A smart safety system for early detection of thermal runaway electric vehicle battery charging

Abstract

The described invention relates to a system for detecting the early-stage risk or presence of thermal runaway in a charging electric vehicle battery. The system comprises a smart safety device 100 inc

Description

Studies show that 93% of all rechargeable Electric Vehicle (“EV”) batteries are Lithium-Iron Batteries (“LIBs”); while LIBs are the mainstay of EV battery production, they are also sensitive to abuse and, in certain circumstances, they represent a serious fire hazard due to their tendency to undergo a thermochemical chain reaction known as Thermal Runaway (hereinafter, “TR”). EV battery abuse comes in 3 mains forms, usually in an escalating manner: • Physical abuse: the most common form of abuse in certain specific cases; this tends to lead to physical battery deformation, which in turn leads to the cause of Internal Short Circuit (ISC) of the EV battery: • Electrical abuse: the main form of electrical abuse is an ISC, typically induced by incorrect charging methods and equipment, or by means of physical abuse as stated already. Electrical abuse can also occur by battery overcharging or over-discharging; in all cases, electrical abuse then induces heat energy release and chemical heat release; • Thermal abuse: this is often caused by high ambient heat in the vicinity of the battery; it is also caused by electrical energy transferring to heat energy, or heat released by chemical reactions, caused by electrical or physical abuse. Once the process of TR is underway, more heat is released which causes more chemical reactions and more heat and so on; TR can therefore be viewed as an escalating positive feedback chain. Once TR has reached a certain point, it becomes a chain reaction which cannot be stopped; the heating of the EV battery causes gases to build up and an overpressure situation internally within the battery, which often makes the battery swell causing deformation, (i.e., physical abuse). Eventually, the gases reach a point where they trigger the release of a safety valve in the battery system, ejecting a mixture of often flammable gases, moving often at supersonic speeds, into the environment around the battery. This gaseous mixture is then capable of ignition, often explosive ignition, if the EV battery continues to be supplied electrical energy and undergoes a TR chain reaction. Once this happens, an explosion and fire often cannot be prevented, with obvious life-threatening consequences. Although TR can happen in any rechargeable battery, including any type of EV battery, the greatest danger of TR lies in the charging of batteries for electric bikes (“E-bikes”) and electric scooters (“E-Scooters”, which for the purposes of this invention would include “mobility” scooters for those with mobility challenges, in addition to the general purpose E-scooter used for transport); this is because these batteries are typically charged indoors (whereas cars are charged outdoors), and also because batteries of E-bikes and E-scooters are more likely to have suffered physical abuse during use (impacts from kerbs, potholes, pavements, steps etc). The other reason that E-bike and E-scooter charging can be dangerous is that it typically happens overnight, when occupants are at their most vulnerable. Research^1) by the UK electrical safety charity Electrical Safety First provides facts and figures on the increasing trend of E-bike and E-scooter fires. • In 2022, there were 116 E-bike or E-scooter battery fires in London alone; • In 2023, the number of E-bike or E-scooter battery fires in London has increased by 60%, resulting in 4 deaths; • In Q1 2023, there have been 59 E-bike or E-scooter battery fires in New York, resulting in 5 deaths. Efforts to mitigate the risk of TR in EV batteries has focused on the battery management system itself, looking for early stage signals within the battery, but this brings certain disadvantages (or foregoes certain advantages), such as: • It does not address the 1 million+ units already in the field (UK, EU, US); • It does not provide an obvious way to integrate with a network to provide alerts to residents. Reducing threat to life is the main aim of this invention; • No network integration also means no way of early alert of Fire &Rescue; • It does nothing to address the key issue of location of the charging point inside a residence (e.g., per the ESF report, often E-bike and E-scooter batteries are often charged in the hallway of a residence, thereby blocking the key exit route in the event of a TR-induced fire). The existing prior art which may bear upon the invention described below is UK Patent GB2576823 (PARFITT, A.). The prior art is aimed at detecting an overheating plug received in a socket and suffers from several drawbacks in relation to the specific case of sensing an early-stage TR hazard; the main thrust of the prior art is the use of contactless thermal imaging. As shown in research^3), this is not an optimum method on its own for TR hazard detection for a number of reasons including accuracy, reliability and speed, and most importantly because contactless thermal imaging requires direct Line Of Sight (LOS) between the sensors incorporated into the socket of the prior art, and the EV battery being charged. It is perfectly possible to imagine any number of real-life situations where LOS would not exist, such as a socket positioned near a corner of a room and the battery positioned around the corner, or a situation where an object (including the E-bike or E-scooter itself) could block or seriously impinge upon the LOS. Additionally, the prior art mentions the detection only of Carbon Monoxide (CO) and Carbon Dioxide (CO2), and neither of these gases are specific gases indicative of a TR process; indeed, CO and CO2 are released by all forms of combustion (meaning they are excellent at detecting any type office risk, but this generality is a handicap when wanting to detect specifically TR-related fires, in order to minimise false alarms and discourage use of the system). Additionally, CO can be released by a faulty appliance, again giving rise to false alarm risk. There is therefore a requirement for a smart system that can reliably and specifically detect a TR hazard at an early stage, and prevent a TR-induced fire from occurring, or at the very least provide an alert and early warning to nearby residents, and the emergency services, that such a fire has indeed occurred. Although much of the focus of this invention is on LIBs (on account of their prevalence), the invention described herein could apply to any rechargeable EV battery, not just the LIB type. In the following description, a “building” means any permanent or temporary structure designed for habitation and a “vessel” means any portable structure or vehicle, crewed or otherwise, intended for transportation of people or cargo. SUMMARY OF THE INVENTION In accordance with a first aspect of the invention there is provided a fire safety system for early detection and prevention of fire caused by TR within an EV battery, comprising: a smart safety device being a device comprising a computer processor configured with a communications device to communicate with one or more sensors configured to detect one or more early TR hazard indicators, wherein the TR hazard indicators may be either or both of: • a gas signature; • an acoustic signature; wherein in each case the smart safety device is incorporated into the faceplate or housing of an electrical conducting device, and wherein the electrical conducting device is connected to the EV battery and is supplying an electric charge to, or receiving an electric charge from, the EV battery and wherein the electrical conducting device in which the smart safety device is incorporated is further configured so as to turn off the flow of electrical current through it upon receiving, from the smart safety device, an instruction to do so when the smart safety device detects a TR hazard nearby, thus eliminating or mitigating the TR hazard. The signature gas and acoustic sensors may be incorporated within or on the electrical conducting device, or directly within or on the smart safety device which is in turn incorporated within or on the electrical conducting device. The essence of the invention is that, by building TR-signature detectors into an electrical conducting device or into a smart safety device incorporated within an electrical conducting device (either of which by definition will be close to a battery being charged, and which does not require direct line of sight between sensor and battery), the early stages of TR can be detected by means of signature gases or sounds which are only typically associated with TR; by incorporating a means to switch off electrical current to the electrical conducting device supplying the charge to the battery in the event of a possible TR hazard, the battery will no longer be supplied with the electrical energy to be turned into heat, and the TR process can be stopped; in the event that the TR process has already passed the stage where it can be stopped, which is a possibility, the detection of a fire at a very early stage can still enable the residents nearby to be alerted, so that they can make their escape as quickly as possible. For the purposes of the invention, the electrical conducting device can be any of: an electrical socket, an electrical adaptor, a power strip or extension lead, or a charger configured