An electric vehicle and control method thereof

Abstract

An electric vehicle (EV) 100 comprises: plural power sources (e.g., hydrogen (H2) fuel cell 102, ultracapacitor / supercapacitor 104, battery 106); a high voltage (HV) power distribution module 108 for

Description

Field The present techniques generally relate to an electric vehicle and a method for controlling the operation of the electric vehicle. In particular, the present techniques provide an electric vehicle which makes more efficient use of power sources that reduces or removes the need for en route charging. Background Commercial vehicles are a crucial business tool for many companies, such as goods delivery companies. Such commercial vehicles need to carry many goods, travel long distances, and arrive at each destination on time. Typically, commercial vehicles, such as delivery vans and lorries, are diesel-powered vehicles. It is desirable to reduce reliance on fossil fuels and therefore, switch to using electric commercial vehicles. However, given that commercial vehicles typically travel much longer distances on a regular basis than cars and carry a much larger load, it is crucial that electric commercial vehicles are able to complete their required journeys without reducing the load they carry or by increasing the journey time significantly. The present applicant has therefore identified the need for an improved electric commercial vehicle. Summary In a first approach of the present techniques, there is provided an electric vehicle comprising: a plurality of power sources; a high voltage power distribution module for directing power from the plurality of power sources to operate the electric vehicle; and at least one processor coupled to memory for: determining an operation state of the electric vehicle, and controlling the high voltage power distribution module to direct power from at least one of the plurality of power sources based on at least the determined operation state. In one arrangement, the plurality of power sources may comprise one or more of any of: a hydrogen fuel cell, an ultracapacitor, and a battery. An advantage of the present techniques is how the plurality of power sources is managed. For example, in cases where the power sources comprise a hydrogen fuel cell, an ultracapacitor, and a battery, the power sources are managed so that a state of charge of the ultracapacitor and battery is such that they are ready for use during a next 'event', and how the fuel cell power is used to bring the ultracapacitor and battery into the required state of charge. The electric vehicle may be a commercial vehicle, i.e. any vehicle used for transporting goods. The commercial vehicle may be used for home deliveries, for instance. Thus, the electric vehicle may enable zero-emissions home deliveries. The electric vehicle may be a truck, articulated lorry, box truck, or van. The electric vehicle may be a large passenger vehicle, such as a sport utility vehicle, SUV, four-by-four vehicle, four-wheel drive vehicle, or off-road vehicle. It will be understood that this is merely an example, non-limiting list of possible vehicle types. Typically, electric vehicles comprise batteries that are charged by coupling the electric vehicle to an electricity source. However, there are a number of disadvantages associated with existing batteries for electric vehicles, including their size and weight, and the impact this has on performance of the electric vehicle. Batteries cannot be charged with a very high current without significantly impacting their lifetime, which means batteries need to be replaced relatively frequently. For example, battery performance often degrades after 3-5 years (depending on usage and charging), and after 6-8 years the degraded battery performance can impact the usability of the vehicle. Some electric vehicles use a fuel cell, such as a hydrogen fuel cell, which are able to generate electricity as long as there is sufficient fuel (e.g. hydrogen) and oxygen. In contrast, the electric vehicle of the present techniques comprises a plurality of power sources: a hydrogen fuel cell, an ultracapacitor and a battery. The combination of power sources enables the electric vehicle of the present techniques to use the most appropriate power source for different operation states of the electric vehicle. This advantageously improves performance of the electric vehicle, as it may be able to accelerate at the same rate, decelerate at the same rate, and reach the same speeds as a diesel-powered vehicle. Furthermore, having a choice of power sources means that it is possible to extend the life of each power source, which is advantageous as it means components such as batteries do not need to be replaced as often. Another advantage of having a plurality of power sources / energy systems on-board the electric vehicle is that they provide long range and high payload, without negatively impacting space and cost (as well as the weight of the vehicle or range of the vehicle), compared to a purely battery-based electric vehicle. Generally speaking, an ultracapacitor can be charged and discharged very quickly (e.g. in seconds), and so may be useful when a lot of power is required, such as when the electric vehicle is accelerating. Ultracapacitors are also typically capable of more than a million charge-discharge cycles, which means ultracapacitors may have a lifetime that is longer than the electric vehicle itself. Thus, using an ultracapacitor in the electric vehicle is advantageous because it overcomes the limitations of a battery. However, it may be useful to also have a battery power source. Having a battery power source as part of the plurality of power sources of the electric vehicle is advantageous because a battery can provide emergency power or a shunting ability. In cases where the electric vehicle needs to be moved without the hydrogen fuel cell, the battery enables the electric vehicle to be moved. The battery may enable the electric vehicle to be moved a short distance, e.g. 3-5km at low speeds, which may be sufficient to move the electric vehicle to a safe location while awaiting a recovery service. Thus, the battery power source provides sustained power in a similar way to the fuel cell, although for less time. As mentioned above, the at least one processor determines an operation state of the electric vehicle, and controls the high voltage power distribution module to direct power from at least one of the plurality of power sources based on at least the determined operation state. Thus, an advantage of the present techniques is that each power source / energy system is used in a way that is optimal for its particular function. Moreover, the present techniques advantageously enable the power sources to be used individually or in combination to meet the power demands for the determined operation state. The ultracapacitor may harvest energy (to recharge) at very high rates which would typically damage a battery, and so in a conventional battery electric vehicle this energy harvesting is often capped. The battery can supply a good steady energy over a long duration which the ultracapacitor cannot, and the hydrogen fuel cell provides "bulk energy" at a very low weight, size and cost. The combination of the power sources and the ability to switch from one to another or to use more than one power source at a time is significantly advantageous over conventional electric vehicles in terms of total cost of ownership (TCO) to the operator (e.g. a goods distribution company or haulage company). The operation states of the electric vehicle may include starting from stationary, accelerating from stationary, accelerating while moving, decelerating, braking, cruising / moving at a steady speed, and stopping. The determined operation state of the electric vehicle may have an associated power demand. Thus, controlling the high voltage power distribution module may comprise controlling the high voltage power distribution module to direct power from at least one of the plurality of power sources that is capable of meeting the power demand. Advantageously, this means that one or more power sources are chosen to control the operation of the electric vehicle at a particular point in time based on the power demand associated with the operation state at that point in time. This ensures that the power demand can be met, and the electric vehicle performs well