A method of monitoring a hydrogen-powered generator system

Abstract

A method of monitoring the performance of a hydrogen-powered generator system comprising a generator and a hydrogen supply system configured to provide hydrogen to the generator, the generator operabl

Description

This invention relates to a method of monitoring the performance of a hydrogen-powered generator system and in particular, but not exclusively, a method of monitoring the efficiency of a hydrogen-powered generator system and / or identify faults associated with a hydrogen-powered generator system. Hydrogen-powered generators use hydrogen to generate power, typically in the form of an electrical current. Hydrogen is stored, typically as a compressed gas or liquid in a gas cylinder, canister or other hydrogen supply system, and provided to the generator. A first type of generator uses hydrogen gas in an internal combustion engine to turn a crankshaft and thereby generate an electrical current using an attached alternator. A second type of generator uses a fuel cell which utilises chemical reactions between hydrogen and oxygen to create water and an electric current. A hydrogen-powered generator in combination with a hydrogen supply system (e.g. a canister) can be referred to as a hydrogen-powered generator system. When the hydrogen in the canister is depleted, the system must be refuelled to enable power generation to continue. Refuelling may be provided by re-filling the canister or replacing the empty canister with a full canister. Refuelling processes can lead to downtime in usage of the generator and can be inefficient and inconvenient. It is generally desirable to understand the behaviour and performance of a generator. For example, it may be beneficial to understand the conversion efficiency at which the generator converts hydrogen into power. Additionally or alternatively, it may be beneficial to reduce or avoid downtime due to faults. SUM MARY OF INVENTION According to a first aspect of the invention there is provided a method of monitoring the performance of a hydrogen-powered generator system comprising a generator and a hydrogen supply system configured to provide hydrogen to the generator, the generator operable to generate electrical power using the provided hydrogen. The method comprises: obtaining pressure data, wherein the pressure data is indicative of a pressure of the hydrogen in the hydrogen supply system, obtaining temperature data, wherein the temperature data is indicative of a temperature of the hydrogen in the hydrogen supply system; obtaining power data, wherein the power data is indicative of the electrical power generated by the generator; and generating performance data based on the pressure data, temperature data and power data, wherein the performance data is indicative of the performance of the hydrogen-powered generator system. The method may be computer implemented. For example, the method may be performed by a monitoring system. The monitoring system may be implemented as a server, for example run on one or more computers. The data may be obtained wirelessly from one or more sensors at the hydrogen-powered generator system. As such, the method may be a method of remotely monitoring the performance of a hydrogen-powered generator system. The generator may generate electrical power in the form of an electric current. The electrical power may be referred to as power or output power. The power data may comprise electrical data. The power data may be referred to as output power data or electrical power data. The generator may be any type of hydrogen-powered electricity generator, for example a fuel cell or an internal combustion engine. Monitoring the performance of a hydrogen-powered generator system is useful for load balancing and general upkeep of the system, for example scheduling refuelling and maintenance. Using pressure data, temperature data and power data enables accurate performance data to be generated, allowing accurate monitoring. It is particularly beneficial to perform this monitoring remotely, for example because it avoids the need for staff to be in hazardous areas, it reduces the risk of user error in taking readings, and because any identified issues or events can be identified automatically and responded to quickly and proactively. Additionally, data obtained from the monitoring can be used in workflow design, reducing the risk of potential human errors. Remote monitoring also allows large real-time datasets to be captured, which can help with algorithm development, such as training a machine learning model to predict issues. Furthermore, the method allows for more scalable deployment of generators, as a remote team can manage multiple generators. The method may further comprise receiving volume data, the volume data indicative of a hydrogen storage volume of the hydrogen supply system. The method may further comprise determining quantity data based on the pressure data and temperature data, wherein the quantity data is indicative of a quantity of hydrogen provided to the generator. The quantity of hydrogen provided to the generator may be estimated to be approximately equal to a quantity of hydrogen depleted from the hydrogen supply system. The quantity of hydrogen depleted from the hydrogen supply system may be determined based on the pressure data and temperature data. The pressure data and temperature data can be used to determine, for example using known gas law equations such as the Van der Waals equation, a quantity of hydrogen contained in the hydrogen supply system or a portion thereof (e.g. a manifold and / or one or more hydrogen storage units thereof). By comparing the determined quantity of hydrogen contained in the hydrogen supply system to a previous quantity of hydrogen contained in the hydrogen supply system, a quantity of hydrogen which has been depleted from the hydrogen supply system can be determined. The previous quantity of hydrogen in the hydrogen supply system may be derived from pressure data and temperature data obtained at a previous time, or from known values such as a maximum possible quantity of hydrogen which the hydrogen supply system can contain. The quantity data may be mass data, wherein the mass data is indicative of an amount of mass of hydrogen provided to the generator. The quantity data may be represented differently, for example in terms of the number of molecules or moles of hydrogen provided to the generator. Generating performance data based on the pressure data, temperature data and power data may comprise generating performance data based on the quantity data and power data. In the absence of any egress of hydrogen to sources other than the generator (e.g. due to a leak), it is assumed that quantity of hydrogen depleted from the hydrogen supply system corresponds to the quantity of hydrogen provided to the generator. That is, it is assumed that all hydrogen has been provided to the generator. In other situations, the quantity of hydrogen provided to the generator may be estimated to be proportional to a quantity of hydrogen depleted from the hydrogen supply system. A proportionality constant can be used, where the proportionality constant represents a transfer efficiency (e.g. a percentage of the hydrogen in the hydrogen supply system which is successfully transferred to the generator). A proportionality constant less than 1 may illustrate a leak or other inefficiency. The quantity data can be represented by a mass, weight, number of molecules, number of moles, a fill ratio of the portion of the hydrogen supply system, or in any other unit which can be used to quantify a fluid. Similarly, the quantity of hydrogen depleted from the hydrogen supply system can be represented by a mass, weight, number of molecules, number of moles, a fill ratio of the portion of the hydrogen supply system, or in any other unit which can be used to quantify a fluid. Determining the quantity data may comprise determining a first quantity of hydrogen at a first time, and a second quantity of hydrogen at a second time. For example, the pressure data and temperature data may be determined at a first time, and the associated quantity data may be determined (first quantity data). Pressure data and temperature data may be determined at a second time, and the associated quantity data may be determined (second quantity data). The difference between the first quantity data and second quantity data may be determined, indicating an amount of hydrogen supplied to the generator between the first time and the second time. The volume data may be constant between the first time and the second time. The performance data may comprise a calculated efficiency of the generator. The calculated efficiency may be calculated based on the quantity data and the power data. The calculation may be based on, or compared to, an