to connect an EV battery to an electrical supply. We now look at the two different detection methods separately, and it is important to note that either method could be incorporated with the other; in preferred embodiments, both methods would be used together in a suitably configured smart safety system, such that the system could comprise detection of either or both of the gas or acoustic signatures of early-stage TR. Gas signatures: The signature gases are gases which are universally or specifically released by TR, as opposed to carbon oxide gases such as carbon monoxide (CO) and carbon dioxide (CO2), which are released by all known forms of combustion and are therefore not reliable indicators of the specific process of TR. There is a wide range of gases to choose from, but research®® indicates that there are certain key gases which are of interest in TR detection, because: • in the context of a normal residential or commercial environment, they are not ambient gases (e.g., unlike Oxygen (02) or 002)); • they are not universally produced by any form of combustion such as non-TR processes (e.g., unlike CO2, CO); • they are produced at an early stage by TR, i.e., at temperatures which do not generally create auto-ignition conditions in materials likely to be in the vicinity of the battery or smart safety device, when first detected. The identified signature gases are: Hydrogen (H2), Methane (CH4), Ethylene (C2H4), Ethane (C2H6), Propylene (C3H6), Propane (C3H8), Hydrogen Fluoride (HF), Phosphoryl Fluoride (P0F3), Fluoroethane (C2H5F), Phosphorus Pentafluoride (PF5), and Butane (C4H10). These signature gases do not exist in a normal ambient atmosphere, and are not produced by appliance leaks or other types of combustion; only TR processes release these gases, therefore their presence is a reliable form of detection for TR; additionally, by not relying on contactless thermal sensing, the detection method is not restricted to fields of view with a direct line of sight between battery and detector (in real world scenarios, the battery might be around the corner from a socket, or there might be an object placed between the battery and the socket); therefore, for the purpose of reliably and specifically detecting the early stages of a TR hazard, this invention makes considerable progress over the prior art. Acoustic Signature: Research^ shows that the safety gas release valve of an EV battery makes a signature sound; high-level characteristics being that the sound is in the 0-8kHz range, with the most common range in the sub-3kHz range; the sound of the safety valve “popping” lasts about 2 seconds, and is amenable to mathematical modelling such as fast Fourier transforms, which lend themselves to de-noising. Appropriate de-noising techniques, well known in the art, can be applied to the treated sound to create a clear acoustic signal, appropriate for detection purposes; it is also a practical matter that the most dangerous time for EV charging is overnight and although acoustic detection suffers from noise interference, in terms of peak vulnerability charging times (overnight, for the resident charging their battery), there will be little or no background noise; thus this is a valid method. Various embodiments of the application of the acoustic method are now discussed. In all cases, these embodiments could be in smart safety device configurations set up for gas detection, or be simply acoustic only. Nothing changes in terms of the acoustic elements; it should be noted that it is a particularly powerful concept to have a smart safety device configured to detect the acoustic signal of a safety valve release and within a few seconds detect the gases which presumably were released, and this combination would make for a virtually foolproof system, which is why it is the preferred embodiment of a smart safety device and system. That said, despite the noise drawbacks in using an acoustic only approach, it is a valid method because, as stated, at the most vulnerable charging times there will usually be very little background noise, and additionally it is a very cost-effective means to achieve TR hazard detection (gas sensors, especially multi-gas sensors, are quite expensive and would put the resultant product out of the price reach of some consumers, especially vulnerable ones). In the simplest embodiment, the electrical conducting device is configured with a microphone (or equivalent sound processing equipment) and a capability for processing sound, typically this is physically achieved by means of appropriate processing hardware on a PCB, but purely software only solutions could be enabled in a processor, or combinations of hardware and software. This embodiment relies on detecting the sound from the nearby battery valve release mechanism (the reasonable presumption being that the socket must be near the device it is charging), then processing the sound and recognising the acoustic signal; appropriate actions can then follow. In other embodiments, the sound processing equipment could be remote from the smart safety device housed in the electrical conducting device, and such devices could be connected to the smart safety device by being in the same smarty safety network, described later. Such remote devices could include a simple device equipped to detect sound, a Digital Voice Assistant (DVA) such as Amazon Alexa® or Apple Siri®, or a smartphone or laptop on which an appropriate app is installed, and any of the foregoing could use a hardware or software solution, or both. In terms of this invention, a remote device is any device which is not a smart safety device, and this would include the electrical conducting device in which the smart safety device is incorporated (given that the electrical conducting device also serves a separate function, being the conduction of electrical current); in this invention, a remote system is a plurality of logically connected remote devices. The app for the smartphone or laptop would include sound recognition software, well known in the art, configured for the parameters of the acoustic signal and with appropriate mathematical processing and de-noising capabilities. The DVA approach would involve much the same process and techniques as the smartphone or laptop method, but would likely need to be developed in collaboration with the DVA manufacturer’s themselves, as it is likely to involve firmware and operating system software. In essence, the above embodiments of laptop, smartphone, and DVA involve turning those items of hardware into “listening devices” for the acoustic signal, and then by means of a wireless network sending the acoustic data to the processor of the smart safety device, to cause the switch off the current supply to the battery. The instruction from the smart safety device to turn off the conduction of electric current through the electrical conducting device could be achieved in a number of means; in preferred embodiments, it would be achieved by means of the computer processor, upon detection of a TR hazard based on the signature data, communicating an instruction to do so to a relay, switch, actuator, or equivalent within the internal mechanisms of the electrical conducting device. In embodiments, the computer processor is further configured to communicate one or more actions in response to the detection of a TR hazard, wherein such actions include but are not limited to an alert of a TR hazard to an authorised user, an instruction to another smart safety device, a remote device or a remote system to perform an action, or sending physical data relating to the TR hazard; such physical data would include, but not be limited to, the data regarding the fire’s location, heat release rate, temperature, toxic gas types and levels, oxygen levels, and smoke levels. In some embodiments, the electrical conducting device in which the smart safety device is located may be equipped with sensors configured to detect a signature gas and the computer processor (of the smart safety device) may communicate with the sensors located within or on the electrical conducting device; in alternative embodiments, the communications device of the computer processor may be configured to receive data regarding a signature gas, and thus the occurrence of a TR hazard, from a remote device in a separate housing equipped with suitable gas sensors. In preferred embodiments wherein the smart safety device is located within the housing of an electrical conducting device, the computer processor, gas sensor, and communications device will likely be mounted on a PCB within the housing; such an arrangement will provide for a low-power configuration and the aforementioned items mounted on the PCB will draw operating power from the electrical supply to which the electrical conducting device are connected. In alternative embodiments, it is also envisaged that the power for the processor, sensors and communications device can be provided by a battery housed within the electrical conducting device. In embodiments, the electrical conducting device which houses the smart safety device and which receives the plug of the EV battery charger could be a mains socket permanently connected to the mains electrical supply, but equally it could be an adapter designed to plug into an electrical mains socket, or an extension lead or power strip designed to plug into an electrical mains socket. The advantage of incorporating the smart safety device in a mains socket is that this encourages the permanent