by an AC output voltage provided by the motor controller 70 which alters the frequency of the AC power so as to vary the speed of the motor and thus the compressor under control of the system controller. The output voltage is also selectively supplied to the fans 30a,34a of the evaporator 30 and condenser 34, e.g. via speed control on the fans via PWM.

[0106] The battery 22,50 and solar panel 16 can be used to provide power to the tractor unit itself, for example where the tractor unit runs on fuel cells or is powered by a diesel ICE. Thus, the PDU may have an output 11 to provide supplementary power to the tractor unit. This is beneficial as the solar power is cheaper than the electricity generated by a fuel cell, e.g. via a DC/DC converter. The PDU may also be able to receive supplementary power, e.g. from axle regeneration, from the tractor unit or trailer via an input for use in powering the compressor.

[0107] A system controller 75 is provided with communication links to the various parts of the TRU 10 to control and monitor the refrigeration process, i.e. to pull down and maintain a set point temperature, and to manage and monitor the various energy sources. Thus, the controller 75 communicates with the power converters 70, 64, 66 and the BMS/contactors of all the battery packs, to control the fans 301,34a, the compressor, the power provided by the PV panels and from the connector, the sensors 44 and voltage sensors, and any other elements of the TRU 10 in order to exchange data and send control signals.

[0108] The system controller 75 may be connected to or incorporates a wireless gateway (e.g. 4G) 76 by which it can exchange data with software 120 running on a remote server or in the cloud 78, which is part of the overall system 5. This may be an “Internet of Things” (IoT) cloud service such as for example Azure IoT Central. Thus, the controller may be a so-called IoT edge device. The controller may also be connected to or incorporate GPS for finding its location, and WiFi or Bluetooth or similar wireless signals for communicating with other external devices in the system.

[0109] A Human Machine Interface (HMI) is provided comprising a display and input means, e.g. a touch screen 80, connected with the system controller 75, e.g. by WiFi, by which an operator can locally see the status of the TRU and provide input/control.

a. Possibly inputs are one or more of:

[0110] setting the temperature set-point

[0111] setting the desired time for the TRU to be at set-point (using energy prediction). Options for this could be fastest possible or at set time in the future

[0112] setting the TRU on/off

b. Possibly status data are one or more of:

[0113] Current air temperatures inside/outside the trailer

[0114] Energy flows around the electrical system 45

[0115] Directions as to which batteries 22 to swap/add/remove

[0116] Whether there is sufficient charge in the batteries 22, 50 for an upcoming journey.

[0117] The system controller 75 can also be controlled directly from the cloud by the software 120, so settings can also be adjusted remotely. The system controller 75, gateway 76, and HMI 80 may be powered by an uninterruptable power supply (UPS) 81 which is a battery, separate from the fixed and swappable batteries and typically smaller. This is useful, for instance, where the trailer is parked up without any load and with partially charged batteries, the sun starts shining, and the contactors to the fixed or swappable batteries must be closed to allow charging with solar power or export to the grid. The controller is therefore “always on” and can be controlled remotely from anywhere via the cloud platform to monitor the available solar energy and if sufficient, turn on battery charging or export of energy to the grid. The UPS battery typically has sufficient capacity to power the controller for 24 to 48 hours, and is recharged automatically when there is power on the DC bus.

[0118] The system 5 uses telematics and instrumentation specific for monitoring, including but not limited to, the electrical system 45 and refrigeration system 29 performance. Air temperature inside the trailer is automatically streamed and recorded and used to warn the driver and the fleet operator if temperatures move outside the desired range. All data is stored in the cloud and is used to teach a machine learning algorithm to create a digital twin of each TRU/trailer. The digital twin is subsequently stored in the cloud and is used to provide future energy prediction for each TRU/trailer and across the fleet.

[0119] As shown by FIG. 6, the system 5 comprises a fleet of trailers 12 and TRUs 14, and optionally at least one centralised static charging location, which in this example is a battery swapping station 100, e.g. located at the depot. The system 5 may use plural batteries per TRU, the number and initial charge of which are configured by the software 120, here an IoT software platform running in the cloud 78 in conjunction with a machine learning algorithm that produces and updates digital twins. This enables the batteries 20, 50 to be small compared with prior art schemes, as the storage capacity can be adapted per journey. The software 120 manages batteries for the fleet of trailers and controls and monitors charging of the batteries at the charging location 100. In the present example, the battery swapping station is adapted to receive the removable batteries 22 when removed from the trailer for charging under the control/monitoring of the software. Alternatively or additionally, batteries 22 may be charged in situ under the control/monitoring of the software 120 by connecting the TRUs 14 to the local electricity system via their connectors 62. In this case, the removable batteries may be swapped directly between the trailers in the fleet rather than via a battery swapping station.