regardless of the operation state of the electric vehicle. Specifically, the high voltage power distribution module may direct power from the following power sources to meet the power demand: the hydrogen fuel cell; the ultracapacitor; the ultracapacitor and the battery; and the ultracapacitor, battery and hydrogen fuel cell. Thus, in some cases the hydrogen fuel cell may be used on its own to operate the vehicle, in other cases the ultracapacitor may be used on its own, and in still other cases, a combination of the power sources may be used. Controlling the high voltage power distribution module may further comprise controlling the high voltage power distribution module to direct power from the hydrogen fuel cell to charge the ultracapacitor and / or battery. Thus, in certain operation states, energy may be released which could be used to recharge the ultracapacitor and / or the battery. This advantageously ensures that energy is not wasted during these operation states, which results in a more efficient electric vehicle. This may also avoid the need to stop en route to recharge the battery and / or ultracapacitor, as recharging may occur while the electric vehicle is in motion. The at least one processor may control the high voltage distribution module to direct power from the hydrogen fuel cell to charge the ultracapacitor and / or battery to maintain a state of charge of the ultracapacitor and / or battery between 20% and 80%. Maintaining a state of charge within this range helps the battery and / or ultracapacitor to achieve the longest lifetime (i.e. increases the overall lifetime of the component), and also reduces heat generation. The electric vehicle may further comprise route information defining a route to be taken by the electric vehicle, wherein the route information comprises road type information and topography information. The route information may be obtained via a GPS route planner, and may provide directions / travel instructions. The road type information may indicate, for each road in the route, the kind of road. Road types / kinds in the UK include minor routes, major routes, A roads, B roads, single carriageways, dual carriageways, and motorways, for example. The type of road along which the electric vehicle travels may impact the operation state of the electric vehicle. For example, an electric vehicle on a motorway may be in a steady state, i.e. may travel at a constant speed, for a significant time, because of the nature of a motorway. Similarly, an electric vehicle on a minor road in a built-up area may routinely accelerate and decelerate due to the variety of other road users on such a road (e.g. cyclists, pedestrians, and slow moving vehicles), and the obstacles that may appear on such a road (e.g. traffic lights, roundabouts, speed breakers, etc.). Topography information may indicate, for example, whether a route involves any hills, ascents, descents, relatively flat sections, and so on. This information may impact the speed of the electric vehicle and any changes in power needed to travel along sections of the route. For example, maintaining a certain speed while travelling up a hill may require more power than maintaining the same speed along a flat (non-inclined) section of road, which may impact which power sources are used to operate the electric vehicle. Similarly, descents may require more braking, which may enable the recharging of the ultracapacitor and / or battery. Topography information may therefore be used to predict the energy demand during various sections of the route. For example, if a route involves lots of downhill sections and then a large uphill section, the processor may determine that the ultracapacitor (and battery) should charge up during the downhill sections so that there is a significant amount of energy available to power the vehicle during the uphill section. Similarly, if there is not enough charge in the battery, the fuel cell could be engaged during the uphill section. The saves fuel while also making the best use of kinetic energy. The at least one processor may therefore control the high voltage power distribution module to direct power from at least one of the plurality of power sources based on the determined operation state and the route information. That is, information on the road along which the electric vehicle is travelling while in the determined operation state may be used to determine which power source(s) to use to control the electric vehicle. Cruising or accelerating up a hill may require multiple power sources to be used, compared to cruising along a flat section of road, for example. The route information may further comprise live traffic information. In this case, the at least one processor may use the live traffic information to control the high voltage power distribution module. For example, when the electric vehicle is travelling along a motorway, it may be assumed that the electric vehicle is going to be in the steady state for a long time. However, the live traffic information may indicate that an accident has occurred further ahead, and so the electric vehicle will need to start decelerating, travel at a lower speed, or move in stop-start traffic. The live traffic information may be used to anticipate which power source is going to be needed in response to the changing road conditions. This can then be used to switch (if necessary) the power source(s) being used to operate the electric vehicle. The at least one processor may determine at least one dynamic control parameter for the electric vehicle, and may use the determined at least one dynamic control parameter to control the high voltage power distribution module. This may be useful to optimise the power consumption of the electric vehicle. The at least one dynamic control parameter may be determined by using one or more of: current state of charge of the ultracapacitor and / or battery; expected future state of charge of the ultracapacitor and / or battery; and route information. The state of charge is defined as the ratio of the available capacity and the maximum possible charge that can be stored in a battery. A fully charged battery has a state of charge of 1 (or 100%), while a fully discharged battery has a state of charge of 0 (or 0%). The current state of charge of the ultracapacitor and / or battery may be less than 100% if the power sources have been used since the electric vehicle began its journey. It may be desirable to, for example, limit the amount that the ultracapacitor and / or battery is used when the state of charge is less than a certain value, which may also depend on how much of the journey is remaining. This may be achieved by having a "floating" state of charge target value, such that the battery and ultracapacitor are ready for the next part / section of a journey or route, without needing to be fully charged, as a state of charge of 1 may limit energy recovery (i.e. recharging during the journey) or impact efficiency. Thus, the at least one dynamic control parameter may be one or more of: a speed limit, an acceleration rate limit, a deceleration rate, and a braking rate. Placing a speed limit or a limit on the acceleration rate on the electric vehicle may reduce the power consumed by the electric vehicle, which may ensure the electric vehicle is able to complete its journey in the most efficient manner (or reach a battery recharging station or a hydrogen filling station). The dynamic control parameter may not be a limitation or restriction. For example, the dynamic control parameter may be a deceleration rate or braking rate. Increasing the braking rate or deceleration rate can release energy that can be used to recharge the ultracapacitor and / or battery. This may be advantageous if the state of charge of the ultracapacitor and / or battery is less than a certain value. As mentioned above, the operation states of the electric vehicle may include starting from stationary, accelerating from stationary, accelerating while moving, decelerating, braking, cruising / moving at a steady speed, and stopping. When the determined operation state is acceleration or moving from stationary, the at least one processor may control the high voltage power distribution module to direct power from the ultracapacitor to operate the electric vehicle. This is because the ultracapacitor can be discharged quickly, i.e. can deliver energy quickly, which is useful during