placement of an “EV charging point” in a safe location within a residence (i.e., by not placing such a socket in an exit route, such as a hallway); the advantage of incorporating the smart safety device in an adapter or power strip is the portability that such a configuration offers. The socket housing the smart safety device need not be connected to a mains supply; the socket could be incorporated in an electric vehicle and connected to the vehicle’s battery, or an adapter configured to be plugged into the socket of an electric vehicle, for example an electric vehicle configured for bi-directional charging where one EV battery supplies electricity to another device or battery. In such a case, the position at which the current is switched off could be within the socket of the electric vehicle, or could be anywhere between the EV battery and the socket located within the electric vehicle; all such equivalents are disclosed. The safety device need not always be in installed in a building or EV; it could be installed into a ship, train, aircraft or oil rig, or indeed any crewed vessel. As already stated, in some embodiments the smart safety device could be incorporated into the casing or housing of the EV battery charger itself, wherein the charger would then plug into a power socket, adaptor or power strip. It is intended in embodiments that the smart safety device will form part of a network of devices (hereinafter, a “smart safety network”), which may comprise one or more other smart safety devices, and may also include one or more remote devices, or a remote system. In embodiments, the smart safety device may send or receive communications from any other component of the smart safety network. The smart safety network may have any topology and architecture; for example, it may be a hierarchical topology wherein some components are configured to be local communications hubs, so as to receive communications from other subcomponents configured to communicate only with their local communication hubs, and wherein the hubs then communicate with each other and with any remote devices or systems; alternatively the smart safety network may be a mesh network (non-hierarchical, peer-to-peer wherein any network component can communicate directly with any other component, remote device or remote system), or the network may be a hybrid between the two approaches; in preferred embodiments, the smart safety network would be a full mesh network. The means of communication within the network may be wireless or wired, though wireless is the preferred embodiment; the wireless network may be radio, infrared (IR), optical, laser or other common means of wireless communication; the wireless network can comprise any combination of the foregoing technologies. The smart safety network may be configured to communicate between one or more smart safety devices and one or more remote devices via a wired or wireless local area network or wide area network, or a wired or wireless public communications network, or a satellite communications network. A remote device could be a stand-alone gas detector, or a gas detector incorporated into another device such as a light-switch, or a smoke or heat detector, or a sound detector as already described; this is not intended to be an exhaustive list and in embodiments, a remote device could be any other smart appliance or a separate component of an electric car, or the EV battery itself. It is obvious to any skilled worker that there are alternative, but equivalent, means by which the aims of this invention could be achieved. Such alternative equivalents could, in the most obvious case, include situating not just the sensors but the entire smart safety device, including signature sensors (in or on the smart safety device, or in or on the remote device housing), in a remote device (such as a light-switch, ceiling-mounted smoke or gas detector, purely as examples), and configuring the communications device of the smart safety device to send an instruction to a standard smart socket or smart adaptor (“standard” meaning that there are no signature gas or acoustic sensors in the socket or adaptor), such that the standard smart socket or smart adaptor then turns off the supply of current to the EV battery; all such equivalents, or combinations thereof, which achieve the same effect as the core invention are considered disclosed by this specification. The preferred embodiment is to have the sensors incorporated in the same housing (being that of the electrical conducting device) as the smart safety device (either in or on the smart safety device, or in or on the housing of the electrical conducting device), since it eliminates the risk of the single point of failure, being the risk that the wireless link between the remote sensor and smart safety device could fail; this is set against the potential advantage of having detectors situated higher in rooms to capture the released gas samples (light-switches being placed higher on walls than sockets), but the risk of communications failure suggests that the preferred embodiment is the best approach, albeit that others can be used. The processor of the smart safety device may determine a TR risk by means of gas or sound detectors physically incorporated into the body or housing of the electrical conducting device, and this is the preferred embodiment of the invention. Alternatively however, the processor may receive data regarding a TR risk from a remote device configured with gas or sound sensors, wherein the remote device forms part of the smart safety network of which the smart safety device is a component. In the event that there is a catastrophic TR event, such that a fire is started even after the electric current to the EV battery is switched off, then the information relating to the TR fire can still be communicated by the smart safety device to another smart safety device, or a remote device or system. A remote device could also be a user device, being any device equipped with a visual display, including in embodiments a smartphone, smart TV, computer screen, or smart watch, and the smart safety device could send information regarding the TR risk to any such user device. The data regarding the TR hazard and, if applicable, any physical data regarding a TR fire can be sent to a remote device, a user device, or a remote system, in the form of a digital representation of the building or vessel in which the TR hazard has occurred. The remote device, remote system or user device could be connected to the smart safety network by means of conventional IOT (Internet Of Things) protocols such as WiFi, Zigbee, Thread, Bluetooth, NFC, NBIOT; an expert working the invention would know that these examples are not exhaustive and any suitable IOT protocol or network topology might be used. For example, it might be necessary in some vessels and for network resilience to use satellite communications between the smart safety network and any remote device or system. The remote system to which the smart safety device might communicate data could be a physical computing platform or a cloud-based platform incorporating one or more virtual machines; a cloud-based platform is the preferred embodiment of any system to which the smart safety network might be connected, for reasons of resilience and scalability; such remote systems could be, purely by way of non- restrictive example, a third-party system such as a building or vessel fire protection I alarm system, a building or vessel fire mitigation system (such as a sprinkler or gas suppression system), a building or vessel management system or a system designed to create predictive computational models of a fire. Any of the systems to which the smart safety network may be connected could incorporate artificial intelligence or machine learning, and one or more types of neural network. The smart safety device might also incorporate supplementary sensors to capture other information relating to the physical characteristics of a TR hazard, additional to the data previously mentioned; the electrical conducting device in which the smart safety device may be housed may therefore be equipped with a contactless thermal sensor, a smoke sensor, and one or more sensors for detecting any of the following gases, which are produced either by all forms of combustion, or by the combustion of plastics which are commonly used as casings for EV batteries; Carbon Monoxide (CO), Hydrogen Cyanide (HCN), Carbon Dioxide (CO2), Hydrogen Chloride (HCI), Oxygen (02), Sulphur Dioxide (SO2), Nitrogen Dioxide (NO2). These sensors could alternatively be incorporated into a remote device linked to the smart safety device through the smart safety network; the purpose of these approaches is to create a connected smart safety system. The thermal sensors used in the smart safety system would each comprise an infrared camera. Preferably, each thermal sensor is an infrared camera comprising an array of thermopile detector pixels. In this way, a highly accurate reading of the temperature of the environment surrounding the sensor may be obtained in order that a fire hazard may be determined. The use of thermal imaging allows for the distribution and change in thermal temperature to be measured, allowing for more information to be gathered so as to provide a more reliable identification of a fire hazard at an earlier stage. Typically, the thermal sensors comprise a lens providing a field of view of greater than 30 degrees. This advantageously provides a