expected conversion efficiency of the generator. There are theoretical models regarding how much energy can be extracted from hydrogen, for example depending on the generator type and a power level that the generator is operating at, and the expected conversion efficiency can be derived from such a theoretical model. The efficiency may be compared to a theoretical maximum possible efficiency or a previous efficiency previously recorded from the generator in question. The calculated efficiency may be referred to as the actual efficiency, for example because it is an efficiency which the generator is actually operating with. The performance data may further comprise an indication of a fault with the hydrogen-powered generator system. The indication of a fault is determined based on the calculated efficiency and a target efficiency of the generator. The target efficiency may be a theoretical maximum efficiency of the generator or a desired efficiency of the generator. The desired efficiency may be, for example, an expected efficiency, i.e. an efficiency which the generator is expected to operate with. The desired or expected efficiency may be determined based on a previously recorded efficiency for the generator, a previously recorded efficiency for a similar generator, and / or based on a theoretical model. The indication of a fault may be determined based on a comparison between the calculated efficiency and the target efficiency. If the calculated efficiency is less than the target efficiency, it may be determined that there is a fault. If the calculated efficiency is substantially equal to the target efficiency, it may be determined that there is no fault. The method may further comprise identifying the fault based on one or more of the power data, pressure data, temperature data, usage data. The method may further comprise providing an instruction to perform an action based on the fault. Certain faults may be identified using additional data such as the power data, pressure data or temperature data. For example, the generator may operate at sub-optimal efficiency at sub-optimal temperatures. In such a case, the fault may be identified as a temperature fault if the temperature data meets a certain condition (e.g. passing a temperature threshold). The temperature fault may be remedied by changing the temperature of the environment the hydrogen-powered generator system is in. In another example, if there is a leak the pressure may quickly decrease. As such, the fault may be identified based on the pressure data. The fault may be identified as a leak if the pressure data meets a certain condition (e.g. passing a pressure threshold). The pressure fault may be remedied by actuating valves or initiating a maintenance process. The fault may be determined based on usage data, such as a usage pattern of the generator, and / or an age of the generator. The action may be, for example, displaying a message, providing an alarm (e.g. a visual or audio alarm), scheduling a maintenance visit, halting operation of the generator, actuating one or more valves or other apparatus of the hydrogen supply system and / or generator. The instruction can be provided to any means for performing the relevant action, for example a display screen, an alarm system, a scheduling system, or a computer configured to control any relevant physical apparatus. Obtaining the power data may comprise obtaining a voltage and a current output from the generator. The current may be obtained using a non-inline sensing device. The power data may comprise more than one voltage, for example a voltage associated with each of multiple output phases (e.g. in a three-phase output). The power data may comprise more than one current, for example a current associated with each of multiple output phases (e.g. in a three-phase output). Optionally the power data may also comprise a phase angle between the current and voltage output from the generator. The performance data may comprise an indication of a power factor of the power output by the generator. The power factor may be determined based on the phase angle (e.g. phase difference) between the current and voltage output from the generator. The power factor may be used to indicate a quality of the power output. A non-inline sensing device may comprise a current transformer, Rogowski coil, or magnetic field sensor. The non-inline sensing device may be connected to an output circuit of the generator. The performance data may comprise an estimated refuelling time. Generating the estimated refuelling time may comprise determining a rate of use of hydrogen based on the pressure data and temperature data and estimating a time at which a quantity of hydrogen contained in the hydrogen supply system will pass a threshold quantity. The threshold quantity may be full depletion, for example when the quantity contained is expected to be zero. Alternatively, a buffer may be provided to ensure timely refuelling. For example, the threshold may represent a quantity more than zero (e.g. 10% of a maximum quantity). The threshold may be variable and depend on the rate of use of hydrogen. For example, the threshold may be set to correspond to a quantity of hydrogen which would enable hydrogen to be continue to used at the current rate for a specified period of time, for example a year or a day. Determining the rate of use of hydrogen may comprise: obtaining the pressure data and temperature data at a first time; obtaining the pressure data and temperature data at a second time; and calculating the rate of use of hydrogen based upon the pressure and temperature data at the first time and the pressure and temperature data at the second time. According to another aspect there is provided a hydrogen-powered generator system configured to provide pressure data, temperature data and power data to a monitoring system configured to perform the method of the first aspect. For example, the hydrogen-powered generator system may comprise a pressure sensor, temperature sensor and means to output power data, and means for transmitting any of the relevant data to the monitoring system. The means to output power data may comprise a voltmeter and current transformer, for example. According to another aspect there is provided a system comprising the hydrogen-powered generator system of claim the previous aspect and at least one additional hydrogen-powered generator system. The at least one additional hydrogen-powered generator system is configured to provide pressure data, temperature data and power data to a monitoring system configured to perform the method of the first aspect. According to another aspect there is provided a monitoring system configured to perform the method of the first aspect. The monitoring system may comprise one or more processors, and a memory comprising instructions thereon that when executed by the one or more processors, cause the one or more processors to perform the method of the first aspect. The monitoring system may be located remotely from the hydrogen-powered generator system. For example, the monitoring system may be located in the cloud (e.g. implemented on one or more computers providing remote server capabilities). The monitoring system may be further configured to receive pressure data, temperature data and power data from a hydrogen-powered generator system and at least one or more additional hydrogen-powered generator systems. The monitoring system may be further configured to perform the method of the first aspect to generate performance data for each of the at least one additional hydrogen-powered generator systems. In this way, multiple hydrogen-powered generator systems can be monitored. Preferably, they may be monitored remotely. The monitoring system may be further configured to estimate a future event occurring at the hydrogen-powered generator system. The estimation may be based on the performance data generated for one or more of the at least one additional hydrogen-powered generator systems. By monitoring multiple generator systems, historical performance data can be collected and analysed to identify patterns. Based on the patterns, future events can be estimated (e.g. predicted and / or anticipated). The future events may include, for example, a refuelling requirement (e.g. an estimated refuelling time), a maintenance requirement (e.g. an anticipated fault such as a leak or a failure), an estimated efficiency change (e.g. a change in efficiency based on temperature, lifetime, or other data). The performance data may be used to train a machine learning model. The machine learning model may be trained to identify future events based on provided performance data. The estimation may include the identification of one or more future event and / or determining the likelihood of any or all of the one or more future events occurring. For