[0120] The static charging station is constantly connected to the grid (AC) 150 to charge the batteries via a charger, or to release surplus energy to the grid for demand side response, grid balancing or other grid connectivity. This may be a bi-directional charger (converts AC to DC one way, and inverts DC to AC the other way) as energy flows both ways. Such converters are known as “V2G” and “grid-tie” in other applications.

[0121] FIG. 7 shows the main process carried out by the software. The software first predicts 710 the battery capacity for the journey. The software takes as input data representing a) the logistics schedule for the fleet, namely the desired temperature setting or profile for the goods being transported, the time and duration of the journey, the payload mass and/or type (i.e. a measure of how much is to be transported), and the number of delivery drops, b) the weather forecast and the hours of daylight of the journey, and c) the initial state of charge of the battery. Typically the itinerary for the journey is obtained from the operator's logistics software, but could be manually entered by the operator; the weather information is obtained from an online source; and the battery charge information is determined from the swapping station and or system controller 75 of the TRU 10 depending on the current location of the battery.

[0122] The software 120 then looks at the TRUs available and selects the one which best matches the requirements for the journey. The software predicts the battery capacity, i.e. energy, required to complete the journey for the best match according to the input parameters, in particular the logistics schedule and weather forecast. It will be appreciated that weather conditions and expected hours of daylight during the journey will influence how much energy is generated via solar during the journey. Ambient air temperature will affect the cooling required. The number of stops for unloading affects loss of cooling, which requires additional energy from the system to compensate for.

[0123] Once the energy required is predicted, in step 720, the software determines how many batteries are required for the journey and how much they need to be charged, taking into account the initial charge of the batteries and expected charging until the trailer must leave. Based on this, the closest match will then be charged according to the predicted energy requirement and/or the operator is instructed to swap, add or remove batteries to adapt the number of batteries if required and allow the operator to vary the on-vehicle battery capacity.

[0124] In particular, if the closest match needs another battery, the operator is instructed to add it. If the closest match has too many batteries on-board they may be removed as they can provide revenue via demand-side-response if they are left in the charging station at the depot. If there is time to charge 20 kWh but only 10 kWh is needed, the 10 kWh may be sold to the national grid or used to charge another TRU in the fleet. Effectively, this means that the TRUs need not leave the depot with redundant battery capacity. Redundant battery capacity is better left at the depot to be used by another TRU or for demand-side-response.

[0125] In step 730, the batteries removed from a TRU 10 are moved to the charging station for offline charging or to another trailer. The software controls the charging of batteries in the battery swapping/charging station and/or in situ in connected TRUs.

[0126] In step 740, the trailer and TRU embarks with its adapted battery capacity.

[0127] FIG. 6a shows a fleet of trailers 12/TRUs 10 with various statuses being managed by the software platform 120. In particular, FIG. 6a shows a TRU 10a with excess energy capacity being sold to the grid, a TRU 10b that requires additional batteries and charging via grid 150 and/or solar to reach the predicted energy capacity, a TRU 10c that will depart soon which has the required number of batteries and is disconnected from the swapping station and is undergoing final solar charging, a TRU 10d arriving back at the depot having finished a journey, and a TRU 10e departing the depot on a journey. Various batteries are shown in the battery swapping station 100 charging.

[0128] FIG. 6b shows a method of energy prediction in more detail. In this technique, a “digital twin” 610 is constructed to model at least one and, in some examples, each TRU 10 in the fleet. A digital twin is a digital representation that simulates virtually a real-life object, process or system. The digital twin comprises data in the form of physics-based mechanisms and models, relating to in particular the refrigerant cycle, heat loss from the trailer, etc; material properties and definitions of the system, such as characteristics of the trailer 12 and refrigeration system 29. The virtual representation of the twinned physical asset enables understanding of the behaviour of the physical object under various circumstances and in a variety of environmental conditions, so that its behaviour in the future can be predicted.