acceleration. When the determined operation state is steady speed, the at least one processor may control the high voltage power distribution module to use the hydrogen fuel cell to control / drive the vehicle, and to direct power from the hydrogen fuel cell to recharge the ultracapacitor and / or battery. This is advantageous because fast energy delivery is not required when travelling at a constant speed, and so power from the hydrogen fuel cell can safely be used to recharge the ultracapacitor and / or battery during this operation state. Furthermore, the fuel cell is most efficient during steady state operation, so this uses the fuel cell optimally. The electric vehicle may further comprise a regenerative braking system. Regenerative braking is an energy recovery mechanism that enables the wasted energy from the process of slowing down a vehicle to be reused for another purpose. In this case, when the determined operation state is braking, the at least one processor may control the high voltage power distribution module to direct power from the regenerative braking system to recharge the ultracapacitor. The at least one processor may control the high voltage power distribution module to direct power from the plurality of power sources to any one or more of: a vehicle controller area network, CAN; a water pump; a brake pump; a steering pump; a motor; a motor driver; a motor controller; a heat exchanger; a vehicle charger; and an oil pump. It will be understood that this is a non-exhaustive list of components within the electric vehicle which may need to be powered. The electric vehicle may further comprise at least one lookup table defining which power source to use for each operation state. The lookup table may include information on which power source to use in different traffic conditions, on different road types, and so on. Additionally or alternatively, the electric vehicle may further comprise an artificial intelligence, AI, module for determining which power source to use based on at least the determined operation state. The AI module comprises a trained AI model. The AI module may continuously or regularly monitor a number of parameters and input these into the trained AI model to determine which power sources should be used at a given point in time, such as the operation state, route information, topography information, road type information, live traffic information, state of charge of the ultracapacitor and / or battery, remaining journey time or remaining journey distance, weather conditions, and so on. As mentioned above, an advantage of the present techniques is how the plurality of power sources is managed so that a state of charge of the ultracapacitor and battery is such that they are ready for use during a next 'event', and how the fuel cell power is used to bring the ultracapacitor and battery into the required state of charge. The AI module has been trained to understand how the electric vehicle is used, and the traffic situation the vehicle is in, so that the AI module can modify the state of charge of the battery and ultracapacitor with this information in mind. This knowledge may also be shared across vehicles (or a set of vehicles belonging to the same user). The AI module may be configured to: predict an upcoming operation state of the electric vehicle; and output a suggested command for controlling the high voltage power distribution module to direct power from at least one of the plurality of power sources based on the determined operation state and the predicted upcoming operation state, to ensure the power sources are ready for the upcoming operation state. The AI module may predict an upcoming operation state using any one or more of: route information; live traffic information; weather information; and remaining journey distance. In cases where the AI module predicts the upcoming operation state using at least the route information, the present techniques advantageously provide a way of managing the power sources for a predetermined route. That is, the route information may correspond to a predetermined route to be taken by the electric vehicle, and the predicting of the upcoming operation state may use the predetermined route to ensure the readiness of the power sources at multiple points along the route. It will be appreciated that the present techniques may also be applied to cases where the prediction of the upcoming operation state uses, for example, live traffic information and weather information. For example, it may not be possible to travel along the predetermined route because an accident or roadworks or other incident appears, meaning that a diversion needs to be followed or an alternative route for some or all of the predetermined route needs to be found. In this case, some or all of a predetermined route is not available. In other cases, a predetermined route may not exist at all between the starting point and destination. The present techniques advantageously provide "on-the-fly" predictions of the upcoming operation state when the electric vehicle is not taking a predetermined route. That is, the upcoming operation state can still be determined when a predetermined route is not being followed, because the AI module can still determine where the vehicle is and where the vehicle is likely to be in a few minutes (e.g. staying on the same motorway, or going around an upcoming roundabout, or approaching traffic lights, or approaching an area with a different speed limit, etc.). In a second approach of the present techniques, there is provided a method for controlling the operation of an electric vehicle, the electric vehicle comprising a plurality of power sources, the method comprising: determining an operation state of the electric vehicle; and controlling a high voltage power distribution module of the electric vehicle to direct power from at least one of the plurality of power sources based on at least the determined operation state. The plurality of power sources may include any one or more of any of: a hydrogen fuel cell, an ultracapacitor, and a battery. The features described above with respect to the first approach apply equally to the second approach and therefore, for the sake of conciseness, are not repeated. As noted above, the determined operation state of the electric vehicle may have an associated power demand. In this case, controlling the high voltage power distribution module may comprise: controlling the high voltage power distribution module to direct power from at least one of the plurality of power sources that is capable of meeting the power demand. Controlling the high voltage power distribution module may further comprise: controlling the high voltage power distribution module to direct power from the hydrogen fuel cell to charge the ultracapacitor and / or battery. The method may further comprise directing power from the hydrogen fuel cell to charge the ultracapacitor and / or battery to maintain a state of charge of the ultracapacitor and / or battery between 20% and 80%. The method may further comprise: obtaining route information defining a route to be taken by the electric vehicle, wherein the route information comprises road type information and topography information, wherein controlling the high voltage power distribution module to direct power from at least one of the plurality of power sources is based on the determined operation state and the route information. The method may further comprise: determining at least one dynamic control parameter for the electric vehicle, and using the determined at least one dynamic control parameter to control the high voltage power distribution module. In a related approach of the present techniques, there is provided a non-transitory data carrier carrying processor control code to implement any of the methods, processes and techniques described herein. As will be appreciated by one skilled in the art, the present techniques may be embodied as a system, method or computer program product. Accordingly, present techniques may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present techniques may take the form of a computer program product embodied in a computer readable medium having computer readable program code embodied thereon. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable medium may be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Computer program code for carrying out operations of the present techniques may be written in any combination of one or more programming languages, including object oriented programming languages and conventional procedural programming languages. Code components may be embodied as procedures, methods or the like, and may comprise sub-components which may take the form of instructions or sequences of instructions at any of the levels of abstraction, from the direct machine instructions of a native instruction set to high-level compiled or interpreted language constructs. Embodiments of the present techniques also provide a non-transitory data carrier carrying code which, when implemented on a processor, causes the processor to carry out any of the methods described herein. The techniques further provide processor control code to implement the above-described methods, for example on a general purpose computer system or on a digital signal processor (DSP). The techniques also provide a carrier carrying processor control code to, when running, implement any of the above methods, in particular on a non-transitory data carrier. The code may be provided on a carrier such as a disk, a microprocessor, CD- or DVD-ROM, programmed memory such as non-volatile memory (e.g. Flash) or read-only memory (firmware), or on a data carrier such as an optical or electrical signal carrier. Code (and / or data) to implement embodiments of the techniques described herein may comprise source, object or executable code in a conventional programming language (interpreted or compiled) such as C, or assembly code, code for setting up or controlling an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array), or code for a hardware description language such as Verilog (RTM) or VHDL (Very high speed integrated circuit Hardware Description Language). As the skilled person will appreciate, such code and / or data may be distributed between a plurality of coupled components in communication with one another. The techniques may comprise a controller which includes a microprocessor, working memory and program memory coupled to one or more of the components of the system. It will also be clear to one of skill in the art that all or part of a logical method according to embodiments of the present techniques may suitably be embodied in a logic apparatus comprising logic elements to perform the steps of the above-described methods, and that such logic elements may comprise components such as logic gates in, for example a programmable logic array or application-specific integrated circuit. Such a logic arrangement may further be embodied in enabling elements for temporarily or permanently establishing logic structures in such an array or circuit using, for example, a virtual hardware descriptor language, which may be stored and transmitted using fixed or transmittable carrier media. In an embodiment, the present techniques may be implemented using multiple processors or control circuits. The present techniques may be adapted to run on, or integrated into, the operating system of an apparatus. In an embodiment, the present techniques may be realised in the form of a data carrier having functional data thereon, said functional data comprising functional computer data structures to, when loaded into a computer system or network and operated upon thereby, enable said computer system to perform all the steps of the above-described method. Brief description of the drawings Implementations of the present techniques will now be described, by way of example only, with reference to the accompanying drawings, in which: Figure 1 is a block diagram of an electric vehicle; Figure 2A is a block diagram of a system for training an AI model; Figure 2B is a schematic diagram showing how the AI model may be trained; Figure 2C is a schematic diagram showing how the trained AI model may be used; and Figure 3 is a flowchart of example steps performed to control the operation of the electric vehicle. Detailed description of the drawings Broadly speaking, embodiments of the present techniques provide an electric commercial vehicle which makes more efficient use of power sources that reduces or removes the need for en route charging. The electric vehicle of the present techniques comprises a plurality of power sources: a hydrogen fuel cell, an ultracapacitor and a battery. The combination of power sources enables the electric vehicle of the present techniques to use the most appropriate power source for different operation states of the electric vehicle. This advantageously improves performance of the electric vehicle, as it may be able to accelerate at the same rate, decelerate at the same rate, and reach the same speeds as a diesel-powered vehicle. Furthermore, having a choice of power sources means that it is possible to extend the life of each power source, which is advantageous as it means components such as batteries do not need to be replaced as often. Typically, vehicle manufacturers design a combustion engine powered vehicle where, due to the abundance of fuel and the flexible nature of the engine, they design a system for maximum power. A mission profile of a vehicle component (e.g. engine) is a simplified representation of all relevant static and dynamic load conditions that a vehicle may be exposed to within its lifecycle. The manufacturer may pick the most arduous part of the mission profile of the vehicle (e.g. Grossglockner for braking and a 12% gradient maximum acceleration), and design the powertrain to suit that because the other load conditions reside under the mission profile curve. Now, vehicle manufacturers are applying the same approach to electric vehicle design, without taking into account the ability to recover energy. In contrast, the present applicant advantageously considered the average power demand (e.g. about 35kW) and realised that a smaller fuel cell could be used for an electric vehicle in combination with additional power sources to provide additional power when needed. The present applicant has selected the fuel cell to be as efficient as possible for the average power demand (which may require a 45-65kW fuel cell). Then, the present applicants consider the rest of the cycle, maximum braking, different driving conditions, and determine an energy swing is required to account for the changing power demands during a journey (which may be from -23kW to 120kW, for example). This information is used to select an ultracapacitor for the swing frequency and amplitude, and selected a battery for the amplitude that the ultracapacitor cannot manage. In this way, a smaller fuel cell may be used which decreases the weight of the electric vehicle, and the additional power sources ensure that the electric vehicle is able to meet power demands (e.g. when accelerating, going uphill, braking, and so on). Figure 1 is a block diagram of an electric vehicle 100. The electric vehicle 100 may be a commercial vehicle, i.e. any vehicle used for transporting goods. The commercial vehicle may be used for home deliveries, for instance. Thus, the electric vehicle 100 may enable zero-emissions home deliveries. The electric vehicle 100 may be a truck, articulated lorry, box truck, or van. It will be understood that this is merely an example, non-limiting list of possible vehicle types. The electric vehicle 100 comprises a plurality of power sources. The plurality of power sources comprise: a hydrogen fuel cell 102, an ultracapacitor 104, and a battery 106. The electric vehicle comprises a high voltage power distribution module, HVPDM, 108. The HVPDM 108 directs power from the plurality of power sources to operate the electric vehicle 100. Generally speaking, the term "operate the electric vehicle" used herein means to operate the components of the electric vehicle which cause the electric vehicle to move, such as, but not limited to, a motor 118. The battery 106 may be any suitable battery and be used to power the electric vehicle during any part of a journey. For example, the battery may be used to power the electric vehicle when the electric vehicle is starting, or may be used to provide emergency power or a shunting ability. Thus, the size or voltage of the battery may depend on what the battery 106 is used for in the electric vehicle. The electric vehicle 100 may comprise at least one processor 110. The at least one processor 102 may comprise one or more of: a microprocessor, a microcontroller, and