example, an estimation may indicate that a leak will occur. Another estimation may indicate that there is a 10% chance that a leak will occur. Another estimation may indicate that there is a 50% chance that any adverse event will occur. The estimation may include a time at which the one or more future event is estimated to occur. According to another aspect there is provided a computer program product comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of the first aspect. DRAWINGS The invention will now be described, by way of non-limiting example only, with reference to the following Figures in which: Figure 1 depicts a hydrogen-powered generator system; Figure 2 depicts a method of determining a quantity of hydrogen; Figure 3 depicts a method of monitoring the performance of a hydrogen-powered generator system; and Figure 4 depicts a schematic of a computer for carrying out the methods of Figures 2 and 3. DETAILED DESCRIPTION Figure 1 depicts a hydrogen-powered generator system 100 and a monitoring system 110. In general, the hydrogen-powered generator system 100 comprises a hydrogen supply system 10 and a generator 11. The hydrogen supply system 10 comprises a manifold 1a configured to control the flow of hydrogen from one or more hydrogen sources to the generator 11. The hydrogen supply system 10 is depicted with hydrogen sources in the form of gas cylinders 1. Three cylinders 1 are depicted in Figure 1, each connected to the manifold 1a by way of a valve 2 at the manifold 1a and a connector 6 running between each cylinder 1 and its associated valve 2. Any means of transporting hydrogen from the cylinders 1 to the manifold 1a may be used, for example the connectors 6 may comprise a hose, channel, pipe or other conduit. The valves 2 may be isolation valves. The valves 2 may be individually actuatable or actuatable in one or more groups. By actuating the valves 2, the flow of hydrogen between each cylinder 1 and the manifold 1a may be controlled. While three cylinders 1 are depicted in Figure 1, any number of hydrogen sources may be used, for example six or sixteen. While gas cylinders 1 have been described, it will be appreciated that any hydrogen storage unit or other source of hydrogen may be used. For example, the hydrogen storage unit may comprise a tube trailers, a reservoir, or a vessel capable of storing and providing hydrogen, for example a canister. Furthermore, while the term cylinder is used, the hydrogen storage unit may take any shape. The manifold 1a is connected to the generator 11 such that hydrogen received by the manifold 1a can be delivered to the generator 11. The connection between the manifold 1a and the generator 11 includes a valve 2a (e.g. an isolation valve), a pressure regulator 4 and a safety relief valve 5 positioned between an outlet of the manifold 1a and an inlet of the generator 11. The valves 2a, 5 and regulators 4 can be used to control the flow of hydrogen from the manifold 1a to the generator 11 and ensure gas safety. Safety features such as the pressure regulator 4 and a safety relief valve 5 may not be necessary for performing the methods described in more detail below, but may optionally be included for general safety of the system and workers. Any or all of the portions of the hydrogen supply system 10 depicted in Figure 1 may be removably connectable. For example, any of the cylinders 1 may be connected to or removed from connection with the manifold 1a. Similarly, the manifold 1a may be connected to or removed from connection with the generator 11. In use, the hydrogen- powered generator system 100 uses hydrogen to the generator 11 for the generation of power. In use, at least one cylinder 1 or other hydrogen source is connected, directly or indirectly (e.g. via the manifold 1a and various valves 2, regulator 4) to the generator 11 such that hydrogen is provided to the generator 11. By controlling the various valves 2, 2a, the manifold 1a may be kept pressurised while one or more cylinders 1 are removed and replaced. As such, the manifold 1a can be used as an intermediate hydrogen storage receptacle so that the generator 11 can be provided with hydrogen even when one or more cylinders 1 are being replaced or refilled. In other arrangements, larger hydrogen sources such as mobile storage units, such as tube trailers, can be implemented in place of cylinders 1. When depleted, a mobile storage unit can be disconnected and moved to a refuelling site for refilling. In this implementation, the manifold 1a may continue to provide hydrogen either using other hydrogen sources or using hydrogen stored within the manifold 1a itself while the mobile storage unit is disconnected. The manifold 1a has a pressure sensor 3 arranged to measure a gas pressure of gas within the manifold 1a. The pressure sensor 3 is associated with the manifold 1a (e.g. the pressure sensor 3 is located in or on or connected to the manifold 1a such that a gas pressure in the manifold 1a can be measured). The measured pressure will typically be hydrogen gas pressure, although it should be understood that other gases may be present, even if only in trace amounts. The pressure sensor 3 may comprise any type of pressure sensor. The pressure sensor 3 may comprise a single pressure sensor or multiple pressure sensors, for example two pressure sensors. If using multiple pressure sensors, said pressure sensors may be positioned to sample the pressure at different locations within the manifold 1a and / or hydrogen supply system 10. From the multiple sampled pressure values, a pressure gradient, average pressure, or any other quantification of pressure values may be determined. Alternatively, or additionally, multiple pressure sensors may be used when the manifold will be connected to different hydrogen sources, the different hydrogen sources having different maximum pressures at source. A tube trailer will typically have a higher maximum pressure than a cylinder. The pressure of hydrogen throughout the hydrogen supply system 10 may be regulated throughout to enable safe and efficient transport of hydrogen from the various hydrogen sources, through the manifold 1a, and to the generator 11. Providing multiple pressure sensors, helps ensure an accurate reading of pressure in the manifold 1a, for example at the outlet of the manifold 1a leading to the inlet of the generator 11. Each of the multiple pressure sensors may be located in a position in the hydrogen supply system 10 which has a different pressure (e.g. an outlet of a cylinder 1 leading to an inlet of the manifold 1a may have a different pressure to an outlet of a tube trailer leading to another inlet of the manifold 1a, and the outlet of the manifold leading to the inlet of the generator 11 may be at another different pressure). Providing individual pressure sensors at each inlet to the manifold 1a also allows the pressure of the hydrogen within individual cylinders or tube trailers that are coupled to each inlet to be determined. This may be used to determine how much hydrogen is initially stored within each cylinder or tube trailer. Each pressure sensor may operate differently over different pressure ranges (for example, by being calibrated differently, being configured to operate over different pressure ranges and / or having different accuracies in different pressure ranges). The pressure sensor 3 may comprise, or be in communication with, a transceiver (not shown) such that pressure data can be provided to the monitoring system 110, which may be a remote monitoring system. The monitoring system 110 may be, for example, a server or other external system providing remote monitoring capabilities. Using the transceiver, data from the pressure sensor 3 comprising measured pressure values from the manifold 1a can be transmitted to the remote monitoring system 110. The transceiver may be wireless, communicating with the monitoring system 110 via for example (but not limited to) 3G, 4G (e.g. 4G LTE), WiFi, Bluetooth Low Energy (BLE), Narrow Band-Internet of Things (NB-loT), Long Range (LoRa) radio communication, ZigBee or satellite. The generator provides electrical power 12. The electrical power 12 can be provided, in the form of electricity (e.g. electrical power), to any system for use or storage. The electrical power 12 is typically provided as a three phase alternating current (AC). From the electrical power 12, power attributes can be generated. The power attributes may be referred to as power data. The power data may include one or more of voltage, current and phase angle of one or more of the three phases of output power 12. The voltage may be measured by sampling from a suitable electrical output terminal, or a busbar (not shown). The voltage may be sampled using any voltage measurement apparatus. A busbar is generally used for electrical power distribution. The voltage of any of the