[0129] Thus, as described above, key usage metrics are continuously streamed from each TRU 10 to the cloud software 120. Data collected is linked to individual TRUs as each TRU will perform slightly differently from another. For instance, there may be different insulation thicknesses in different trailers, and/or damaged insulation on one trailer and not on another. There may be a TRU with a faulty/less efficient refrigeration cycle than another. A solar panel on one TRU may be damaged or dirty. Thus, the digital twin may be created for each TRU in the fleet, or at least different types of TRU/trailers if those in the fleet can be sub-divided into categories.

[0130] This streamed usage data creates historical data 620 of the performance of a TRU 10, i.e. the “response” of the system in terms of energy usage and temperature profile achieved, based on the “stimuli” to the system, i.e. the input data to the system as described above, i.e. the start time and duration of the trip, the desired temperature set-point, the prevailing weather conditions, the number, times, and durations of delivery drops, the payload capacity/utilisation, the location during the journey. This historical data is used to train the Digital Twin model via machine learning algorithms 640. The trained digital twin represents the digital behaviour blueprint of an individual TRU capturing the real-life response caused by indefinite combinations of stimuli at any given point in time.

[0131] FIG. 6c shows in more detail the digital twin models 610 being used to predict energy usage for a TRU 10 and across the fleet 10a . . . 10n. The software 120 continuously monitors the present state 650 of the individual TRUs 10. This includes GPS location and battery state of charge. The present state provides a first input to the model 610. Future stimuli 655 are also input into the model 610, including the logistics schedule, i.e. when the TRUs must depart, number of deliveries, set-point temperature, time and duration of trip; and the weather forecast, i.e. to estimate ambient temperatures and solar energy available for the PV panels. Based on these input parameters 650,655 to the digital twin model 610 in the IoT software platform 120, the software can predict the energy required for a particular TRU to make the scheduled journey. Thus the data that is used to create the digital twin 610 using machine learning 640 which is then used to make predictions 660 for a specific TRU/journey and to manage the batteries across the fleet.

[0132] In particular, the model predicts 660 (i) how long it takes to pull down, (ii) how much energy is needed to pull down, (iii) how much energy is needed for a specific journey, which is used to generate one or more actions 670 to adjust the battery capacity and/or charging required. “Pull down” usually takes place ahead of loading goods, either starting at the depot or on the way to the first pickup. So by predicting how long it takes to pull-down the software 120 can match when the trailer 12 needs to be ready for the goods. This minimises the time when the empty trailer is cooled to the required temperature ahead of the first pickup. The temperature may be pulled down whilst connected to the grid so less battery charging/capacity is needed.

[0133] FIG. 4 shows typical temperature profile over time when operating a TRU 10. Initially, the TRU “pulls down” temperature inside the trailer 12 from ambient temperature 410 to a desired set point temperature 420 as required by the goods when being transported. When the TRU 10 is started from an ambient temperature the refrigerant is dense and so the pressure of the refrigerant is high. At high pressures, compressors 32 require a lot of power, especially during start-up, due to the high inrush currents. For a typical set point of 5 degrees C. or −20 degrees C., pull down from ambient temperature, say 20 degrees C., may take 30 minutes or more. Once the set point is reached, smaller energy output is required to keep the trailer at the desired temperature, depending again on ambient temperature and effectiveness of insulation of the trailer and resultant heat ingress.

[0134] The amount of energy required over time for a particular journey, for instance shown by FIG. 5, is predicted using the energy prediction software 120 using the digital twin. It can be seen that the (in this example 10 kWh) fixed battery 505 is sufficient for pulldown and maintaining the setpoint for an initial period of time 510. The adaptive battery capacity provided by the swappable batteries 515 is selected to provide sufficient energy to fulfil the total journey requirements, i.e. to move the line 515 along the axis such that the total battery capacity is adapted to the total journey time and requirements.

[0135] Returning to FIG. 6c, the digital twin models 610 provide energy prediction 660 for the fleet. This comprises both what energy is available in the fleet (how much, where and when), and what energy is needed in the fleet (how much, where and when). Decisions are made by the software to balance the requirements, taking into account also the cost of electricity 680 and the potential profit from feeding energy back into the grid 150.

