Air purification system

Abstract

An air purification system includes: an air purification device; and a network server. The air purification apparatus includes: an air purification unit configured to be operated to purify air in the vicinity of the air purification apparatus; and a communication interface, wherein the air purification device is configured to communicate with the network server via the communication interface. The air purification device is configured to operate the air purification unit according to a control algorithm that uses, as an input, sensor data representative of a composition of air in the vicinity of the air purification device. The network server is configured to update a control algorithm in accordance with sensor data representative of a composition of air in the vicinity of the air purification device to change how the air purification device operates the air purification unit.

Description

Technical Field

[0001] The invention relates to an air purification system. Background Art

[0002] In order to maintain or improve the air quality of an indoor environment, it is known to use air purification devices. An air purification device generally comprises an air purification unit which is configured to be operated to purify the air in the vicinity of the air purification device.

[0003] The air purification unit may, for example, include a fan assembly that draws air into the air purification device, and a purification assembly configured to purify the air drawn into the air purification device by the fan.

[0004] The purification component can be configured to purify air (drawn into the air purification device by the fan) by, for example, reducing the concentration of one or more chemical elements, chemical compounds, types of chemical compounds, and / or other substances (e.g., particulate matter) present in the air passing through the device. For example, the chemical compounds may include volatile organic compounds (VOCs), nitrogen oxides, formaldehyde, carbon dioxide, carbon monoxide. One or more other materials may include smoke and / or particulate matter. One or more chemical elements may include lead particles. The purification component may include one or more filter elements.

[0005] The known air cleaning device comprises a sensor unit which is configured to provide sensor data representing the composition of air in the vicinity of the air cleaning device, and an air cleaning system of the air cleaning device is controlled depending on the sensor data provided by the sensor unit.

[0006] The inventors have observed that a problem with known air purification devices is that the control algorithms used to control the air purification devices based on sensor data are very simple and tend to turn on the device when the measured value exceeds a threshold. In addition, these algorithms tend to be fixed and do not change based on user preferences or newly identified harmful compounds. The inventors have observed that this results in users having difficulty controlling what odors are removed by their air purification devices.

[0007] The present invention has been designed in view of the above considerations.

[0008] WO2019 / 074256 discloses an odor mitigation system for a vehicle that can use a sensor such as an electronic nose to detect tastes or odors in a vehicle. The user of the vehicle can indicate whether the detected odor is pleasant or unpleasant, so that mitigation methods for unpleasant or undesirable odors can be recommended.

[0009] The present invention has been designed in view of the above considerations. Summary of the invention

[0010] In a first aspect, the present invention provides an air purification system, comprising:

[0011] Air purification equipment; and

[0012] Network servers;

[0013] Among them, air purification equipment includes:

[0014] an air purification unit configured to be operated to purify air in the vicinity of the air purification device; and

[0015] a communication interface, wherein the air purification device is configured to communicate with a network server via the communication interface;

[0016] wherein the air purification device is configured to operate the air purification unit according to a control algorithm, the control algorithm using as input sensor data representing the composition of air in the vicinity of the air purification device;

[0017] The network server is configured to update a control algorithm to change how the air purification device operates an air purification unit based on sensor data representing a composition of air near the air purification device.

[0018] In this way, the network server can change how the air purification device responds to sensor data, for example, in order to improve the performance of the air purification device, for example, by changing to a newly identified harmful compound.

[0019] In some examples, the network server can be configured to update the control algorithm based on sensor data associated with chemical elements, chemical compounds, classes of chemical compounds, and / or other materials that have been identified as harmful to the user. For example, an entity controlling the network server (e.g., a supplier of an air purification device) may have identified a new compound or class of compounds that has been identified as harmful, and the control algorithm can be updated so that the air purification device operates to reduce the amount of the compound or class of compounds present in the vicinity of the air purification device.

[0020] In some examples, the network server may be configured to update the control algorithm based on sensor data associated with user input data indicating a user preference regarding the composition of air near the air purification device.

[0021] In this way, the control algorithm can be updated in a way that matches the user's preferences.

[0022] In some examples, the air purification device may include a user interface, and the user input data may be obtained from user input provided through the user interface.

[0023] In some examples, the air purification device may include a user interface. In other examples, the user interface may be included in a device (e.g., a mobile device such as a smartphone) connected to the air purification device (e.g., via a communication interface of the air purification device).

[0024] The user interface may be configured to allow the user to provide a "positive opinion" input indicating that the user has a positive opinion of the smell of the air near the device when the "positive opinion" input is provided. The user input data may include data associating the user's positive opinion with sensor data representing the composition of the air near the air purification device when the "positive opinion" input is provided.