phases can be sampled from the busbar. The current of any of the phases may be sampled using one or more current transformers. The current transformers can be installed around power cables that carry the electrical power 12 from the generator 11. Such power cables are typically connected to the generator 11 via the busbar. The current transformers provide an indirect measure of the current of the electrical output 12. Because of its indirect sampling method, a current transformer can be used to measure the output current for high electrical powers, for example greater than 40kW e.g. 100 kW or even greater. Other methods of sampling current, for example in-line current measurement, may be limited to lower power outputs, e.g. 40kW or lower. Sampling current using an indirect sampling method, such as a current transformer, is particularly useful for high power use cases e.g. the generation of electrical power for a power. Furthermore, indirect sampling methods, such as the use of a current transformer, can be used around insulated cables. The power data may be communicated to the monitoring system 110, for example by a transceiver, similar to the pressure data from the pressure sensor 3. The generator 11 may use the same transceiver as the pressure sensor 3 to communicate power data, or it may use a different transceiver. The power data can be used to determine the power generated by the generator 11. For example, multiplying the measured voltage (in volts) with the measured current (in amps) gives the power (in Watts) generated by the generator 11. The hydrogen-powered generator system 100 may further comprise a temperature sensor 13. The temperature sensor 13 is arranged to provide an estimated temperature of the hydrogen at the hydrogen supply system 10. As such, the temperature sensor 13 may be positioned on, in or close to the hydrogen supply system 10. A hydrogen-powered generator system will typically give rise to one or more ATEX zones, where ATEX is an acronym for "Appareils destines a etre utilises en Atmosphere Explosive” (“Equipment intended for use in explosive atmospheres”). It can be difficult to locate a temperature sensor 13 within the manifold 1a which may, for example, comprise an ATEX zone (e.g. an ATEX Zone 0). However, it has been found by the inventors that arranging the temperature sensor 13 to measure the environmental temperature of the hydrogen supply system 10 provides a good approximation of the hydrogen temperature in the hydrogen supply system 10 (e.g. in the manifold 1a). For example, placing a temperature sensor 13 at the edge of an ATEX zone provides a good approximation of the temperature of the hydrogen. The temperature may be sampled outside of the ATEX zone 0, or even outside of any ATEX zone (outside a perimeter or environment which forms an ATEX zone 2, for example). The temperature of the hydrogen in the hydrogen supply system may be estimated to be substantially the same as the temperature at the temperature sensor 13. It has been found by the inventors that this estimation provides an good enough approximation of the temperature of the hydrogen to accurately perform the methods described herein. The temperature of the hydrogen may be determined, based on the temperature at the temperature sensor 13, using another relationship. For example, a theoretical or empirical model may be used to determine the temperature of the hydrogen in the hydrogen supply system 10 based on the temperature at the temperature sensor 13. In another example, the environmental temperature of the hydrogen supply system and therefore the temperature of the hydrogen in the hydrogen supply system could be estimated based on local weather information. Temperature data can be communicated to the monitoring system 110, for example by a transceiver, as described above for the power data and / or pressure data. Figure 2 depicts a method 200 of determining a mass of hydrogen using a remote monitoring system. For illustrative purposes, the method 200 is described below with reference to determining a mass of hydrogen in the hydrogen supply system 10 of Figure 1 using the remote monitoring system 110 of Figure 1. However, the method 200 can be applied mass determination in other hydrogen-powered generation systems. In a first step 201, the monitoring system 110 receives data indicative of a pressure. The data indicative of a pressure may be referred to as pressure data. The pressure data may be representative of a gas pressure in the manifold 1a of the hydrogen supply system 10. The pressure data may have been measured using the pressure sensor 3. The pressure data may be received by the monitoring system 110 by any means, for example the transmission means described above with reference to the pressure sensor 3 and associated transceiver. Preferably, the pressure data is received by the monitoring system 110 wirelessly. The pressure data is denoted “Pressure” in Figure 1. As described above, more than one pressure sensor 3 may be provided at the hydrogen supply system 10. The pressure data may therefore comprise multiple pressure readings recorded by the pressure sensors and / or one or more values determined based on the multiple pressure readings. The determined values may include, for example, a pressure gradient, an average pressure across the manifold 1a, a calibrated or corrected pressure. In situations where all hydrogen storage units and the manifold 1a are at the same pressure and backflow is enabled between different vessels, a single pressure sensor and single pressure value may be sufficient. However, where different units are at different pressures, more accurate pressure data may be provided through the use of multiple pressure sensors and multiple pressure readings. In an example, each of multiple cylinders 1 connected to the manifold 1a are at different pressures. A pressure reading may be taken at the outlet of each cylinder 1 and the outlet of the manifold 1a so as to determine an appropriate measure of pressure for use in calculating mass data as described in the third step 203 below. In a second step 202, the monitoring system 110 receives data indicative of a temperature. The data indicative of a temperature may be referred to as temperature data. The temperature data is indicative of a temperature at the hydrogen supply system 10, e.g. indicative of the temperature of the hydrogen at the hydrogen supply system. The temperature data may have been measured using the temperature sensor 13. As described above, the temperature data may be based on a measurement of the environmental temperature of the hydrogen supply system 10. The temperature data may be received by monitoring system 110 by any means, for example the transmission means described above with reference to the temperature sensor 13 and associated transceiver. Preferably, the temperature data is received by the monitoring system 110 wirelessly. The temperature data is denoted “Temperature” in Figure 1. In a third step 203, the monitoring system 110 determines data indicative of a quantity of the hydrogen based on the received pressure data and temperature data. The data indicative of the quantity of hydrogen may be referred to as quantity data. The quantity may be represented as data indicative of a mass of the hydrogen. The data indicative of a mass of hydrogen may be referred to as mass data. The mass data may indicate a mass of hydrogen present in (e.g. contained within) the manifold 1a. The mass data may indicate a mass of hydrogen present in (e.g. contained within) the manifold 1a and any connected cylinders 1. The mass data may be represented in terms of a mass, weight, number of molecules, or any other way of representing an amount of hydrogen. As such, the third step 203 may be broadly referred to as obtaining quantity data, wherein quantity data may comprise mass data. The further description of the third step 203 below will be discussed in the context of mass data, but it should be understood that it may be implemented with any quantity data rather than being solely limited to mass data. The calculation of the mass data may use, as an input, volume data, wherein volume data is indicative of the hydrogen storage volume of the hydrogen supply system 10 being monitored (for example a volume of the manifold 1a and any connected cylinders 1). That is, the third step 203 may further comprise determining data indicative of a mass of the hydrogen based on the received pressure data and temperature data and further based on volume data. It should be understood that, for a given system, this volume will be a fixed value. However, when different hydrogen storage units are interchanged (e.g. a different size of cylinder is introduced) the volume in question may change. As such, the monitoring system 110 may receive data indicative of the volume of the hydrogen storage