wide field of view, allowing for reliable detection of a fire hazard. Such a wide field of view advantageously allows monitoring of large areas such as rooms, corridors and stairwells, and provides the best chance to capture images of the EV battery itself. Preferably, the thermal sensors comprise a lens providing a field of view of between 30 and 90 degrees, preferably around 60 degrees. The smart safety device may communicate an alert to a user by means of communicating with a user device forming part of the smart safety network; alternatively, the smart safety network may comprise one or more audible indicators arranged within the building, configured to audibly indicate an alert and information regarding the nature of the TR hazard, including a pre-set range of safety instructions and recommended actions (“stay put”, “evacuate immediately”, being just two examples). Such audible indicators may be in the form of an alarm sounder (which may be incorporated in the smart safety device or the electrical conducting device), or in a remote device, for example, a speaker configured to play a pre-recorded announcement, or a DVAs such as Alexa®, Siri®, Cortana® or Google Assistant®. Alternatively, or in addition to the audible indicators located within the building discussed above, the smart safety device may send an instruction to a remote device in the smart safety network; such a remote device may be, for example, a remotely controlled valve, a router or hub, a docking station for a mobile phone, a fire alarm, a smoke alarm, a sprinkler system, a fire-door, a shut-off valve to any of; a mains electrical supply, a mains water supply, a mains gas supply (i.e. a supply of gas intended for use as a fuel, or “fuel gas”). In the case of shutting of any of the aforementioned mains supplies, this could include shutting off those supplies to a single appliance, a plurality of appliances, a single compartment within the building or vessel, a plurality of compartments within the building or vessel, a floor within the building or vessel, a plurality of floors within the building or vessel, or the entire building or vessel. Another option is to configure the smart safety device with its own wireless voice capability, by installing components within the PCB, or within the housing of the electrical conducting device, which can be configured so as to enable real-time voice communications between a user in the vicinity of the device, and a remote device, system or person; a microphone would need to be installed within or the electrical conducting device to process such voice communications. Preferably, each of the smart safety devices is connected to a battery power source. Such a battery power back-up ensures that if the mains electricity fails or is switched off in response to the detected fire, the smart safety devices continue to operate for a certain amount of time. Each smart safety device is configured to transmit data relating to a TR hazard to a processing unit for analysis to determine an appropriate response. The processor may be located locally within the building, or within the smart safety device itself, or as has already been stated, may be provided as a distributed system (e.g., “Cloud” system). In some examples, the smart safety system may comprise a local processing unit for processing the data received from the sensors and the system may further be configured to send data to a remote processing unit in the cloud, whereby the location in which processing takes place may be selected based on the particular task, the processing requirements, or the current network status. In other (preferred) embodiments, each smart safety device comprises an internal processing unit configured to process data collected by a sensor to determine TR hazard information. As has been stated, the data may be transmitted from each smart safety device either directly or indirectly; for example, each smart safety device may comprise a SIM or Narrowband IOT connection for direct communication with another component of the smart safety network, or the processing unit, over the internet or over the wireless (i.e., cellular) public communications network. Each smart safety device may transmit its respective data to the processing unit indirectly. For example, a smart safety device may be in communication with a node (“hub”) within the smart safety network, which may itself be another smart safety device specially assigned for that purpose, where the hub may communicate the data to the processing unit over an external network, such as the internet or public cellular network. Additionally, a component of the smart safety network may, depending on the network configuration (i.e., mesh versus hierarchical) communicate as appropriate with a remote device or system via the wired public communications network, or via a satellite communications network. Preferably, the smart safety system is configured such that the processing unit receives data from the smart safety devices in real time. In this way, the actual data from the fire can be sent to the processing unit in real time or near real time (e.g., within 1 or 2 seconds of the event). The smart safety system may be additionally configured to automatically contact the emergency services upon determining the presence of a fire. The system may also send an alert to a user device to confirm that the emergency services have been contacted; the system may also send an alert to a user operating a remote device or a remote system, prompting the user to take action based on the alert. Each smart safety device may comprise an illumination device, for example a high-power LED light, configured to illuminate in the case of loss of power. In particular, if a hazard is detected in a dark room, the smart safety device may be instructed to illuminate their respective illumination devices. Figures 1 and 2 illustrate exemplary smart safety systems 1000 (multi-occupancy building or vessel) and 1001 (single occupancy building or vessel) according to embodiments of the present invention; the term “occupancy” here meaning not the number of people, but the number of different IT networks within the building or vessel; for example, hotels, cruise liners, university campuses, hospitals, single organisation office blocks are all examples of single occupancy because in each case there will be a single IT network for the building or vessel, whereas residential blocks and multi-organisational office blocks are examples of multi-occupancy, because each different resident is likely to have their own independent network and connection to the internet; that said, it is possible to install a single-occupancy IT network in a residential building or vessel for the purposes of creating a smart safety network, and this is well-known in the art. Notwithstanding the foregoing distinction between single and multi-occupancy buildings or vessels, systems 1000 and 1001 are otherwise identical in all respects and, unless specified, a reference to one is a reference to either. The smart safety system includes a plurality of smart safety devices and possibly remote devices arranged throughout a building or vessel; in the present embodiment (for purposes purely of example), the smart safety devices are integrated within a plurality of sockets 100, adaptors 200, and charger 300 (only one of each shown, for illustration purposes). A smart safety system according to the invention may comprise one or more housing types and also remote devices. The smart safety devices 100, 200, and 300 may each comprise a communications link so as to be able to communicate with each other via wireless connectivity, for example radio narrow band frequency, Wi-Fi (RTM) or Bluetooth (RTM); any other wireless protocol or technology is equally acceptable, some examples having been given earlier. Preferably the smart safety devices are each configured to communicate over two communication channels such that all of the safety devices can operate on two different types of network, as a failsafe. In the drawings 1 and 2, the safety devices can communicate via Wi-Fi (RTM) 401 and a radio mesh network 402, for example 868MHz. By providing two communications networks, if one network goes down, data obtained from the Appliances arranged throughout the building may still be communicated in order that the location of a fire hazard and the locations of the occupants of the building may be determined. In Figures 1 and 2, lines connecting safety devices 100, 200, 300 and a hub 400 represent a radio mesh network 402 and a Wi-Fi (RTM) mesh network 401. The smart hub 400 could be any secure router capable of wireless communications, and could also be incorporated into a smart safety device within an electrical conducting device (as in Figures 1, where it takes the form of a double-socket housing), or as in Figure 2 where the hub 400 is a separate router. The connectivity of each of the safety devices in the local network may be managed by a smart hub 400 which is connected to a central router (not shown) within the building I vessel. If the building I vessel is multi-occupancy, the smart hub 400 itself may be a smarty safety device itself (as per Figure 1). In preferred embodiments in the case of a multi-occupancy building I vessel, the smart hub is integrated within one of the smart safety devices 100, 200. In cases where the building I vessel is single occupancy, it is likely and preferable that the hub 400 will be one or more separate router devices inside the building