[0025] The user interface may be configured to allow the user to provide a "negative opinion" input indicating that the user has a negative opinion of the smell of the air near the device when the "negative opinion" input is provided. The user input data may include data associating the user's negative opinion with sensor data representing the composition of the air near the air purification device when the "negative opinion" input is provided.

[0026] The user interface may be configured to allow the user to provide a "neutral opinion" input indicating that the user has a neutral opinion (i.e., neither a positive opinion nor a negative opinion) about the smell of the air near the device when the "neutral opinion" input is provided. The user input data may include data associating the user's neutral opinion with sensor data representing the composition of the air near the air purification device when the "neutral opinion" input is provided.

[0027] In some examples, there may be only one type of "positive opinion" input (e.g., corresponding to a thumbs up button) and one type of "negative opinion" input (e.g., corresponding to a thumbs down button). However, in other examples, the user interface may be configured to allow the user to provide further "opinion" input. For example, a five-point rating of "very negative," "negative," "neutral," "positive," and "very positive" may be provided. Alternative ratings of "positive opinion" and / or "negative opinion" may also be used.

[0028] The association of the input (indicating that the user had a given opinion about the smell of the air near the device at the time the input was provided) with the sensor data may be achieved by any suitable means known to those skilled in the art.

[0029] In some examples, the sensor data used to update the control algorithm includes aggregated sensor data associated with a plurality of users, wherein the plurality of users includes at least one user of an air purification device other than the air purification device.

[0030] In this way, sensor data from multiple users of multiple air purification devices may be used to improve the performance of the air purification devices without having to rely on input provided by the users of the air purification devices.

[0031] In some examples, multiple users may belong to a limited group of people that share at least one characteristic with the user of the air purification device.

[0032] In this way, sensor data from multiple users may be utilized to improve the performance of an air purification device in a manner that reflects the user preferences (by sharing at least one characteristic), without having to rely on the users to provide input regarding their preferences.

[0033] For example, members of a crowd may share a geographic location feature, for example, members may belong to the same geographic region (e.g., country, group of countries, continent). This may be useful if people in the geographic region share preferences regarding desired / undesirable flavors.

[0034] For example, members of a population may share an age profile, e.g., members may belong to the same age group. This may be useful if people in that age group have common preferences for desirable / undesirable flavors.

[0035] For example, members of a crowd may share gender characteristics, e.g., members may belong to the same gender category (e.g., male, female, gender neutral, heterosexual, transgender). This may be useful if people in that gender category share preferences regarding desired / undesirable flavors.

[0036] For example, members of a crowd may share user preference characteristics regarding the composition of the air near an air purification device. For example, members of a crowd may all have similar preferences regarding smells they like and / or don't like (not necessarily belonging to the same geographic region, age group, or gender).

[0037] The air purification system may be configured to categorize users of the air purification devices into populations (that share user preference characteristics regarding the composition of the air near the air purification devices) based on user input responsive to the composition of the air near the air purification devices.

[0038] The air purification system can be configured to use a machine learning model to form crowds and classify users into crowds. The machine learning model used to form crowds based on shared user preference features can be a clustering algorithm.

[0039] In some examples:

[0040] The control algorithm may include a machine learning model; and

[0041] The network server can be configured to update the control algorithm by updating the machine learning model using the training data.

[0042] The training data may include sensor data associated with user input data indicating a user preference regarding the composition of air proximate the air purification device.

[0043] The training data may include aggregated sensor data associated with a plurality of users, wherein the plurality of users includes at least one user of an air purification device other than the air purification device.

[0044] The training data may include sensor data associated with a chemical element, chemical compound, or class of chemical compounds that has been identified as being harmful to a user (eg, by an original equipment manufacturer OEM).

[0045] In some examples, the machine learning model is located at a network server, and the network server is configured to instruct the air purification unit to operate according to the output of the machine learning model.

[0046] In some examples, the machine learning model is located at the air purification device, and the network server is configured to update the control algorithm to change how the air purification device operates the air purification unit by updating the machine learning model located at the air purification device.

[0047] In some examples where the machine learning model is located at the air purification device, the network server can be configured to update the control algorithm by sending an updated machine learning model to the air purification device for use by the air purification device to change how the air purification device operates the air purification unit.

[0048] In some examples where the machine learning model is located at the air purification device, the network server can be configured to update the control algorithm by sending training data to the air purification device to change how the air purification device operates the air purification unit, so that the air purification device updates the machine learning model using the training data.