volume being monitored. The volume data may be received as a single or infrequent input, for example during calibration of a hydrogen supply system 10 or when a new hydrogen storage unit (e.g. a cylinder 1) is connected to the manifold 1a. Alternatively, the hydrogen supply system 10 may periodically provide (e.g. by transmission using a transceiver or transmitter) current volume data. The current volume data may be entered manually or may be obtained using device data, for example a hydrogen storage unit may include device data comprising data indicative of its volume, and means to provide the data indicative of its volume to the remote monitoring system 110. The device data may be stored in the form of an RFID tag, a barcode, a QR code, or any other type of code on the hydrogen storage unit. The device data may, for example, be in the form of an alphanumeric code or pictorial representation on the hydrogen storage unit. The device data may be extracted from the relevant code, representation, tag by any known means, for example optical character recognition, a barcode reader, a QR code reader, an RFID reader. Alternatively, the device data may be provided by an external system, for example a device data repository. The device data repository may contain information regarding hydrogen storage units which have been delivered to a particular location and / or which are due to be connected. The examples of device data provision detailed above are particularly suitable for use in combination with hydrogen-powered generator systems because they are non-invasive and so are more safe to use in an ATEX zone than other methods. For example, a method used in other scenarios may be including an electrical connection for the purposes of transferring device data when a hydrogen storage unit is connected to the manifold 1a, but an electrical connection can give rise to sparks which may be dangerous in ATEX zones. Alternatively, or additionally, multiple pressure sensors may be used when the manifold will be connected to different hydrogen sources, the different hydrogen sources having different maximum pressures at source. For example, a tube trailer will typically have a higher maximum pressure than a cylinder., which may operate differently over different pressure ranges (for example, by being calibrated differently, being configured to operate over different pressure ranges and / or having different accuracies in different pressure ranges), helps ensure an accurate reading of pressure in the manifold 1a. Furthermore, as described below in more detail, multiple pressure sensors may aid in accurately determining the hydrogen storage volume within the hydrogen supply system 10. Each cylinder 1 (or other hydrogen storage unit) may be filled to a slightly different pressure. As such, knowledge of the volume of the unit and the pressure and temperature of the gas therein can provide a particularly accurate determination of the mass of gas contained therein. The calculation of the mass data may be determined using any suitable algorithm. The ideal gas law may be used, or any other gas law that will be known to those skilled in the art. For example, the Van Der Waals equation has been found by the inventors to be sufficiently accurate, even at high pressures. The mass data may additionally or alternatively indicate a mass of hydrogen used in contrast to a mass of hydrogen remaining in the hydrogen supply system 10. The amount of hydrogen used will generally correspond to the amount of hydrogen provided to the generator 11. The mass of hydrogen used may be determined based on a mass of hydrogen remaining in the hydrogen supply system 10 compared to a known maximum amount of hydrogen contained in the hydrogen supply system 10 and / or a known amount of hydrogen in the hydrogen supply system 10 at a previous time. The mass data may be represented in terms of a fill ratio (e.g. a ratio of the mass of hydrogen compared to a maximum possible mass of hydrogen which may be contained in the hydrogen supply system). The method 200 may be repeated to obtain time series data. At a first time the first, second and third steps 201-203 may be performed to obtain first pressure data, first temperature data and first mass data, each indicating the pressure, temperature and mass at the first time. The data obtained at the first time may be referred to as first time data. At a second time the first, second and third steps 201-203 may be repeated to obtain second pressure data, second temperature data and second mass data, each indicating the pressure, temperature and mass at the second time. The data obtained at the second time may be referred to as second time data. The first and second time data may also include the volume data, which remains constant between the first time and the second time. Based on the first and second time data, a rate of change of any of the monitored and / or determined properties can be determined. Based on the first and second time data and using extrapolation, for example based on said rate of change, a prediction may be made. The prediction may represent a future event or future state of the hydrogen supply system 10. The prediction may be, for example, a depletion time representing a time at which it is predicted the hydrogen supply system 10 will be depleted of hydrogen (i.e. contain zero or a negligible mass of hydrogen). The prediction may be, for example, a refill time representing a time at which it is preferable that the hydrogen supply system 10 should be refilled. The refill time may be, for example, a time at which the mass of hydrogen in the hydrogen supply system 10 is predicted to fall below a threshold mass (e.g. 10% of the known maximum mass). The refill time may be, for example, a time which is predicted to occur more than a threshold amount of time from the depletion time (e.g. one hour before the depletion time). An action can be performed in response to the mass data. The mass data can be compared to one or more conditions and an action can be performed if the mass data matches one of the one or more conditions. In a first example, the mass data comprises a mass value representing the mass of hydrogen present in the hydrogen supply system 10. A first condition is selected, the first condition being a predetermined mass value of hydrogen present in the hydrogen supply system 10, for example a mass value of 10% of the maximum mass of hydrogen which could be contained by the hydrogen supply system 10. If the mass value of the mass data is equal to or below the predetermined mass value, it is determined that the first condition has been met. In response to the first condition being met, an action can be performed. For example, the action may be an alert that the hydrogen supply system 10 requires refuelling. The alert can be sent by the monitoring system 110, for example, to a user of the system 100. Alternatively or additionally, the alert can be sent to a third party, such as a fuel provider or distributer. On receipt of the alert, the fuel provider or distributer may, using an appropriate Application Programming Interface (API) with a scheduling system, automatically schedule a refuelling appointment with the user. The alert can be sent wirelessly to the user. For example, the alert can be sent to a personal device of the user, such as a mobile computing device (smartphone, smartwatch). The alert may include a message, a sound, a visual cue, or any other means of alerting a user. In a second example, the mass data comprises a predicted depletion time. The predicted depletion time can be communicated, by the remote monitoring system 110, to a third party (for example, a user or fuel provider) and / or a scheduling system. The predicted depletion time can be communicated as a matter of course. Additionally or alternatively, the predicted depletion time can be communicated if the depletion time will occur in less than a threshold amount of time, for example less than one day. In this instance, a second condition is selected, the second condition being, for example, a time less than one day. The action (e.g. communicating the predicted completion time) is performed if the second condition is met, i.e. if the predicted completion time is less than the threshold amount of time selected as the second condition. The action may include the scheduling of a refuelling event, or the issuing of an alert to indicate that refuelling should occur. In a third example, the mass data comprises mass values representing the mass of hydrogen present in the hydrogen supply system 10 at a first time and a second time. A rate of change of mass can be calculated based on the mass values. A third condition is selected, the third condition being a predetermined rate of change of mass. If the calculated rate of change of mass is greater than the predetermined rate