an integrated circuit. The electric vehicle may comprise memory 112. The memory 112 may comprise volatile memory, such as random access memory (RAM), for use as temporary memory, and / or non-volatile memory such as Flash, read only memory (ROM), or electrically erasable programmable ROM (EEPROM), for storing data, programs, or instructions, for example. The at least one processor 110 may determine an operation state of the electric vehicle 100, and control the high voltage power distribution module 108 to direct power from at least one of the plurality of power sources based on at least the determined operation state. The operation states of the electric vehicle 100 may include starting from stationary, accelerating from stationary, accelerating while moving, decelerating, braking, cruising / moving at a steady speed, and stopping. The operation state of the electric vehicle 100 may be determined by the processor 110 using information gathered from sensors or other devices in the electric vehicle 100. For example, a speedometer of the electric vehicle 100 may be used to determine the operation state. Similarly, a braking system of the electric vehicle 100, or sensors associated with a braking system, may be used to determine the operation state. The determined operation state of the electric vehicle 100 may have an associated power demand. Thus, the at least one processor 110 may control the high voltage power distribution module to direct power from at least one of the plurality of power sources that is capable of meeting the power demand. Advantageously, this means that one or more power sources are chosen to control the operation of the electric vehicle at a particular point in time based on the power demand associated with the operation state at that point in time. This ensures that the power demand can be met, and the electric vehicle performs well regardless of the operation state of the electric vehicle. Specifically, the high voltage power distribution module 108 may direct power from the following power sources to meet the power demand: the hydrogen fuel cell 102; the ultracapacitor 104; the ultracapacitor 104 and the battery 106; and the ultracapacitor 104, battery 106 and hydrogen fuel cell 102. Thus, in some cases the hydrogen fuel cell 102 may be used on its own to operate the vehicle, in other cases the ultracapacitor 104 may be used on its own, and in still other cases, a combination of the power sources may be used. The at least one processor 110 may control the high voltage power distribution module 108 to direct power from the hydrogen fuel cell 102 to charge the ultracapacitor 104 and / or battery 106. Thus, in certain operation states, energy may be released which could be used to recharge the ultracapacitor 104 and / or the battery 106. This advantageously ensures that energy is not wasted during these operation states, which results in a more efficient electric vehicle. This may also avoid the need to stop en route to recharge the battery and / or ultracapacitor, as recharging may occur while the electric vehicle is in motion. The at least one processor 110 may control the high voltage distribution module 108 to direct power from the hydrogen fuel cell 102 to charge the ultracapacitor 104 and / or battery 106 to maintain a state of charge of the ultracapacitor and / or battery between 20% and 80%. Maintaining a state of charge within this range helps the battery and / or ultracapacitor to achieve the longest lifetime (i.e. increases the overall lifetime of the component), and also reduces heat generation. The electric vehicle 100 may further comprise route information 114 defining a route to be taken by the electric vehicle 100. The route information 114 may comprise road type information and topography information. The route information 114 may be obtained via a GPS route planner, and may provide directions / travel instructions. The road type information may indicate, for each road in the route, the kind of road. Road types / kinds in the UK include minor routes, major routes, A roads, B roads, single carriageways, dual carriageways, and motorways, for example. The type of road along which the electric vehicle travels may impact the operation state of the electric vehicle. For example, an electric vehicle on a motorway may be in a steady state, i.e. may travel at a constant speed, for a significant time, because of the nature of a motorway. Similarly, an electric vehicle on a minor road in a built-up area may routinely accelerate and decelerate due to the variety of other road users on such a road (e.g. cyclists, pedestrians, and slow moving vehicles), and the obstacles that may appear on such a road (e.g. traffic lights, roundabouts, speed breakers, etc.). Topography information may indicate, for example, whether a route involves any hills, ascents, descents, relatively flat sections, and so on. This information may impact the speed of the electric vehicle and any changes in power needed to travel along sections of the route. For example, maintaining a certain speed while travelling up a hill may require more power than maintaining the same speed along a flat (non-inclined) section of road, which may impact which power sources are used to operate the electric vehicle. Similarly, descents may require more braking, which may enable the recharging of the ultracapacitor and / or battery. The at least one processor 110 may therefore control the high voltage power distribution module 108 to direct power from at least one of the plurality of power sources based on the determined operation state and the route information. That is, information on the road along which the electric vehicle is travelling while in the determined operation state may be used to determine which power source(s) to use to control the electric vehicle. Cruising or accelerating up a hill may require multiple power sources to be used, compared to cruising along a flat section of road, for example. The route information 114 may further comprise live traffic information. In this case, the at least one processor 110 may use the live traffic information to control the high voltage power distribution module 108. For example, when the electric vehicle is travelling along a motorway, it may be assumed that the electric vehicle is going to be in the steady state for a long time. However, the live traffic information may indicate that an accident has occurred further ahead, and so the electric vehicle will need to start decelerating, travel at a lower speed, or move in stop-start traffic. The live traffic information may be used to anticipate which power source is going to be needed in response to the changing road conditions. This can then be used to switch (if necessary) the power source(s) being used to operate the electric vehicle. The at least one processor 110 may determine at least one dynamic control parameter for the electric vehicle 100, and may use the determined at least one dynamic control parameter to control the high voltage power distribution module 108. This may be useful to optimise the power consumption of the electric vehicle 100. The at least one dynamic control parameter may be determined by using one or more of: current state of charge of the ultracapacitor and / or battery; expected future state of charge of the ultracapacitor and / or battery; and route information. The state of charge is defined as the ratio of the available capacity and the maximum possible charge that can be stored in a battery. A fully charged battery has a state of charge of 1 or 100%, while a fully discharged battery has a state of charge of 0 or 0%. The current state of charge of the ultracapacitor and / or battery may be less than 100% if the power sources have been used since the electric vehicle began its journey. It may be desirable to, for example, limit the amount that the ultracapacitor and / or battery is used when the state of charge is less than a certain value, which may also depend on how much of the journey is remaining. Thus, the at least one dynamic control parameter may be one or more of: a speed limit, an acceleration rate limit, a deceleration rate, and a braking rate. Placing a speed limit or a limit on the acceleration rate on the electric vehicle may reduce the power consumed by the electric vehicle, which may ensure the electric vehicle is able to complete its journey (or reach a battery recharging station). The