[0049] In some examples, operating the air purification unit according to the control algorithm includes the air purification system:

[0050] determining from the sensor data whether air in the vicinity of the air purification device contains an undesirable chemical element, chemical compound, or class of chemical compounds; and

[0051] If it is determined that the air near the air purification device contains an undesirable chemical element, chemical compound, or class of chemical compounds, the air purification unit is operated to purify the air near the air purification device.

[0052] The chemical element, chemical compound, or class of chemical compounds may be determined to be undesirable based at least in part on user input indicating a user preference regarding the composition of air proximate the air purification device.

[0053] The chemical element, the chemical compound, or the class of chemical compounds may be determined to be undesirable based at least in part on the chemical element, the chemical compound, or the class of chemical compounds being identified (eg, by a manufacturer of the device) as being harmful to a user.

[0054] In some examples, operating the air purification unit according to the control algorithm includes the air purification system:

[0055] determining from the generated sensor data whether air in the vicinity of the air purification device contains a desired chemical element, chemical compound, or class of chemical compounds;

[0056] If it is determined that the air in the vicinity of the air purification device contains the desired chemical element, chemical compound, or class of chemical compounds, operation of the air purification unit is terminated.

[0057] The chemical element, chemical compound, or class of chemical compounds may be determined to be desirable based at least in part on user input indicating a user preference regarding the composition of air proximate the air purification device.

[0058] In some examples, an air purification unit may include a fan assembly that draws air into the air purification device, and a purification assembly configured to purify the air drawn into the air purification unit by the fan assembly.

[0059] In some examples, air purification devices may be used in buildings.

[0060] In some examples, the air purification device may be wearable. The air purification device may be a wearable mask. If the air purification device is a wearable mask, sensor data representing the composition of the air near the air purification device may be obtained by appropriately positioned sensors on the mask.

[0061] In some examples, an air purification system may include a sensor unit configured to provide sensor data representative of a composition of air proximate to the air purification device.

[0062] In some examples, the air purification device may include a sensor unit.

[0063] In some examples, the air purification system may include a sensor unit separate from the air purification device. The independent sensor unit may be configured to transmit sensor data to the air purification device via a communication interface. The independent sensor unit may be, for example, a separate sensor unit, or may be incorporated into another device, such as a wall clock or a smart watch.

[0064] In some examples, the sensor unit may include an electronic nose sensor.

[0065] In some examples, the sensor unit may include a VOC (volatile organic compound) sensor or a VIC (volatile inorganic compound) sensor.

[0066] In some examples, the sensor unit may include a particle counter.

[0067] In some examples, the sensor unit may include a temperature sensor.

[0068] In some examples, the sensor unit may include a humidity sensor.

[0069] In some examples, the sensor unit may include a pressure sensor.

[0070] In a second aspect of the present invention, there may be provided an air purification device suitable for (eg configured for) an air purification system according to the first aspect of the present invention.

[0071] The present invention includes any combination of the described aspects and preferred features unless such a combination is expressly impermissible or explicitly avoided. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Embodiments and experiments illustrating the principles of the present invention will now be discussed with reference to the accompanying drawings, in which:

[0073] Figure 1 An example air purification system is shown.

[0074] Figure 2A-2C An example air purification device for use in the air purification system of FIG. 2 is shown.

[0075] Figure 3-Figure 5 Shows Figure 1 Different examples of how the air purification device 110 of the air purification system 100 can use a control algorithm that includes a machine learning model. DETAILED DESCRIPTION

[0076] Aspects and embodiments of the present invention will now be discussed with reference to the accompanying drawings. Other aspects and embodiments will be apparent to those skilled in the art. All documents mentioned herein are incorporated herein by reference.

[0077] Figure 1 An example air purification system 100 is shown.

[0078] The air purification system 100 includes an air purification device 110 and a network server 180 .

[0079] The air purification device 110 includes an air purification unit 120 configured to operate to purify air near the air purification device 110 and a communication interface 112 , wherein the air purification device 110 is configured to communicate with the network server 180 via the communication interface 112 .

[0080] The communication interface 112 may be, for example, a wireless network interface, for example configured to communicate with the Internet via one or more known wireless network protocols.

[0081] The air purification unit 120 includes a fan assembly 122 that draws air into the air purification unit 120 , and a purification assembly 124 configured to purify the air drawn into the air purification unit 120 by the fan assembly 122 .

[0082] The air purification device 110 also includes a processing system 140 , which may include, for example, a memory 142 and one or more processing units 144 .