I vessel (as per Figure 2). Although not shown in Figures 1 and 2, the smart safety devices 100, 200, 300 will be configured to link to the nearest hub 400 in their vicinity, but since there will be a plurality of devices 100, 200, 300 and hubs 400 arranged throughout the building, in the event of any devices 100, 200, 300 losing connectivity to their nearest hub 400, said devices will automatically “roam” and link to the next nearest hub 400, in order to maintain connectivity to the cloud platform 601; alternatively, some wireless technologies (Thread™ being an example) include the capability for a single smart safety device to be “nominated” by other devices as the hub, and in the event of the failure of that hub the network automatically nominates a replacement device to act as hub, as part of the features of the mesh network). As will be described in more detail later, in the event that all the hubs 400 are unavailable for any device 100, 200, 300 then said devices will communicate directly with the cloud platform 601 by means of an independent wireless communications capability which all devices will be configured with (such as a SIM capable of being linked directly to the public cellular network or satellite communications), and this “failover” independent link is shown in Figures 1 and 2. As the network of safety devices are in local communication with each other, alarms or other safety notifications may be initiated quickly in response to a detected hazard. For example, if a TR hazard is detected that is indicative of the presence of a fire, this information may be communicated to the other safety devices within the building over the local smart safety networks 401, 402. In response, the smart safety devices 100, 200, 300 may initiate an integrated alarm sounder to warn occupants of the fire hazard; alternatively, an alarm can be sounded by connecting the smart safety devices to digital voice assistants, a feature which can be offered as part of a “user app”. In Figures 1 and 2, the local mesh network 402 is shown as having single connections between different nodes; this is for ease of illustration only. Typically, such networks will communicate by two different means, usually Wi-Fi (RTM) and a radio network on an 866Mhz or 868Mhz frequency, for instance, such that if one network fails then communications between devices can continue on the other. Both communications networks 401 and 402, described in the foregoing, are included in the mesh network and although not separated out in Figures 1 and 2, nonetheless they are understood to be different, redundant networks in the same mesh network. Additionally, it is likely that the smart safety network and the communication network with the cloud is fully secured, using an encrypted VPN (purely as an example, any equivalent alternative would be acceptable). The data processed by a cloud-based processing unit 500 may be communicated to one or more remote devices 700 via a communications device 510, or alternative means. A particular remote device may be a smartphone 700 or laptop 701. The smartphone (or other smart user device) may run an app with which the user can receive alerts and notifications from the processing unit 500, via a communications device 510, which are indicative of the location of the fire hazard and instructions regarding what to do next. This information may be displayed on the remote device, for example a smartphone 700, configured to receive data from the device 510, which is displayed to the user on the smartphone screen. In the event of a loss of connectivity between the communications hub and the cloud (601), the smart safety devices can (through Wi-Fi (RTM) or Bluetooth (RTM) or other “localised” communication means, link to any smart device on which the “app” is installed, in order to provide the same functions as if the loss of connectivity to the cloud had not occurred. In this scenario, the smart device would (temporarily at least) form part of the mesh network. Typically, a remote device such as a smart phone or other smart device 700 may run one or two forms of software, or “app”: (1) a user app (intended for residents), showing alerts, messages, evacuation information and instructions and (2) a responders’ app, (intended for emergency services and building authorities) showing the status of the fire as a whole and the evacuees throughout the building. The software may also utilise a smart device’s inbuilt navigation I tracking system (e.g., integrated GNSS and inertial sensors) to determine metrics of a building occupant’s motion (e.g., position I direction of motion) in relation to the determined safe route out of the building, in order to assist the user in a safe evacuation. The smart safety system may comprise a communications device in the form of one or more speakers 777, which in preferred embodiments may also be in the form of device configured to interact with a voice assistant such as Alexa™ or Siri™. The speakers may be configured to sound an announcement describing evacuation instructions, which may be changed in real-time as the fire develops, avoiding the detected location of the first hazard. The speakers 777 each comprise a wireless communications link for receiving a signal transmitted from the processing unit 500, whereby the speakers 777 may be actuated. In other embodiments, a local processor, typically located on the smart hub 400, may communicate with the speakers or voice assistant 777 (e.g., over the smart safety network). In some embodiments, each device 100, 200, 300 may comprise communication means configured to communicate directly with the processing unit 500. For example, each device may comprise a SIM card, preferably an NBIOT card (Narrowband Internet of Things) allowing it to connect to the internet. This would particularly be the case when the device is in the form of a portable item, such as an adapter or power strip; in embodiments, it is envisaged that all smart safety devices may connect first to a local network (such as WiFi, Bluetooth etc) as the primary communications configuration, with the use of a SIM connected to the cellular network as a backup; in the case where the smart safety device cannot connect to a WiFi (or equivalent), the SIM would be the default alternative. BRIEF DESCRIPTION OF THE DRAWINGS: Embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which: Figures 1 and 2 illustrate exemplary smart safety system according to embodiments of the invention in (1) a multi-occupancy building and (2) a single occupancy building, as have been already described in detail; Figure 3 illustrates an example of a smart safety device incorporated within a socket faceplate 100 from the outside, front view; Figure 4 illustrates an example of a network architecture for a smart safety system based on the acoustic signature method; In all cases, where the drawings refer to a cloud-based system, it is re-stated that the processing platform need not be cloud-based; it could perfectly well be a physical computing platform (for example, in a ship or in a data-centre). DETAILED DESCRIPTION OF FIGURES 3 AND 4: The embodiment shown in Figure 3 is that of a smart safety device incorporated within a socket, incorporating sensors for signature gases and supplementary sensors (wherein the sensors are incorporated either directly in or on the smart safety device, or within or on the socket, neither of which are shown); in the example of Figure 3, the socket is intended to be a mains socket but as previously described, the illustration in Figure 3 could be a socket installed in an electric vehicle configured for bi-directional charging, or could be a socket in a vessel. Signature gas and supplementary sensors are incorporated into the faceplate. Preferably, the sensors may be positioned on the front of a socket faceplate 100 in order to achieve the best vantage point to detect a TR hazard in the vicinity of the socket; through the combination of the two types of sensors (signature gas and supplementary), a smart safety device housed in a socket faceplate 100 can, in the same device, detect the presence of one or more signature gases, elevated temperatures and smoke levels that are indicative of the early-stage presence or risk of a fire (a TR hazard). In the example of Figure 3, the signature gas 120 is positioned within a housing 150 of the device 100 at two points, on the upper left and upper right corners of the faceplate. The configuration shown in Figure 3 is for two separate signature gas sensors (which could be H2 and CH4, or any other combination of any of the disclosed signature gases) if mono-gas sensors, although is it perfectly possible for multiple gases to be sensed by multi-gas sensors designed to detect a plurality of signature gases, in which case the configuration of sensors 120 is intended to provide a positioning to enable the fastest gas detection system possible, by virtue of having sensors in more than one location on the front of the faceplate. The supplementary sensor 130 is located in the middle upper portion of the faceplate housing 150, directly above the switches; this sensor could be a monogas sensor configured to detect any of the supplementary gases previously disclosed; equally, as for sensors 120, the sensor 130 could be a multi-gas sensor to detect a plurality of supplementary gases. In the case of the gas sensors 120 and 130 as shown in Figure 3, the sensors are mounted on a PCB (not shown) and the areas attributed as sensors as shown in Figure 