dynamic control parameter may not be a limitation or restriction. For example, the dynamic control parameter may be a deceleration rate or braking rate. Increasing the braking rate or deceleration rate can release energy that can be used to recharge the ultracapacitor and / or battery. This may be advantageous if the state of charge of the ultracapacitor and / or battery is less than a certain value. As mentioned above, the operation states of the electric vehicle 100 may include starting from stationary, accelerating from stationary, accelerating while moving, decelerating, braking, cruising / moving at a steady speed, and stopping. When the determined operation state is acceleration or moving from stationary, the at least one processor 110 may control the high voltage power distribution module 108 to direct power from the ultracapacitor 104 to operate the electric vehicle 100. This is because the ultracapacitor 104 can be discharged quickly, i.e. can deliver energy quickly, which is useful during acceleration. When the determined operation state is steady speed, the at least one processor 110 may control the high voltage power distribution module 108 to use the hydrogen fuel cell to control / drive the vehicle, and to direct power from the hydrogen fuel cell 102 to recharge the ultracapacitor 104 and / or battery 106. This is advantageous because fast energy delivery is not required when travelling at a constant speed, and so power from the hydrogen fuel cell can safely be used to recharge the ultracapacitor and / or battery during this operation state. For example, when the electric vehicle is in the steady state, the power demand of the electric vehicle may be 37kW. The fuel cell may be run at 45kW, such that the excess is used to recharge the ultracapacitor 104 and / or battery 106 depending on the state of charge. The recharging may also occur when the predicted power demand of the next stage / section of the journey is high. For instance, if a route involves lots of downhill sections and then a large uphill section, the processor may determine that the ultracapacitor (and battery) should charge up during the downhill sections so that there is a significant amount of energy available to power the vehicle during the uphill section. The electric vehicle 100 may further comprise a regenerative braking system 116. Regenerative braking is an energy recovery mechanism that enables the wasted energy from the process of slowing down a vehicle to be reused for another purpose. In this case, when the determined operation state is braking, the at least one processor 110 may control the high voltage power distribution module 108 to direct power from the regenerative braking system 116 to recharge the ultracapacitor 104. The at least one processor 100 may control the high voltage power distribution module 108 to direct power from the plurality of power sources to any one or more of: a vehicle controller area network, CAN; a water pump; a brake pump; a steering pump; a motor 118; a motor driver; a motor controller; a heat exchanger; a vehicle charger; and an oil pump. It will be understood that this is a non-exhaustive list of components within the electric vehicle which may need to be powered. The electric vehicle may further comprise at least one lookup table 120 defining which power source to use for each operation state. The lookup table(s) 120 may include information on which power source to use in different traffic conditions, on different road types, and so on. Additionally or alternatively, the electric vehicle may further comprise an artificial intelligence, AI, module 122 for determining which power source to use based on at least the determined operation state. The AI module 122 comprises a trained AI model 124. As explained below, the AI model 124 may be trained by the electric vehicle 100 or by a central server. In any case, knowledge learned by one electric vehicle 100 may be shared with other electric vehicles, e.g. other electric vehicles owned or used by the same user / customer, so that all vehicles in a fleet benefit from the knowledge. The AI module 122 may continuously or regularly monitor a number of parameters and input these into the AI model 124 to determine which power sources should be used at a given point in time, such as the operation state, route information, topography information, road type information, live traffic information, state of charge of the ultracapacitor and / or battery, remaining journey time or remaining journey distance, weather conditions, and so on. The AI model 124 may also be able to predict the power needed to complete the remaining journey or an upcoming section of the journey, and use this prediction to determine whether, for example, the ultracapacitor and / or battery should be recharged now. As mentioned above, an advantage of the present techniques is how the plurality of power sources is managed so that a state of charge of the ultracapacitor and battery is such that they are ready for use during a next 'event', and how the fuel cell power is used to bring the ultracapacitor and battery into the required state of charge. An event may be a stage of a journey which requires a different power usage. For example, if a vehicle is travelling at a steady speed (current event) but there is traffic or traffic lights ahead, the next event will be a slowing down (deceleration) or stopping of the vehicle. Similarly, if a vehicle is travelling on a flat road (current event) but there is a hill coming-up, the next event will be the additional power required to drive up the hill / incline at the same speed. The AI module has been trained to understand how the electric vehicle is used, and the traffic situation the vehicle is in, so that the AI module can modify the state of charge of the battery and ultracapacitor with this information in mind. This knowledge may also be shared across vehicles (or a set of vehicles belonging to the same user). The AI module may be able to predict, based on, for example, the current operation state of the vehicle 100 (such as a current speed), on live traffic information, on topography information, the remaining journey time or distance, an upcoming section of the journey, etc, what event is likely to occur next, and use this prediction to ensure that the plurality of power sources are ready for the next event. Figure 2A is a block diagram of a system for training an AI model used by the AI module 122 of the electric vehicle. The system comprises a plurality of electric vehicles 100-1, 100-2, 100-3, which are electric vehicles of the type shown in Figure 1. Although only three vehicles are shown in Figure 2 for the sake of simplicity, it will be understood that the system may comprise any number of electric vehicles. The system may comprise a central server 200. The central server 200 may comprise an AI model 202. The AI model may be trained using data collected by the plurality of electric vehicles. For example, each electric vehicle may be able to share its data with the central server 200, as indicated by the arrows in Figure 2A. Figure 2B is a schematic diagram showing how the AI model may be trained using data collected by the plurality of electric vehicles. The data 204 (shown as "Base Data" in Figure 2B) collected by each electric vehicle may include, for example, data on the power sources of the electric vehicle (e.g. fuel cell, ultracapacitor and / or battery performance), data on the load (weight) of the vehicle during each journey or during different parts of the journey, data on the type of electric vehicle (e.g. model type, size, weight without goods on board, whether it is refrigerated, hydrogen tank size, hydrogen tank fill time, and so on), data on journeys travelled by the vehicle, data on journeys travelled by other vehicles (i.e. fleet tracking information), the power source(s) used during different parts of the journey, whether the vehicle was able to complete its journey (i.e. whether there was sufficient energy available to complete the journey), vehicle performance during each journey or during segments of a journey, vehicle range, whether the ultracapacitor and / or battery were able to be charged during a journey, etc. It will be understood that this is a non-limiting and non-exhaustive list of types of data that may be used to train the AI model 202. Such data may enable an AI 202 model to be trained, retrained