[0083] In this example, the air purification device 110 includes a sensor unit 150 configured to provide sensor data representative of a composition of air in the vicinity of the air purification device 110 .

[0084] In other examples, the sensor unit 150 may be separate from the air purification device 110. The independent sensor unit may be configured to transmit sensor data to the air purification device 110 via the communication interface 112. The independent sensor unit may be, for example, a stand-alone sensor unit, or may be incorporated into another device, such as a wall clock, a mobile device, or a smart watch.

[0085] In this example, sensor unit 150 includes an electronic nose ("e-nose") sensor 152 for generating sensor data representative of the chemical content of air near the air purification device. The electronic nose can be used to detect chemical compounds of interest (including but not necessarily limited to chemical compounds that can be smelled by a person's nose) and can be used as an electronic substitute for a person's nose.

[0086] An electronic nose sensor typically includes multiple (eg, 5-64) individual sensors (pixels) that are used to generate sensor data representative of the chemical content of the air in the vicinity of the electronic nose (and therefore, in the vicinity of the air purification device if the electronic nose is in the vicinity of the air purification device).

[0087] The electronic nose sensor may generate sensor data in the form of raw signal data acquired from individual sensors in the electronic nose, or in the form of data derived from the raw signal data. If a particular chemical compound is present in the air near the electronic nose sensor (and therefore near the air purification device), the sensor data generated by the electronic nose at that time may include a signature indicative of the chemical compound.

[0088] Typical sensor types for electronic noses can include electrical sensors (e.g., chemiresistor sensors that measure changes in resistance or impedance), mechanical sensors (e.g., quartz crystal microbalances that measure changes in resonant frequency), or optical sensors (measuring changes in fluorescence, photoconductivity, or color of a chemically responsive component).

[0089] Individual sensors within the electronic nose sensor 152 can be selective so that the sensor array responds to different chemicals or groups of chemicals. Each individual sensor can be configured to be sensitive to a single compound (e.g., formaldehyde) or a wide range of compounds (e.g., volatile organic compounds (VOCs)). In particular, the electronic nose sensor 152 can include a selective membrane (e.g., a molecular size / shape selective membrane) covering one or more sensors to limit which molecules can hit each sensor. Additionally or alternatively, the sensors themselves can be individually tuned so that each sensor responds only to certain molecules, chemicals, or groups thereof.

[0090] In some examples, the electronic nose sensor 152 can be a silicon photonics-based electronic nose sensor having a Mach-Zehnder interferometer array. An example of such an electronic nose sensor is available from Aryballe (https: / / aryballe.com / our-products / device-solutions / ).

[0091] In other examples, the electronic nose sensor 152 may be a chemiresistor array. An example of such an electronic nose sensor is SafeScent from Stratuscent (https: / / www.stratuscent.com / safescent).

[0092] Of course, as mentioned above, other types of electronic nose sensors are also available.

[0093] The sensor unit 150 may include one or more other sensors, such as a particle counter, a temperature sensor, a humidity sensor, a pressure sensor.

[0094] In this example, the air purification device 110 includes a user interface 160 configured to allow a user to provide user input data indicating a user preference regarding the composition of air in the vicinity of the air purification device.

[0095] In this example, the user interface 160 includes a preference button 162 that includes a “thumbs up” button (see Figure 2A element 262A) and the "thumbs down" button (see Figure 2A element 262B).

[0096] The "thumbs-up" button is configured to allow a user to provide a "positive opinion" input by pressing the "thumbs-up" button. The "positive opinion" input indicates that the user has a positive opinion of the smell of the air near the device when the "thumbs-up" button is pressed. The user input data may include data associating the user's positive opinion with sensor data representing the composition of the air near the air purification device when the "positive opinion" input is provided.

[0097] The "thumbs down" button is configured to allow a user to provide a "negative opinion" input by pressing the "thumbs down" button. The "negative opinion" input indicates that the user has a negative opinion of the smell of the air near the device when the "thumbs down" button is pressed. The user input data may include data associating the user's negative opinion with sensor data representing the composition of the air near the air purification device when the "positive opinion" input is provided.

[0098] The user interface 160 may include additional control buttons 164 for initiating other operations to be performed by the air purification device (eg, turning the power on / off or turning a wireless connection on / off).

[0099] In other examples (not shown), the user interface may be provided by a separate device (eg, a mobile device such as a mobile phone) connected to the air purification device 110 via the communication interface 112 of the air purification device 110 .

[0100] The network server 180 includes a communication interface 182 and a processing system 190 .

[0101] The communication interface 182 may be, for example, a wired network interface, eg, configured to communicate with the Internet via one or more known wired network protocols.