3 are in reality grilles or other equivalent mechanisms for the ingestion of air samples within the actual sensors located on the PCB inside the faceplate; this is purely by way of example, other methods of sample capture may be used. The microphone 180 is a two-way microphone for receiving voice communications from one or more persons in the vicinity of the faceplate 100 and for broadcasting voice communications from a remote person or system; in other embodiments, the broadcasting of voice communication may come from the speaker of a voice assistant, such as Alexa™, to which the microphone 180 can be connected in a smart safety network. The foregoing configuration enables the smart safety device to be housed in the socket faceplate and enable the device to provide its full range of functions, without stopping the faceplate from fulfilling its other function of enabling the transmission of electrical power to an appliance, by means of a plug; this is advantageous in that it reduces the number of devices that are required in a typical residence by utilising the connections and locations of other devices, such as sockets, which are always present in a room in any event. Equally, for the acoustic method, the microphone 180 may be a one-way input only device, whereby it is used simply to detect an acoustic signature of the battery safety valve being released. In the example of Figure 3 the socket face plate 100 comprises a double socket housing 150, affixed to a surface by screws 101, but such fire and occupant safety devices could equally have a single socket housing 150 or a greater number of sockets in the housing 150, and could be affixed by means other than screws. In the example of Figure 3, the thermal sensor 141 is placed at the bottom of the centre of the faceplate, to have the best and widest overall field of view of the surroundings; the thermal sensor 141 therefore allows for the early detection of temperature increases associated with possible fire hazards, typically within a compartment of the building or vessel. In embodiments, in addition to the wireless mesh network connection between the device 100 and any hub 400, through which the smart safety devices communicate with the processing unit 500 (within the cloud platform 601), each device 100 is configured so as to be capable of wireless communication directly with the processing unit 500. This will typically be achieved by configuring the devices 100 with a SIM of some kind, most likely a Narrowband Internet of Things (NBIOT) SIM card; in this scenario, any loss of communication between a particular device 100 and the processing unit 500 due to, for example, any of; a catastrophic failure of all devices 100, or all hubs 400, or of all third-party systems, or of all wired or wireless connectivity from the building to the internet, can be overcome by means of this independent “failover” communication capability between the device 100 and the processing unit 500. This is a form of redundancy of the last resort. In the event of a TR hazard in the vicinity of the smart safety device in socket 100, the same local processor inside the smart safety device (not shown) located within the faceplate device 100 determines the presence of a possible fire risk and can take a number of actions; the socket 100 firstly comprises an audible alarm sounder 140 which is configured to sound when a hazard is detected in order to alert the occupants in the immediate surrounding area; in some embodiments, said processor in the faceplate 100 might send an alert a voice assistant such as Alex™, so that the occupants can be alerted to the threat of the fire elsewhere. The data regarding the fire is then fed to a hub 400 to which the device 100 is connected, and thence to all other devices in the smart safety network and any third-party systems to which the mesh network is connected (in the manner previously described) and ultimately to the processing unit 500 in the cloud 601; the processing unit 500 could refer to an alert policy and further alerts sent to other occupants (e.g., to a registered user’s smartphone), in line with a pre-determined alert policy and the threat level of the fire. The socket housing of the smart safety device 100 could contain a reset switch (not shown in Figure 3), for resetting the device 100 or silencing the alarm 140 when it is sounding. The electrical safety device 100 further comprises a series of status LEDs to indicate to a user that the device 100 is functioning correctly. In particular the device of Figure 3 includes a corresponding LED to indicate the status for the network connectivity, the power to the device and the sounding of the alarm. The series of LEDs 142 are provided on the surface of the housing 150 to provide visual alert to the user. The series of LEDs 142 are provided on the surface of the housing 150 to provide visual alert to the user; in preferable embodiments, the green LEDS (indicative of a fully functional system) can be disabled so as not to shine, at night, and give off an intrusive light, if the device is installed in a room in which persons may be sleeping; this may be accomplished by a manual switch, or configured in an app downloaded on a smart phone, or may be automatically configured using a photosensitive cell to detect conditions of darkness. In addition to the LED indicators 142, in certain embodiments there may be a single white light LED (not shown) incorporated into the housing, which is capable of emitting a bright light in the case where a fire is detected in conjunction with poor ambient visibility (whether due to darkness or thick smoke), in order to help illuminate the surroundings for occupants trying to gain their bearings. The smart safety device 100 may equally be configured to communicate via the wireless communications link with other user devices such as a smart TV, smart watch, or voice assistant such as Alexa™ or Siri™ or other devices to indicate the presence of a potential hazard and provide details on the hazard detected. The device 100 is also configured to send and receive voice communications, via the microphone 180 (if suitably configured), between one or more persons in the vicinity of the device and a remote system or persons, in order to establish the medical condition of any persons and to allow said remote device or person to communicate safety information to said persons. The local processor of the device 100 is configured to determine the presence of a potential hazard by identifying when the value of a sensed parameter (e.g., ambient levels of a signature or supplementary gas, temperature from the thermal sensor, smoke from the smoke sensor) exceeds a predetermined threshold value. However more complex processing may be used to identify the presence of a hazard, for example by identifying a rate of change of a sensed parameter or where a sensed parameter change displays a particular behaviour or pattern associated with an increased risk of a hazard. The local processor can also be configured to determine the presence of a hazard based on a combination of sensor outputs in order to identify a risk more reliably. For example, the processor can use more complex algorithms, such as machine learning based algorithms which take the output from multiple sensors in order to determine an elevated risk. For example, in a situation where the signature or supplementary gas sensor and the thermal sensor are lower than their corresponding individual thresholds, the behaviour of the sensor readings in combination may signify a developing hazard and therefore this can be detected at an earlier stage than with a single sensor. Similarly, an unusual rate of change of one or more parameters may indicate the presence of a hazard. Data from the plurality of further sensors may be communicated to the processing unit 500 or via the device communications link and the smart hub 400 for further analysis and identification of the hazard location. The fact that many of the signature gases and supplementary gases are both hazardous (toxic and I or flammable) and not found in normal ambient atmospheres means that a simple and effective threshold for detecting a TR hazard would be any level above a trace (trivial) level, such levels known within the art and easily sourced in publicly available publications. In the case of a thermal sensor 141, the detector is an array of pixels sensitive to infra-red radiation, most likely in the Long Wavelength Infra-Red (LWIR) spectrum (8 to 14 micrometre range) arranged behind a suitable lens; each such thermal sensor is provided by an infrared sensor, in particular an infrared camera comprising an array of infrared detector pixels. The infrared array sensor may comprise an 8x8 grid array of thermopile elements that detect absolute temperature by measuring the emitted infrared radiation. The infrared array sensors are able to provide thermal images by measuring actual temperature and temperature gradients, allowing highly precise measurements of surface temperature and identification of changes in temperature. Such a large viewing angle is also useful for monitoring large spaces such as rooms, corridors within a building I vessel. The lens may comprise an integral silicon lens which provides a viewing angle of around 60 degrees. There are several ways of detecting signature and supplementary gases in an air sample, some examples being; Resistive detection, also known as chemiresistor detection, which offers high sensitivity and stability; Non-Dispersive