or updated to improve the AI model's ability to pick the correct power source(s) for the vehicle, depending on any of the parameters mentioned above. The training process may also be termed a "learning" process (as shown in Figure 2B). During the training process, the AI model 202 may learn relationships between the data 204 and the electric vehicle performance. These learnt relationships may be used to determine which power sources should be used at a given point in time and / or to predict the power needed to complete the remaining journey or an upcoming section of the journey. For example, the AI model may learn the effect that an air conditioning system of the electric vehicle has on the fuel cell, ultracapacitor and / or battery performance. Using this learnt relationship, the AI model may be able to determine which power sources should be used at a given point in time e.g. if a given proportion of the ultracapacitor's power should be directed to the air conditioning system. It will be appreciated that the AI model may learn any number of such relationships or effects, such as driving style effect, fuel cell transient time, Cooling capability versus external conditions, regen capability, motor efficiency, system efficiency, air conditions effect, driving style effect, effect of Load, effect of traffic, effect of Geography and road conditions. Whilst the training process of Figure 2B is described in relation the training of the AI model on the central server 200, it will be appreciated that the AI model may also be trained on any of the electric vehicles 100-1, 100-2, 100-3. The central server 200 may use data obtained from the electric vehicles to train, retrain or update the AI model 202 centrally. Once trained, retrained or updated, the central server 200 may provide the AI model 202 to each electric vehicle. A local instance of the AI model 202 may be stored on each electric vehicle 100 and used by the AI module 122. This advantageously means that the AI model 202 can be trained using data from a plurality of vehicles, i.e. a fleet of vehicles (which may be the same or different). Figure 2C is a schematic diagram showing how an AI module may be used on each electric vehicle. As noted above, each electric vehicle may further comprise an AI module 122 having a trained AI model 124. The AI model trained centrally 202 may be provided to each vehicle and stored as the trained model 124 in the AI module 122. The AI module may continuously or regularly monitor a number of parameters 126 and provide them as input to the AI model 124. These parameters may be, for example, the state of charge of the battery, the fuel cell power, the level of hydrogen in the hydrogen tank, the ultracapacitor state of charge, vehicle speed, vehicle load, route information, and so on. The output 128 of the AI module 122 may be a determination of which power sources to use based on at least the determined operation state and / or a prediction of the power needed to complete the remaining journey or an upcoming section of the journey. Returning to Figure 2A, alternatively, the electric vehicles may share their data with the other electric vehicles in the system, such that each electric vehicle can update its own AI model based on the data from the other vehicles. Further alternatively, the electric vehicles may share their data with a subgroup of the other electric vehicles in the system. For example, the electric vehicles may share their data with the other electric vehicles in a fleet of vehicles used for providing delivery services, wherein the fleet of vehicles corresponds to a sub-group of all other vehicles in the system. Further alternatively, the electric vehicles may share their data with a single other electric vehicle in the system, which assumes the role of a central server for the vehicles. This electric vehicle gathers all the data from the other electric vehicles, uses this data to train, retrain or update its AI model. Once trained, retrained or updated, this electric vehicle transmits the AI model to the other vehicles in the system for use. In other words, the AI model may be trained using federated learning. It will be understood that whichever training scheme is used, each vehicle may share raw data, or may only share model parameters (e.g. model weights based on the vehicle's own data). Figure 3 is a flowchart of example steps performed to control the operation of the electric vehicle of Figure 1. The method comprises: determining an operation state of the electric vehicle (step S100). The operation states of the electric vehicle 100 may include starting from stationary, accelerating from stationary, accelerating while moving, decelerating, braking, cruising / moving at a steady speed, and stopping. The operation state of the electric vehicle 100 may be determined using information gathered from sensors or other devices in the electric vehicle 100. For example, a speedometer of the electric vehicle 100 may be used to determine the operation state. Similarly, a braking system of the electric vehicle 100, or sensors associated with a braking system, may be used to determine the operation state. The method may comprise controlling a high voltage power distribution module 108 of the electric vehicle 100 to direct power from at least one of the plurality of power sources based on at least the determined operation state (step S102). For example, if it is determined that a change in velocity over time is increasing (i.e. the electric vehicle is accelerating), the hydrogen fuel cell 102 may be used to keep the electric vehicle moving if the hydrogen fuel cell 102 can, on its own, cope with the power demand associated with this operation state. If it is determined that the hydrogen fuel cell 102 cannot cope with the power demand itself, then the hydrogen fuel cell 102 may be used in combination with the ultracapacitor 102 and / or battery 104 may be used to keep the electric vehicle moving and meet the power demand associated with this operation state. In another example, if it is determined that a change in velocity over time is decreasing (i.e. the electric vehicle is decelerating), the power drawn from the hydrogen fuel cell 102 may be ramped down. Advantageously, the present techniques utilise the ramp down energy by using it to recharge the ultracapacitor 104 and / or battery 106. In contrast, conventional systems waste the ramp-down energy through a discharge resistor. Thus, the present electric vehicle is more fuel and energy efficient. In another example, if it is determined that a braking rate is increasing over time (i.e. the electric vehicle's brakes are being applied and the electric vehicle is decelerating), the power drawn from the hydrogen fuel cell 102 may be ramped down, and the ultracapacitor 104 and / or battery 106 is coupled to the high voltage power distribution module 108 for recharging, as mentioned above. Step S102 of Figure 3 may be performed using at least one lookup table 120 defining which power source to use for each operation state. The lookup table(s) may include information on which power source to use in different traffic conditions, on different road types, and so on. Additionally or alternatively, step S102 may be performed using an artificial intelligence, AI, module 122 for determining which power source to use based on at least the determined operation state. The AI module 122 may continuously or regularly monitor a number of parameters and input these into the trained AI model 124 to determine which power sources should be used at a given point in time, such as the operation state, route information, topography information, road type information, live traffic information, state of charge of the ultracapacitor and / or battery, remaining journey time or remaining journey distance, weather conditions, and so on. Those skilled in the art will appreciate that while the foregoing has described what is considered to be the best mode and where appropriate other modes of performing present techniques, the present techniques should not be limited to the specific configurations and methods disclosed in this description of the preferred embodiment. Those skilled in the art will recognise that present techniques have a broad range of applications, and that the embodiments may take a wide range of modifications without departing from any inventive concept as defined in the appended claims.