[0102] The processing system 190 of the network server 180 may include, for example, a memory 192 and one or more processing units 194 .

[0103] although Figure 1It is shown that the air purification device 110 and the network server 180 communicate via the Internet, but the air purification device 110 and the network server 180 can also communicate directly, that is, without involving the Internet.

[0104] Air purification system 100 may include one or more additional air purification devices 110A, 110B . . . 110N, each of which may include components corresponding to air purification device 110 .

[0105] Figure 2A-2B Shown in Figure 1 An example air purification device 210 for use in an air purification system 100 (eg, as air purification device 110 ).

[0106] The air purification device 210 includes a main body 214, at least one removable purification assembly (not shown) installed in the main body, and a nozzle 218 installed on the vent / opening, the main body 214 includes an air inlet 216, the main airflow enters the main body 214 through the air inlet 216, and the main airflow leaves the main body 214 through the vent / opening. The main fan assembly (not shown) is arranged in the main body 214 to draw air from the air inlet 216 to the nozzle 218 through the purification assembly. The nozzle 218 includes air outlets 218a, 218b for emitting the main airflow from the air purification device 210. The nozzle 218 also defines a center / inner opening / hole 219, and the air from the outside of the air purification device 210 is sucked in by the main airflow emitted from the outlets 218a, 218b through the opening / hole. Therefore, the nozzle 218 forms a ring extending around the hole 219.

[0107] The purification assembly includes one or more filter elements (e.g., filters) that are used to reduce the concentration of one or more compounds and / or other substances (e.g., particulate matter) when the main fan assembly is operating. In particular, the filter element includes one or more filter media to filter the air sucked into the air purification device 210 by the main fan assembly.

[0108] The main fan assembly and the purification assembly together form an air purification system, which is configured to be operated to purify the air near the air purification device. In this example, the air purification system operates by operating the main fan assembly to draw air from the air inlet 216 to the nozzle 218 through the purification assembly.

[0109] Features of the air cleaning device 210 relating to fans and air cleaning are described in more detail in GB2568979A and need not be repeated here.

[0110] Figure 2C Shows Figure 2A-2B The sensor air flow channel 226 within the air purification device 210.

[0111] The sensor air flow passage 226 extends from the sensor air flow inlet 219a to the sensor air flow outlet 219b.

[0112] The sensor airflow channel 226 includes a secondary fan assembly 227, which is configured to be operated to cause a secondary airflow to flow from the sensor airflow inlet 219a through the sensor airflow channel to the sensor airflow outlet 219b. The secondary fan assembly 227 is configured to operate independently of the primary airflow described above, which allows the secondary airflow to flow even when the primary fan assembly is not operating.

[0113] In this example, Figure 2A-2B The air purification device 210 includes an electronic nose sensor 252 configured to provide sensor data representative of a composition of air in the vicinity of the air purification device.

[0114] The electronic nose sensor 252 may be as described above with respect to Figure 1 discussed.

[0115] The airflow in the sensor airflow channel 226 is preferably ambient air that has not been filtered / heated / cooled by the air purification device. To achieve this, the sensor airflow inlet 219a and the sensor airflow outlet 219b can be appropriately located on the outer surface of the body of the air purification device 210 so that the purified / heated / cooled air from the main airflow does not pass over the sensor airflow inlet 219a and the sensor airflow outlet 219b. Figure 2B In the example shown in FIG. 2 , the sensor air flow inlet 219 a and the sensor air flow outlet 219 b are located on one side of the main body of the air purification device 210 to achieve this effect.

[0116] In some examples, the electronic nose sensor 252 may be the only sensor in the air purification device 210, in which case the electronic nose sensor 252 may (alone) form a sensor unit of the air purification device 210. In other examples, one or more other sensors may be incorporated into the air purification device 210, with the electronic nose sensor and other sensors together forming a sensor unit of the air purification device 210. One or more other sensors may be included in the sensor airflow channel 226, such as Figure 2C as shown, or in an additional sensor airflow channel ( Figure 2A-2C not shown).

[0117] Different from the air purification equipment described in GB2568979A, Figure 2A-2BThe air purification device 210 shown in includes a preference button in the form of a thumb up button 262A and a thumb down button 262B. As discussed in more detail above, these preference buttons allow a user to provide user input data indicating user preferences regarding the composition of the air near the air purification device.

[0118] The air purification device 210 also includes the communication interface ( Figure 2A-2C ) and an on / off power button.

[0119] Now return to reference Figure 1 The control algorithm according to which the air purification device 110 operates may include a machine learning model. The network server 180 may be configured to update the control algorithm by updating the machine learning model using training data.