of change of mass, the third condition is met. In some circumstances, such a high rate of change of mass may indicate a leak or other adverse event associated with the hydrogen supply system 10. As such, if the third condition is met, an action may be performed including providing an alert to indicate an adverse event such as a leak. The alert may include a warning that personnel should vacate the location. The action may include actuating safety equipment or protocols, for example opening or closing valves. The above described first, second and third examples are illustrative in nature and are not intended to be limiting. It can be understood that any number of actions may be performed in response to any number of conditions, either of which may be set depending on the needs of the user and the available data yielded by the method 200. The method 200 described above may be used to determine the mass of hydrogen in a system other than the hydrogen supply system 10 depicted in Figure 1. For example, the method 200 may be used to monitor the mass of hydrogen in a hydrogen supply system with no manifold, or where a different gas other than hydrogen is used. Figure 3 depicts a method 300 of monitoring the performance of a hydrogen-powered generator system. For illustrative purposes, the method 300 is described below with reference to monitoring the performance of the hydrogen-powered generator system 100 of Figure 1 using the remote monitoring system 110 of Figure 1. However, the method 300 can be applied to performance monitoring in other hydrogen generation systems. In a first step 301, the monitoring system 110 obtains data indicative of a quantity of hydrogen provided to the generator 11. The data indicative of a quantity of the hydrogen provided to the generator 11 can be, or be derived from, the mass value determined using the above described method 200. The quantity data may comprise mass data and so the first step 301 may alternatively comprise obtaining mass data. In a second step 302, the monitoring system 110 obtains data indicative of the electrical power 12 of the generator 11. The data indicative of the electrical power may be referred to as electrical data or power data. Obtaining power data is described above with reference to Figure 1. The power data is denoted “Power” in Figure I.The power data can include a voltage and a current output from the generator 11. As described above, a hydrogen-powered generator may provide three-phase output. The power data can therefore include data indicative of the electrical output of each phase of the three-phase output. For example, the power data may include a first voltage and first current associated with the first phase of the three-phase output, a second voltage and second current associated with the second phase of the three-phase output and a third voltage and third current associated with the third phase of the three-phase output. In a third step 303, the monitoring system determines performance data associated with the hydrogen-powered generator system 100. The performance data is determined based on the obtained data indicative of the quantity of hydrogen and received electrical data. The performance data may be an efficiency of the generator 11. The efficiency can be determined based on how much hydrogen is provided to the generator 11 by the supply system 10 and how much power is output by the generator 11. The quantity of hydrogen provided to the generator 11 is assumed to be equivalent to the quantity of hydrogen which leaves the manifold 1a. By calculating the mass of hydrogen in the manifold 1a at a first and second time, a mass difference is obtained. The quantity of hydrogen provided to the generator 11 is assumed to be equal to the mass difference. Hydrogen-powered generators typically have known efficiencies, and so, given a certain quantity of hydrogen, there is an expected amount of power that a generator should be capable of outputting. However, the actual power output may not match the expected power output. By comparing the actual power output by the generator for the quantity of hydrogen believed to be provided to the generator and the expected power output of a generator for the same quantity of hydrogen, a calculated efficiency can be determined. The calculated efficiency may be less than the theoretical maximum efficiency. The power output can be represented using a ratio of actual power output compared to expected power output. The generator efficiency can be represented using a ratio of the actual efficiency compared to an expected efficiency (e.g. a desired efficiency or theoretical maximum efficiency). The actual power output can be determined based on the power data output from the generator 11 as described above with reference to Figure 1. If the calculated efficiency is less than the theoretical maximum, this can indicate a fault in the system 100. The fault may be due to the generator 11 operating sub-optimally, which may for example be because it requires some maintenance or is operating in sub-optimal conditions, e.g. a sub-optimal temperature. Alternatively, the fault may be due to the actual quantity of hydrogen received by the generator 11 being less than the assumed quantity. The actual quantity of hydrogen received by the generator 11 may be less than the assumed quantity if, for example, there is a leak. Secondary indicators can be used to determine the type of fault that is present. In the first instance where the generator 11 is operating sub-optimally, data such as the temperature may indicate a temperature fault. In the second instance where there is a leak, a sudden drop in pressure may be expected to accompany a leak. Additionally or alternatively, where multiple pressure sensors are provided, a difference between a pressure at a first location and pressure at a second location of a particular volume (e.g. the manifold 1a) may also indicate a leak. As such, by considering the pressure data, temperature data, power data etc, the source of the fault may be identified. In some instances, various data can be used as a secondary indicator to determine that the fault does not correspond to a particular type of fault. That is, data can be used to exclude a particular fault. For example, if the fault is initially suspected to be a leak, pressure data and / or data from an additional sensor may be considered and, if the pressure data and / or data from the additional sensor are not consistent with a leak, the fault may be determined to be not a leak. Once the fault has been determined to not be a leak, the fault may be suspected to be the next most likely fault, for example. A secondary indication used to determine the type of fault that is present may be the magnitude of the calculated efficiency. If the calculated efficiency, for example, falls to zero, this may indicate a failure of the generator 11 (e.g. indicating that no power is being output). In some instances, an efficiency history may be maintained (e.g. data indicating the previous efficiency of the generator 11 at one or more times prior to a current time). By comparing the calculated efficiency to the efficiency history (e.g. comparing the calculated efficiency to one or more efficiency values in the history or a value derived therefrom, for example an average efficiency), a fault can be determined to be of a particular type. A sudden change in efficiency (e.g. a large and / or sudden difference between the calculated efficiency and the efficiency history) may indicate a leak, whereas a gradual change in efficiency may indicate a lowering performance of the generator. The above described implementations are non-exhaustive and other secondary indicators may be used to determine the type of fault that is present. The performance data may, in addition or alternatively, comprise an indication of a power factor of the electrical power. The power factor may depend on properties (e.g. current and voltage) of the electrical power output from the generator. For example, the power factor may depend on a phase angle between the output current and output voltage. An optimal power factor may be achieved when the output current and output voltage are in sync, for example when the phase angle is zero. The power factor can be a useful way of understanding the efficiency of power usage of the hydrogen-powered generator system 100. The power factor provides a quantification of the real power output (e.g. relating to actual usable power) compared to the apparent power output (e.g. the RMS of current and voltage, some of which may be lost through resistive heating). The power factor may be represented in terms of a phase angle, a percentage, or a ratio between 0 and 1. A measurement of the power factor of the power output may be, for example, 80%. The performance data may, in addition or alternatively, be an