Infra-Red detection (NDIR) often used to capture CO2 and combustible gases; Photoacoustic detection; Metal Oxide Semiconductor detection (MOS); Electrochemical detection often used to capture CO and 02; Catalytic detection often used to detect CH4 and C3H8; Photoionisation detection (PID) often used to detect Volatile Organic Compounds, also known as VOCs. An advantage of many of the above is that they can be used to detect the presence of many different gas molecules in a given air sample, which is of particular value in terms of providing supplementary data on other toxic gases typically produced by combustion and have a long operating lifespan (10 years plus). The signature and supplementary sensors can use any of the above methods, or indeed other methods not disclosed in the list of above examples can be used; key considerations would be to optimise the sensors by capturing as many signature and supplementary gases as possible with the minimum number of sensors, subject to considerations of cost and physical constraints created by the need to fit such sensors on a PCB capable of being housed inside an electrical conducting device, or within the electrical conducting device itself; different housings may therefore lead to different combinations of gas being detected, for reasons of physical space constraints and practicality. In embodiments, the signature and supplementary gas sensors 120 and 130 can be located in multiple points on the surface of, or within the interior of, the socket. Any of the sensors, LEDs, microphones, re-set switches of the socket 100, as per Figure 3, could be located in any other configuration; although Figure 3 shows a “Type G” socket, it is recognised that there are multiple different form-factors and regulations for sockets in different legal jurisdictions, so it is not possible to be prescriptive in any drawing on the exact shape of a socket, or the configuration of sensors and other functionality on or inside the socket; it is expected that a person working the invention in a particular jurisdiction will experiment with different locations and types of sensors to arrive at an optimum configuration, being a configuration which captures as many gas samples as quickly as possible, provides for the widest field of view of thermal sensors, enables communications between the smart safety device and other devices in the smart safety network, and enables the foregoing to be achieved within minimum power usage and wherein the various components of the smart safety system are installed within the socket such that the socket can still receive a plug and conduct electric current. The embodiment of Figure 4 shows an example of a network architecture configured to implement the acoustic signal detection method, using remote devices to detect the sound of the safety release valve in operation; in the example of Figure 4, the remote devices include a DVA 777 and a laptop 701 (on which an appropriate sound processing app (not shown) is installed) and this is purely exemplary, as in reality it may be that only one of these devices would be used or that another device might be used (for example, a smartphone 700). The remote devices will be part of the smart safety network, but it is likely that they will not themselves have direct access failover access to the cloud platform 601 and will send the acoustic data to the socket 100 and / or the hub 400, leaving those devices to take appropriate local action and communicate with the cloud 601. The more listening devices are used to detect the acoustic signal, the more accurate the system will likely be due to the ability to cross-reference the same event, detected simultaneously on several devices. The example of Figure 4 shows an acoustic signal architecture in a multi-occupancy environment 1000, but it is the case that the acoustic approach could be deployed in a single-occupancy environment 1001 in the same way, with the hub 400 being a separate router rather than being integrated within a device 100, as an example. As previously described, the socket 100 could be configured with its own sound processing capability, and could also have signature gas detection capability and neither of these are shown in Figure 4, for reasons of clarity, but are understood to be there (although, the acoustic method could easily run in standalone mode). Having described the essential hardware and network components, we now briefly describe an exemplary architecture for the cloud platform, a preferred embodiment for the processing unit 500. Cloud Platform 601 In preferred embodiments, the processing unit 500 is a virtual computing device (a computational system wherein the computing processor, memory and storage is separated from the underlying physical hardware) and which has access to data and executable code relating to said data, located in the cloud 601 and connected to the internet by means of a communications device, or “layer”, which may itself also be a virtual device or cloud server. The cloud environment may be “private” (i.e., the computing software and memory is run on an array of dedicated physical hardware inside a specific datacentre), or “public”, wherein the computing platforms are “elastic”, meaning infinitely and almost instantly scalable and run on non-dedicated hardware simultaneously in multiple and physically separate data-centres), or they may be “hybrid” which is a mixture of both types previously mentioned, wherein the main processing is undertaken by the private cloud environment and the public cloud is used for “burst-outs”, whereby a sudden and dramatic spike in computational resource requirement, such as in the event of a fire, is dealt with by utilising the elastic computing capability of the public cloud. For the purposes of example only, the following description is based on a high-level design for deploying the processing unit 500 inside a public cloud. As illustrated by Figures 1 and 2, the connections from devices 100, 200, 300 or the one or more hubs 400 to the cloud 601 are likely to be a combination of wireless or wired, depending on whether the building or vessel is multi-occupancy or single occupancy, and wherein the wireless network is the internet, the public cellular network, or a satellite communications network and wherein the wired network may be the public communications network. The distinction between single and multi-occupancy buildings I vessels is important from a network architecture perspective; a single occupancy building I vessel may be a single digital entity in the form of a Local Area Network (LAN) which pervades the entire building / vessel through a local mesh network which is wired, wireless or more likely both and will have an interface with the internet such as a router, which by virtue of the single occupancy can be stored in a safe and secure location. In such a case, the devices can communicate with the router by means of the LAN and thence to the cloud and receive communications back from the cloud by the same route. Alternatively, the smart safety system could be configured to have its own separate network (preferably mesh) and its own internet router for the purpose described in the foregoing and, for reasons of system resiliency and redundancy, this is the recommended approach. As previously stated, the smart safety devices will always have an independent means to wirelessly communicate with the internet via SIMs (e.g., NBIOT), which is necessary for system redundancy in case the router connectivity to the internet is lost (for example, a fire, water leak or power cut could disable the router). It is also possible to configure each smart safety device to communicate via the mobile cellular network (i.e., via a SIM) as the default communication method. As already stated in the case of a multi-occupancy building I vessel, there will usually not be a single network within it, and this is most often the case when the multi-occupancy building I vessel is residential. In such a case, there may not be a suitable and safe location for an internet router and there may be no LAN for the building I vessel as a single “digital entity”; in this example, the preferred architecture is for each of the devices to have wireless communication directly with the cloud 601, preferably by using SIMs (e.g., NBIOT); it is also desirable to have backup to the public cellular network, which could be achieved by configuring all the SIMs to communicate via satellite communications, if necessary. When a fire is detected by any device 100, 200, 300, the information gathered by the appropriate sensors detecting the fire is fed back to the processing unit 500, and can then be sent to any connected remote device or system. References: GB 2578238 A (PARFITT) 2020, UK IPO; See whole document; Electrical Safety First, “Battery Breakdown Report (2023)"', w Wang (Z) et al, “Gas Sensing Technology for the Detection and Early Warning of Battery Thermal Runaway; A Review.” Published 27th May 2022, American Chemical Society (https: / / doi.org / 10.1021 / acs.energyfuels.2c01121); (3) Wang (K) et al, “Early Warning Method and Fire Extinguishing Technology of Lithium-Iron Battery Thermal Runaway; A Review”. Published 23rd March 2023, MDPI (https: / / doi.org / 10.3390 / en16072960); (4) Jin (Y) et al, “safety warning of lithium-iron battery energy storage station via venting acoustic signal detection for grid application”. Published 6th April 2021, Elsevier Ltd (https: / / doi.Org / 10.1016 / j.est.2021.102498).