[0120] Figure 3-Figure 5 Shows Figure 1 Different examples of how the air purification device 110 of the air purification system 100 can use a control algorithm that includes a machine learning model.

[0121] exist Figure 3 In the example shown, the machine learning model 144 used as part of the control algorithm is located at the air purification device 110, and the network server 180 is configured to update the control algorithm by updating the machine learning model 144 located at the air purification device to change how the air purification device 110 operates the air purification unit 120.

[0122] The machine learning model 144 used here is a model configured to determine whether to operate the air purification unit 120 of the air purification device 110 based on sensor data representing the composition of the air near the air purification device. The model is located in the processing system 140 of the air purification device 110 and stored in the memory 142.

[0123] In this example, the processing system 140 of the air purification device is configured to send a control signal to instruct the air purification unit 120 to operate (or not operate) based on the output of the machine learning model 144.

[0124] The network server 180 is configured to update the control algorithm by updating the machine learning model 144 using the training data.

[0125] In this example, the network server 180 is configured to update the machine learning model 144 located at the air purification device 110 by generating an updated version of the machine learning model 184 at the server 180, and then transmitting the updated version of the machine learning model 184 to the processing system of the air purification device, in which the machine learning model 184 is stored in the memory 142 as a new version of the machine learning model 144 to replace the previous version of the machine learning model 144.

[0126] In some examples, a new version of the machine learning model 144 can be added to the air purification device via a separate communication channel used to transmit sensor data to a network server. For example, a new version of the machine learning model can be made downloadable to a removable storage device, which can then be used to transfer the new version of the machine learning model to the air purification device. This can help reduce the reliance of the air purification device operation on the reliability of the network connection.

[0127] The network server 180 is configured to generate an updated machine learning model 184 using the training data.

[0128] In detail, the training database 186 at the network server 180 stores training data for training the machine learning model 184. In this example, the training data includes:

[0129] (i) User input data and sensor data associated with the user input data transmitted from the air purification device 110 .

[0130] (ii) aggregated sensor data associated with user input data of multiple users. Thus, the aggregated sensor data may include data from other air purification devices (e.g. Figure 1 In some examples, the aggregated sensor data may include only one type of "positive opinion" input (corresponding to a "negative opinion" input). Figure 2A Thumbs up button 262A shown) and one type of "negative opinion" input (corresponding to Figure 2B 262B). However, in other examples, more than one type of "positive opinion" input or "negative opinion" input (e.g., different levels of positive and / or negative opinions) can be used. For example, a five-point scale of "very negative", "negative", "neutral", "positive", and "very positive" can be provided.

[0131] (iii) Original Equipment Manufacturer ("OEM") data from the manufacturer of the air purification device 110, 110A-110N. The OEM data may include sensor data associated with a chemical element, chemical compound, or class of chemical compounds that has been identified as being harmful to a user.

[0132] The training data is used to train or retrain the machine learning model 184 in the "Model Training" functional block so that:

[0133] (A) the machine learning model 184 is trained to determine that the air purification unit should be turned on when sensor data associated with a "negative opinion" from a user of the air purification device 110 or one of a plurality of users associated with the aggregated data is provided as input to the machine learning model; and / or

[0134] (B) the machine learning model 184 is trained to determine that the air purification unit should be turned on when sensor data related to a chemical element, chemical compound, or class of chemical compounds that has been identified in the OEM data as being harmful to the user is provided as input to the machine learning model; and / or

[0135] (C) The machine learning model is trained to determine that the air purification unit should be shut down when sensor data associated with a “positive opinion” from a user of the air purification device 110 or one of multiple users associated with the aggregated data is provided as input to the machine learning model.

[0136] Based on the disclosure herein, training a machine learning model in this manner is well within the capabilities of the technician, such as using known machine learning training techniques and libraries. For example, the machine learning model 144 (and model 184) can be appropriately selected as a decision tree algorithm, which can be used as a data point together with the individual static values ​​of each sensor element in the electronic nose under different circumstances. The above training can be posed as a nonlinear step-based problem, in which a decision can be made through a series of conditions, and these conditions are learned using underlying real data. A decision tree is the simplest algorithm that can perform conditional-based learning, and it may require less computation than other nonlinear methods. The machine learning model can be built using a standard machine learning library, such as the scikit learn python library, which can be licensed under a permissive simplified BSD license, which allows it to be used in commercial applications.