estimated refuelling time, e.g. a time at which refuelling will be required. This may be determined as described above, based on the determined depletion time. Based on the performance data, an action can be performed. The action may be the provision of an alert, for example indicating that a fault is present, the type of fault, that a refuelling event is required, that a maintenance event is required. The performance data may be displayed, for example on a display screen associated with the monitoring system whether directly connected or remotely in communication with the monitoring system. The action may be to schedule an event, for example a refuelling event or a maintenance visit. The action may be a determination of a health of the generator. The health of the generator may be determined based on, for example, the energy conversion efficiency of the generator. The health of the generator may be represented by a health score. Based on the health score, a value (e.g. a monetary value) and / or a remaining lifetime of the generator may be determined. Furthermore, a single monitoring system can be used to monitor multiple hydrogen-powered generator systems (or, in some instances, multiple generators supplied by a single hydrogen supply system / By monitoring multiple systems, historical data, such as power data, pressure data, temperature data, volume data, performance data, can be collected and analysed from many hydrogen-powered generator system to identify patterns. Based on the patterns, future events can be estimated (e.g. predicted and / or anticipated events). The future events may include, for example, a refuelling requirement (e.g. an estimated refuelling time), a maintenance requirement (e.g. an anticipated fault such as a leak or a failure), an estimated efficiency change (e.g. a change in efficiency based on temperature, lifetime, or other data). The data may be used to train a machine learning model. The machine learning model may be trained to identify future events based on provided performance data. That is, once trained, the machine learning model may take as input data from a specific hydrogen power generator system, such as power data, pressure data, temperature data, volume data, etc., and output a likelihood of a specific event occurring, such as a failure. The monitoring system 110 is described and illustrated herein as a single item, exemplified as a server. It should be understood that such a system (e.g. a server) can be implemented as one or more computers in one or more locations. As such, different data may be communicated to different portions of the external system 110 for monitoring purposes. The monitoring system 110 may be remote, in that it may be located in a geographically different location to the hydrogen-powered generator system 100. Figure 4 shows a computer 400 in further detail. The computer 400 may be one of the one or more computers on which the monitoring system 110 is implemented. It can be seen that the computer 400 comprises a CPU 41 which is configured to read and execute instructions stored in a volatile memory 42 which takes the form of a random access memory (RAM). The volatile memory 42 stores instructions for execution by the CPU 41 and data used by those instructions. For example, in use, pressure data, temperature data, quantity data and / or performance data may be stored in the volatile memory 42. The computer 400 further comprises non-volatile storage in the form of a solid state drive 43. Data such as gas constants for use in calculation algorithms and / or volume data may be stored in the non-volatile storage 43. The computer 400 further comprises an input / output (I / O) interface 44 to which are connected peripheral devices used in connection with operation of the computer 400. More particularly, the I / O interface is connected to the pressure sensor 3, temperature sensor 13 and generator 11 (or measurement systems arranged to monitor the output power of the generator 11, for example a current transformer and voltmeter) such that pressure data, temperature data and power data can be obtained. Furthermore, a display 45 is configured so as to display output from the computer 400. The display 45 may, for example, display a representation of the performance data using a graphical user interface. The display 45 may display images generated by the various input and output data. Other input devices are also connected to the I / O interface 44. Such input devices include a keyboard 46 and a mouse 47 which allow user interaction with the computer 400. Alternatively the input devices may comprise, for example, a touch screen of a personal user device (e.g. smartphone, watch). The various peripheral devices may be directly connected or indirectly connected (e.g. through the use of wireless communication) to the I / O interface 44. It should be understood that the peripheral devices may themselves be associated with other computer systems. For example, the keyboard 46, mouse 47 and display 45 may be part of a user device used to interface with the computer 400 (e.g. to control processes and / or view data). Such a user device may be implemented remotely from the I / O interface 44 of the computer 44. Similarly, the pressure sensor 3, temperature sensor 13 and generator 11 may form part of one or more different computer systems each configured to provide the relevant data to the I / O interface 44. A network interface 48 allows the computer 400 to be connected to an appropriate computer network so as to receive and transmit data from and to other computing devices. For example, the computer 400 may be remotely controlled by a remote computer via the Internet. The CPU 41, volatile memory 42, solid state drive 43, I / O interface 44, and network interface 48, are connected together by a bus 49. Alternatively, in other arrangements where portions of the computer 400 are implemented separately, for example as a system of distributed computers, connection between the various components of the computer 400 may be implemented wirelessly. In the described and illustrated implementations described herein, the hydrogen supply system comprises a manifold 1a and hydrogen sources in the form of cylinders 1. However, other arrangements may be implemented which still utilise the methods described herein. For example, the generator 11 may receive hydrogen directly from a hydrogen source and not include a manifold 1a. In these arrangements, the pressure sensor 3 may be positioned at any appropriate position such that a pressure of the gas being supplied to the generator 11 may be measured. In the systems described herein, various safety methods may be employed to ensure the safety of the system and any personnel in view of the high gas pressures, high voltages and high currents typically present in a hydrogen-powered generator system. For example, any gas storage or transportation equipment may include safety valves, flow restrictors, relief valves, regulators or any other gas control mechanisms. An Intrinsically Safe Barrier, such as a Zener barrier (e.g. comprising a Zener diode, fuse and resistor), galvanic isolators, etc. may be used to limit the voltage and current entering an ATEX zone. The pressure sensor 3, and / or temperature sensor 13, could be implemented without a transceiver, for example it could comprise an analogue or digital pressure sensor which is read by a user who inputs the data for analysis, or image recognition system records the data for analysis. However, for autonomous monitoring it is advantageous to implement as a digital pressure sensor and temperature sensor 13 with automatic transmission of data to the remote monitoring system 110. Transceivers are referred to above. However, these can be implemented separately as transmitters and / or receivers. In some scenarios, for example where data is being output by an apparatus (e.g. the pressure sensor 3, the temperature sensor 13), only a transmitter and no receiver may be required and so a transmitter may be implemented without a receiver, or only the transmission capabilities of the transceiver may be utilised. In alternative implementations, rather than immediately transmitting data to the remote monitoring system 110, data can be stored locally and periodically transferred to the remote monitoring system, either wirelessly or using a wired connection. The hydrogen-powered generator systems described herein are referred to as generating power, or electrical power, in the form of an electrical current. They may analogously be referred to as generating energy. For example, the system converts energy (e.g. chemical energy) from the hydrogen fuel into electrical energy. It should be understood that power is a function of energy over time. As such, terms such as energy efficiency may analogously be understood in terms of power efficiencies.