Claims

1. A smart safety system for early detection of thermal runaway during electric vehicle battery charging, comprising; one or more smart safety devices (hereinafter, a “safety device”) comprising a computer processor configured with a communications device, wherein the computer processor is in communication with one or more sensors configured to detect Thermal Runaway (hereinafter, “TR”) by means of detecting the ambient levels of one or more gases (hereinafter, “signature gases”) typically released during TR and indicative of the risk or early-stage presence of a TR-induced fire (hereinafter, a “TR hazard”) in the vicinity of the safety device, wherein the one or more signature gases comprise at least one of the following:Hydrogen (H2), Methane (CH4), Ethylene (C2H4), Ethane (C2H6), Propylene (CsHe), Propane (CsHs), Hydrogen Fluoride (HF), Phosphoryl Fluoride (POF3), Fluoroethane (C2H5F), Phosphorus Pentafluoride (PF5), Butane (C4H10), and;wherein the safety device is located within or on the faceplate or housing of an electrical conducting device configured to conduct an electrical current into, or from, an electric vehicle battery (hereinafter, an “EV battery”), and wherein the computer processor is further configured, upon detection of a TR hazard, to communicate an instruction to the electrical conducting device to turn off the supply of electrical current to, or from, the EV battery.

2. The system of claim 1, wherein the electrical conducting device is any of: an electrical socket, being any of:a mains socket within a building;a socket within a vessel;a socket within an electric vehicle;an electrical adaptor;a power strip, or extension lead;a charger configured to connect the EV battery to an electrical supply.

3. The system of claim 1 or claim 2, wherein the signature gas sensors are located within or on any of:the safety device;a remote device, being any of:the electrical conducting device;any other device that is not the safety device.

4. The system of any of the preceding claims, wherein the safety device is in communication with an auditory sensor for detecting an acoustic signal indicative of a TR hazard (hereinafter, an “acoustic signature”), and wherein the computer processor is configured, upon detection of an acoustic signature, to turn off the supply of electric current to, or from, the EV battery.

5. The system of claim 4, wherein the auditory sensors are hardware or software based, or both, and located or installed within or on any of: the safety device;a remote device, being any of:the electrical conducting device;any other device that is not the safety device.

6. The system of claim 5, wherein the other device is any of:a digital voice assistant;a smartphone;a laptop.

7. The system of any of the preceding claims, wherein the safety device is in communication with one or more supplementary sensors configured to detect supplementary data regarding other aspects of the TR hazard, said supplementary sensors to include the following:one or more thermal sensors;one or more smoke sensors;one or more sensors to detect any of the following gases:carbon monoxide;carbon dioxide;hydrogen cyanide;hydrogen chloride;oxygen;sulphur dioxide;nitrogen dioxide; andwherein the supplementary sensors are incorporated within or on the safety device or the electrical conducting device.

8. The system of any of the preceding claims, wherein an audible alarm is incorporated within or on the electrical conducting device, the alarm configured to sound a warning upon detection by the computer processor of a TR hazard.

9. The system of any of the preceding claims, wherein the communications device is configured to communicate data relating to the TR hazard to any of: another safety device;a remote device;a remote system, being a plurality of logically connected remote devices;a user operating a remote device or system.

10. The system of claim 9, wherein the communications device is configured to communicate:an instruction to another safety device;an instruction to a remote device or system;an alert to a user operating a remote device or system;wherein the instruction or alert relates to a TR hazard and wherein the communication can be in the form of data or voice.

11. The system of claim 10, wherein the communications device is configured to communicate:an instruction to another safety device;an instruction to a remote device or system;an alert to a user operating a remote device or system;to shut down one or more mains supplies wherein said mains supplies comprise any of:electrical current;fuel gas;water;to any of;a single appliance;a plurality of appliances;a single compartment within a building or vessel;a plurality of compartments within a building or vessel;a floor within a building or vessel;a plurality of floors within a building or vessel;an entire building or vessel;in which a TR hazard is detected.

12. The system of any of claims 9 to 11, wherein the communications device is configured to communicate any of the data relating to a TR hazard to a remote device or system, wherein the data is in the form of a digital representation of a building or vessel.

13. The system of any of claims 9 to 12, wherein the communications device is configured to act as a component of a secure network (hereinafter, a “smart safety network), wherein the smart safety network is a wired or wireless network within a building or vessel, and the smart safety network is configured to be any of:a full mesh network, being a non-hierarchical, peer-to-peer network wherein any network component can communicate directly with another, and any component can communicate with a remote device or system; a hierarchical network, being a network where some components communicate only with a nominated local hub, and the local hub then communicates with any other hub and any remote device or system.

14. The system of claim 13, wherein the smart safety network communicates with a remote device or system by means of any of:a wired or wireless local area network;a wired or wireless wide area network;5 a wired or wireless public communications network;the internet;a satellite communications network.

15. The system of any of claims 9 to 14, wherein the remote device or system 10 includes any of:a system for notifying the emergency services of a TR hazard;a third-party system, being any of:a fire-protection or fire-alarm system for a building or vessel;a fire mitigation system for a building or vessel;15 a building or vessel management system;a building or vessel evacuation system;a system configured to create predictive computer models of a fire.

Images