[0137] After (re)training, the (re)trained model is then tested to see if it has the desired level of accuracy, see functional block 188. If the (re)trained model has the desired level of accuracy, the weights of the machine learning model 184 are updated accordingly to generate an updated machine learning model 184. This updated version of the machine learning model 184 is then transmitted to the air purification device, as described above.

[0138] If the (re)trained model lacks the desired level of accuracy, the model needs to be further trained. This retraining can optionally include obtaining more data in the database, which can be used as part of the retraining.

[0139] Note here that the machine learning model does not need to identify the chemical or smell based on the sensor data. Instead, the machine learning model has been trained to operate the air purification unit (or not) based on the sensor data and how to use the training data to train it. In other words, the machine learning model can determine whether a chemical or smell is desired or undesirable without directly and / or specifically identifying the chemical or smell.

[0140] In some examples, the training database may include metadata about the air purification device 110, such as the device model, software version, IP address location. For example, this metadata may affect how sensor data from the device is interpreted. The IP address location may indicate the location of the air purification device, which may be used to aggregate data from geographically similar users, as discussed further below.

[0141] Preferably, the plurality of users (with which the aggregated sensor data is associated) belong to a limited population that shares at least one characteristic with the user of the air purification device.

[0142] For example, members of a population may share geographic location characteristics, age characteristics, gender characteristics, and / or user preference characteristics.

[0143] Geolocation characteristics (e.g., country) may be provided by the user during the registration process of their air purification device, or via the IP address of the air purification device (since the geolocation can be derived from the IP address, although this may not always be accurate, for example because the user may use a VPN).

[0144] In this way, sensor data from multiple users may be used to improve the performance of an air purification device in a manner that reflects the user preferences (by sharing at least one characteristic) without having to rely on the users of the air purification device 110 to provide any input regarding their preferences.

[0145] In some examples, members of a crowd may share user preference characteristics regarding the composition of the air near an air purification device. For example, members of a crowd may all have similar preferences regarding smells they like and / or don't like (not necessarily belonging to the same geographic region, age group, or gender).

[0146] The air purification system may be configured to categorize users of the air purification devices into populations (that share user preference characteristics regarding the composition of the air near the air purification devices) based on user input responsive to the composition of the air near the air purification devices.

[0147] The air purification system can be configured to use a machine learning model to form crowds and classify users into crowds. The machine learning model used to form crowds based on shared user preference features can be a clustering algorithm.

[0148] Therefore, machine learning can be used to form a group of people who share user preference characteristics with the users of the air purification device, and then the group of people is used to train the machine learning model used by the air purification device. This helps to speed up the training of the machine learning model based on the preferences of the users of the air purification device without having to wait for the users of the air purification device to provide a statistically significant amount of training data to the network server.

[0149] exist Figure 3 In the example (different from the one discussed below) Figure 4 ), even if the wireless communication interface of the air purification device cannot communicate with the wireless communication interface of the server, the air purification device can continue to operate. Figure 5 Unlike the example above, the processing required to (re)train the machine learning model is offloaded to a network server.

[0150] exist Figure 4 In the example of , the machine learning model 144 used as part of the control algorithm is located at a network server, and the network server 180 is configured to instruct the air purification unit to operate according to the output of the machine learning model 144.

[0151] Here the steps associated with training a machine learning model are Figure 3 The examples are very similar. Figure 4 In the example, the machine learning model 144 stored in the network server receives sensor data from the sensor unit of the air purification device, determines whether to operate the air purification unit 120, and then sends a control signal to the air purification unit based on the output of the machine learning model stored in the network server to control the operation of the air purification unit.

[0152] exist Figure 5In the example shown, a machine learning model used as part of a control algorithm is located at the air purification device, and the network server is configured to update the control algorithm by sending training data to the air purification device to change how the air purification device operates the air purification unit, so that the air purification device updates the machine learning model using the training data. In this example, the training work is done with Figure 3 The same example.

[0153] and Figure 3 Compared with the layout of Figure 5 More processing is required in the arrangement, and the OEM and the collected data need to be sent directly to the air purification device. Figure 3 The example strikes a good balance between being able to respond to network issues and minimizing the processing required for the air purification equipment.

[0154] for Figure 3 and Figure 4 For example, each air purification device 110, 110A-110N may have its own machine learning model 184 stored on a network server, with each purification device 110, 110A-110N being trained differently based on one or more characteristics of the user of the purification device 110, 110A-110N. This does not apply to Figure 5 An example where each air purification device 110, 110A-110N will have its own machine learning model.