[0136] Finally, various actions are taken by the energy prediction software 120 based on the balancing calculation. As described, instructions can be issued to an operator to vary the number of swappable batteries 22 on the on-vehicle battery rack 20, based on expected consumption and hence the energy storage capacity. Unused batteries 22 are moved to the charging station 100 or other trailers 12 in the depot for offline charging under the control of the software so as to be ready for other upcoming journeys and/or participation in DSR services, i.e. the modification of consumer demand for energy and/or selling energy back to the national grid based on signals received from a utility company. The software controls charging and discharging of the connected batteries.

[0137] Referring again to FIG. 6a, the software platform 120 and charging station 100 has the capacity to manage a fleet of TRUs. This provides the ability for multiple TRUs to communicate with each other and share energy through battery swapping and vehicle to grid (V2G) technology. Vehicle-to-grid (V2G) describes a system in which plug-in electric vehicles, such as battery electric vehicles (BEV), plug-in hybrids (PHEV) or hydrogen fuel cell electric vehicles (FCEV), communicate with the power grid to sell demand response services by returning electricity to the grid. The power distribution unit 60 in each TRU and/or the power management system in the battery swapping station may be connected to the local AC grid, and thence to the grid 150, to charge the batteries or to provide V2G functionality, i.e. via an on-board or off-board bidirectional charger. A smart energy meter is mounted on the on-vehicle electrical circuit and the static charging station electrical circuit to monitor energy usage from the refrigeration system and supply energy usage information to a server.

[0138] Plural TRUs may be connected together to become a “micro grid” and may be controlled through the software platform that uses energy prediction to instruct TRUs to either share energy (from solar or their batteries) with other TRUs (directly and/or via the battery swapping station), or charge their own battery depending on (i) photovoltaic power production (ii) grid electricity cost (iii) grid capacity, (v) demand from other TRUs or (vi) logistics schedule.

[0139] Returning to the example architecture of FIG. 3, whilst a single compressor is used in this example, other arrangements are possible. As mentioned above, the compressors have a high power requirement during the initial pulldown phase. In embodiments where an inverter/motor controller is used to power the compressor motor, ordinarily, very large and expensive inverters would be needed to allow for sufficient peak power capacity to drive the compressors at start-up, as smaller units would be prone to failure. Various measures may be taken to mitigate this problem and avoid the need for large expensive inverters.

[0140] In a further example, a variable speed drive 70 is used under the control of the system controller 75 to modify the frequency of the AC power supplied from the inverter part of the power management system from the battery system 22,50 and/or solar panel system 16 to the AC motor of the compressor 32 at start up so that the power applied by the motor gradually ramps up during start up, keeping the power requirement within the safe capabilities of the motor controller. In this example, the compressor is a reciprocating type.

[0141] In another example, as shown by FIG. 8, there is an initial single compressor 32a and one or more additional compressors 32b,32c may be added in parallel to the initial compressor. In this example, the compressor is a scroll type. The controller 75 (not specifically shown in FIG. 8) is arranged to turn on one compressor 32a, and once the suction pressure as measured by a sensor 44 has dropped sufficiently, the next compressor 32b can be engaged, once the suction pressure has dropped further, the next compressor 32c can be engaged, and so on.

[0142] Thus, in the example in FIG. 5, the electrical system 60 may be designed for a capacity of 10 kW 518 with perhaps a maximum peak capacity of 12 kW for short bursts. A first compressor is engaged at start up for approx. 10 minutes 520 requiring 5 kW from the inverter consuming 0.8 kWh, staying well within the inverter capacity. After approx. 10 minutes, with the refrigerant now less dense and the compressor requiring less power, a second compressor 32b is engaged. The combined compressors operate for 10 minutes 540 requiring 7.5 kW from the inverter consuming 1.3 kWh, staying within the inverter capacity. After another approx. 10 minutes, with the refrigerant less dense again, a third compressor 32c is engaged. The combined compressors 32a,b,c operate for approx. 10 minutes 560 requiring approx. 10 kW from the inverter consuming 1.7 kWh, staying at or around the inverter capacity.

[0143] Once the set point is reached, compressors can be disengaged and a single compressor 32a is operated for periods of time to maintain the set point temperature. Thus, the system of plural compressors reduces inrush start-up current and provides capacity modulation.

[0144] Embodiments of the present disclosure have been described with particular reference to the examples illustrated. However, it will be appreciated that variations and modifications may be made to the examples described within the scope of the present claims.