Claims

1. An electric vehicle comprising:a plurality of power sources;a high voltage power distribution module for automatically and dynamically directing power from the plurality of power sources to operate the electric vehicle using the most appropriate power source for different operation states of the electric vehicle; andat least one processor coupled to memory for:determining an operation state of the electric vehicle, andcontrolling the high voltage power distribution module to direct power from at least one of the plurality of power sources based on at least the determined operation state.

2. The electric vehicle as claimed in claim 1 wherein the plurality of power sources comprise one or more of any of: a hydrogen fuel cell, an ultracapacitor, and a battery.

3. The electric vehicle as claimed in claim 1 or 2 wherein:the determined operation state of the electric vehicle has an associated power demand; andcontrolling the high voltage power distribution module comprises controlling the high voltage power distribution module to direct power from at least one of the plurality of power sources that is capable of meeting the power demand.

4. The electric vehicle as claimed in claim 3 when dependent on claim 2, wherein the high voltage power distribution module directs power from the following power sources to meet the power demand: the hydrogen fuel cell; the ultracapacitor; the ultracapacitor and the battery; and the ultracapacitor, battery and hydrogen fuel cell.

5. The electric vehicle as claimed in claim 3 or 4 wherein controlling the high voltage power distribution module further comprises controlling the high voltage power distribution module to direct power from the hydrogen fuel cell to charge the ultracapacitor and / or battery.

6. The electric vehicle as claimed in claim 5 wherein the at least one processor controls the high voltage distribution module to direct power from the hydrogen fuel cell to charge the ultracapacitor and / or battery to maintain a state of charge of the ultracapacitor and / or battery between 20% and 80% for prolonging lifetime of the ultracapacitor and / or battery.

7. The electric vehicle as claimed in any preceding claim further comprising route information defining a route to be taken by the electric vehicle, wherein the route information comprises road type information and topography information.

8. The electric vehicle as claimed in claim 7 wherein the at least one processor controls the high voltage power distribution module to direct power from at least one of the plurality of power sources based on the determined operation state and the route information.

9. The electric vehicle as claimed in claim 8 wherein the route information further comprises live traffic information, and the at least one processor uses the live traffic information to control the high voltage power distribution module.

10. The electric vehicle as claimed in any preceding claim wherein the at least one processor determines at least one dynamic control parameter for the electric vehicle, and uses the determined at least one dynamic control parameter to control the high voltage power distribution module.

11. The electric vehicle as claimed in claim 10 wherein the processor determines the at least one dynamic control parameter using one or more of: current state of charge of the ultracapacitor and / or battery; expected future state of charge of the ultracapacitor and / or battery; and route information.

12. The electric vehicle as claimed in claim 10 or 11 wherein the at least one dynamic control parameter is one or more of: a speed limit, an acceleration rate limit, a deceleration rate, and a braking rate.

13. The electric vehicle as claimed in any one of claims 1 to 12 wherein the determined operation state is acceleration or moving from stationary, and the at least one processor controls the high voltage power distribution module to direct power from the ultracapacitor to operate the electric vehicle.

14. The electric vehicle as claimed in any one of claims 1 to 12 wherein the determined operation state is steady speed, and the at least one processor controls the high voltage power distribution module to direct power from the hydrogen fuel cell to recharge the ultracapacitor and / or battery.

15. The electric vehicle as claimed in any preceding claim further comprising a regenerative braking system.

16. The electric vehicle as claimed in claim 15 wherein the determined operation state is braking, and the at least one processor controls the high voltage power distribution module to direct power from the regenerative braking system to recharge the ultracapacitor.

17. The electric vehicle as claimed in any preceding claim wherein the at least one processor controls the high voltage power distribution module to direct power from the plurality of power sources to any one or more of: a vehicle controller area network, CAN; a water pump; a brake pump; a steering pump; a motor; a motor driver; a motor controller; a heat exchanger; a vehicle charger; and an oil pump.

18. The electric vehicle as claimed in any one of claims 1 to 17 further comprising at least one lookup table defining which power source to use for each operation state.

19. The electric vehicle as claimed in any one of claims 1 to 17 further comprising an artificial intelligence, AI, module for determining which power source to use based on at least the determined operation state.

20. The electric vehicle as claimed in claim 19 wherein the AI module is configured to:predict an upcoming operation state of the electric vehicle; andoutput a suggested command for controlling the high voltage power distribution module to direct power from at least one of the plurality of power sources based on the determined operation state and the predicted upcoming operation state, to ensure the power sources are ready for the upcoming operation state.

21. The electric vehicle as claimed in claim 20 wherein the AI module predicts an upcoming operation state using any one or more of: route information; live traffic information; weather information; and remaining journey distance.

22. A method for controlling the operation of an electric vehicle, the electric vehicle comprising a plurality of power sources, the method comprising:determining an operation state of the electric vehicle; andcontrolling a high voltage power distribution module of the electric vehicle to direct power from at least one of the plurality of power sources based on at least the determined operation state.

23. The method as claimed in claim 22 wherein the determined operation state of the electric vehicle has an associated power demand, and controlling the high voltage power distribution module comprises:controlling the high voltage power distribution module to direct power from at least one of the plurality of power sources that is capable of meeting the power demand.

24. The method as claimed in any one of claims 22 or 23 further comprising:obtaining route information defining a route to be taken by the electric vehicle, wherein the route information comprises road type information and topography information,wherein controlling the high voltage power distribution module to direct power from at least one of the plurality of power sources is based on the determined operation state and the route information.

25. The method as claimed in any one of claims 19 to 23 further comprising:determining at least one dynamic control parameter for the electric vehicle, andusing the determined at least one dynamic control parameter to control the high voltage power distribution module.

Images