[0155] In some embodiments, more than one type of air purification system may be used based on the specific air purification device. Figure 3-Figure 5 For example, air purification equipment with lower processing capacity can use Figure 3 or preferred Figure 4 The arrangement of air purification equipment with higher processing capacity can be used Figure 5 In the case of poor connection quality between the server and the air purification device, the air purification device can be configured not to use Figure 4 In this way, each air purification device of the air purification system can utilize the most suitable operation mode.

[0156] ***

[0157] The features disclosed in the preceding description, or in the following claims, or in the accompanying drawings, expressed in their specific form or as devices for performing the disclosed functions, or as methods or processes for obtaining the disclosed results, may, where appropriate, be used alone or in any combination of these features to implement the invention in its different forms.

[0158] Although the present invention has been described in conjunction with the above exemplary embodiments, many equivalent modifications and variations will be apparent to those skilled in the art when this disclosure is given. Therefore, the above exemplary embodiments of the present invention are considered to be illustrative rather than restrictive. Various changes may be made to the described embodiments without departing from the spirit and scope of the present invention.

[0159] For the avoidance of any doubt, any theoretical explanations provided herein are intended to enhance the reader's understanding. The inventors do not wish to be bound by any of these theoretical explanations.

[0160] Any section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described.

[0161] Throughout the specification, including the following claims, unless the context requires otherwise, the words "comprise" and "comprising", and variations such as "comprises", "including" and "comprising", will be understood to imply the inclusion of stated integers or steps or integers or groups of steps but not the exclusion of any other integers or steps or integers or groups of steps.

[0162] It must be noted that, as used in the specification and the appended claims, the singular forms "a", "an", and "the" include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from "about" one particular value, and / or to "about" another particular value. When such a range is expressed, another embodiment includes from the one particular value and / or to the other particular value. Similarly, when a value is expressed as an approximation by using the antecedent "about," it will be understood that the particular value forms another embodiment. The term "about" in relation to a numerical value is optional and means, for example, + / - 10%.

[0163] References

[0164] Many publications are cited above to more fully describe and disclose the present invention and the current state of the art to which the present invention belongs. The complete citations of these references are as follows. The entire contents of each of these references are hereby introduced.

[0165] WO2019 / 074256

[0166] GB2568979A

Claims

1. An air purification system, comprising: Air purification equipment; and Network servers; Wherein, the air purification equipment comprises: an air purification unit configured to be operated to purify air in the vicinity of the air purification device; and a communication interface, wherein the air purification device is configured to communicate with the network server via the communication interface; wherein the air cleaning device is configured to operate the air cleaning unit according to a control algorithm that uses as input sensor data representative of the composition of air in the vicinity of the air cleaning device; Wherein the network server is configured to update the control algorithm to change how the air purification device operates the air purification unit based on sensor data representing the composition of air near the air purification device.

2. The air purification system of claim 1, wherein the network server is configured to update the control algorithm based on sensor data associated with user input data indicating a user preference regarding the composition of the air near the air purification device.

3. The air purification system of claim 2, wherein the sensor data used to update the control algorithm comprises aggregated sensor data associated with a plurality of users, wherein the plurality of users includes at least one user other than a user of the air purification device.

4. The air purification system of claim 3, wherein the plurality of users belong to a limited group of people that share at least one characteristic with the user of the air purification device.

5. The air purification system of claim 4, wherein members of the group share a geographic location characteristic.

6. An air purification system according to claim 4 or 5, wherein members of the group share age characteristics.

7. An air purification system according to any one of claims 4 to 6, wherein members of the group share gender characteristics.

8. An air purification system according to any one of claims 4 to 7, wherein members of the group share user preference characteristics regarding the composition of the air in the vicinity of the air purification device.

9. The air purification system of claim 8, wherein the air purification system is configured to classify users of the air purification device into the population based on user input responsive to a composition of air near the air purification device.

10. An air purification system according to claim 8 or 9, wherein the air purification system is configured to use a machine learning model to form the population and classify the user into the population.

11. An air purification system according to any one of the preceding claims, wherein: The control algorithm includes a machine learning model; The network server is configured to update the control algorithm by updating the machine learning model using training data.

12. An air purification system according to any one of the preceding claims, wherein: The air purification unit includes a fan assembly that draws air into the air purification device, and a purification assembly configured to purify the air drawn into the air purification unit by the fan.

13. An air purification system according to any one of the preceding claims, wherein: The air purification device is used in a building.

14. An air purification system according to any one of the preceding claims, wherein: The air purification system comprises a sensor unit configured to provide sensor data representative of a composition of air in the vicinity of the air purification device.

15. An air purification device configured for use in an air purification system according to any one of the preceding claims.

Images