Claims

:

1. A method of monitoring the performance of a hydrogen-powered generator system comprising a generator and a hydrogen supply system configured to provide 5 hydrogen to the generator, the generator operable to generate electrical power using the provided hydrogen, the method comprising:obtaining pressure data, wherein the pressure data is indicative of a pressure of the hydrogen in the hydrogen supply system;obtaining temperature data, wherein the temperature data is indicative of a 10 temperature of the hydrogen in the hydrogen supply system; andobtaining power data, wherein the power data is indicative of the electrical power generated by the generator;generating performance data based on the pressure data, temperature data and power data, wherein the performance data is indicative of the performance of the C\l 15 hydrogen-powered generator system.

002. The method of claim 1, further comprising determining quantity data based on QQ the pressure data and temperature data, wherein the quantity data is indicative of aquantity of hydrogen provided to the generator.

203. The method of claim 2, wherein the performance data comprises a calculated efficiency of the generator, wherein the calculated efficiency is calculated based on the quantity data and the power data.25 4. The method of claim 3, wherein the performance data further comprises anindication of a fault with the hydrogen-powered generator system, and wherein the indication of a fault is determined based on the calculated efficiency and a target efficiency of the generator.30 5. The method of claim 4, further comprising identifying the fault based on one ormore of the power data, pressure data, temperature data, usage data, and optionally providing an instruction to perform an action based on the fault.

6. The method of any preceding claim, wherein obtaining the power data comprises obtaining a voltage and a current output from the generator, and optionally wherein the current is obtained using a non-inline sensing device.5 7. The method of claim 6, wherein the performance data comprises an indicationof a power factor of the power output by the generator.10158. The method of any preceding claim, wherein the performance data comprises an estimated refuelling time, and wherein generating the estimated refuelling time comprises:determining a rate of use of hydrogen based on the pressure data and temperature data; andestimating a time at which a quantity of hydrogen contained in the hydrogen supply system will pass a threshold quantity.

9. The method of claim 8, wherein determining the rate of use of hydrogen comprises:obtaining the pressure data and temperature data at a first time;obtaining the pressure data and temperature data at a second time;calculating the rate of use of hydrogen based upon the pressure and temperature data at the first time and the pressure and temperature data at the second time.

10. A hydrogen-powered generator system configured to provide pressure data, 25 temperature data and power data to a monitoring system configured to perform the method of any of claims 1 to 9.

11. A system comprising the hydrogen-powered generator system of claim 10 and at least one additional hydrogen-powered generator system, the at least one additional 30 hydrogen-powered generator system configured to provide pressure data, temperature data and power data to a monitoring system configured to perform the method of any of claims 1 to 9.

12. A monitoring system configured to perform the method of any of claims 1 to 9.

13. The monitoring system of claim 12, further configured to:receive pressure data, temperature data and power data from a hydrogen-5 powered generator system and at least one or more additional hydrogen-powered generator systems; andperform the method of any of claims 1 to 9 to generate performance data for each of the at least one additional hydrogen-powered generator systems.10 14. The monitoring system of claim 13, wherein the monitoring system is furtherconfigured to estimate a future event occurring at the hydrogen-powered generator system, the estimation based on the performance data generated for one or more of the at least one additional hydrogen-powered generator systems.

15. A computer program product comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of any of claims 1 to 9.

Images