Air purifiers

An air purifier with multiple sensors and a processor detects sneezes or coughs to enhance its cleaning rate, efficiently removing airborne droplets and reducing infection risk by automatically increasing fan speed.

JP2024542479APending Publication Date: 2024-11-15ブルーエア·アーベー
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
JP2024529610
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2021-11-17
Filing Date
2022-11-14
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

Existing air purifiers lack the ability to automatically respond to sudden increases in droplet contamination, such as those caused by a person sneezing or coughing in the same room, necessitating a more responsive and accurate detection system.

Method used

An air purifier equipped with multiple sensors, including acoustic, carbon dioxide, particulate, and humidity sensors, processes inputs to detect sneezes or coughs and adjusts its operation accordingly, using a processor to increase cleaning rate and fan speed to efficiently remove medium-sized droplets.

Benefits of technology

The system effectively removes medium-sized droplets within minutes of a sneeze or cough, reducing the risk of infection by activating the air purifier promptly and accurately distinguishing between sneezes and background noise.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024542479000002
    Figure 2024542479000002
  • Figure 2024542479000001
    Figure 2024542479000001
  • Figure 2024542479000003
    Figure 2024542479000003
Patent Text Reader

Abstract

1. An air purifier comprising a fan, a motor, an air inlet and an air outlet, a removable particulate or gas filter, an acoustic sensor, a further sensor, means for transmitting information from the acoustic sensor and the further sensor to a processor for processing input from the acoustic sensor and the further sensor, and means for controlling the fan based on calculations by the processor, wherein the further sensor is selected from a carbon dioxide sensor, a particulate sensor, a humidity sensor, and a volatile organic carbon sensor.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an improved air cleaner with automated features based on situational awareness. [Background technology]

[0002] Patent Document 1 (Panasonic Corporation) discloses an air purifier that effectively prevents the spread of viruses and the like by checking the occurrence and location of coughs and sneezes in two dimensions, and performs intensive air purification at the most polluted location. The air purifier includes a main body provided with at least three microphones, and is configured to detect the sound of a cough or sneeze among the sounds collected by the microphones, and controls a blower fan, a longitudinal louver, a horizontal louver, and an ion generator according to the sound information. This structure enables the air purifier to check the occurrence and location of coughs and sneezes in a room in two dimensions, and perform intensive air purification at the most polluted location, thus achieving rapid prevention of the spread of viruses and the like.

[0003] US Patent No. 5,399,633 (Koninklijke Philips) discloses an air treatment system that combines a conventional air treatment device (such as an air purifier, humidifier, or dehumidifier) ​​with an ultrasonic device for modifying the air flow generated by the air treatment device, thereby enabling reduced noise and / or power consumption to achieve the desired air treatment function. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2011 / 174624 A [Patent Document 2] European Patent Application Publication No. 3 578 886 Summary of the Invention [Problem to be solved by the invention]

[0005] Notwithstanding the prior art, there remains a need for an air purifier that can operate automatically in response to the surrounding environment, in particular, an air purifier that can respond to sudden increases in water droplet contamination of the surrounding air, for example, when someone in the same room as the purifier sneezes or coughs. [Means for solving the problem]

[0006] Thus, in a first aspect, there is provided an air purifier comprising a fan, a motor, an air inlet and an air outlet, a removable particulate or gas filter, an acoustic sensor, a further sensor, means for transmitting information from the acoustic sensor and the further sensor to a processor for processing input from the acoustic sensor and the further sensor, and means for controlling the fan based on calculations by the processor, wherein the further sensor is selected from a carbon dioxide sensor, a particulate sensor, a humidity sensor, and a volatile organic carbon sensor.

[0007] Surprisingly, it has been found that droplets generated within a domestic environment can be efficiently removed if a nearby air purifier is activated as soon as possible after a cough or sneeze.

[0008] When a person coughs or sneezes, a wide variety of droplet sizes are expelled. Generally, there are three droplet size categories: large droplets (>20 microns) fall quickly to the floor and do not travel far from the person coughing or sneezing; medium-sized droplets (3-20 microns) remain airborne for a short period of time, and smaller droplets (0.6 microns to 3 microns) generally remain airborne indefinitely.

[0009] Small droplets will be removed periodically by any air cleaning process, but medium-sized droplets will only be removed if the air is sufficiently agitated and passes through the air cleaner within 2 minutes of the droplet's emission.

[0010] The accuracy of detection is dramatically improved by the use of an additional sensor in addition to the acoustic sensor. In particular, where the assessment of the sound produced through a noise incident is ambiguous as to whether it was a sneeze or not, processing the data from the additional sensor increases the likelihood that a sneeze will be accurately detected. By this accuracy, the correct identification of a noise incident that is not a sneeze will also be accurately assessed. A purifier that identifies any noise as a sneeze and triggers itself accordingly will likely be shut off.

[0011] The air purifier preferably includes a humidity sensor, which may be part of the air purifier or may be part of a separate electronic device.

[0012] The air cleaner preferably includes a particulate sensor, which may be part of the air cleaner or may be part of a separate electronic device.

[0013] The air purifier preferably includes a carbon dioxide sensor, which may be part of the air purifier or may be part of a separate electronic device.

[0014] The air purifier preferably includes a volatile organic carbon sensor, which may be part of the air purifier or may be part of a separate electronic device.

[0015] In order for the air purifier to modify its performance in case of a sneeze or cough, it must detect sounds in the surrounding environment. Such an acoustic sensor is preferably a microphone or similar device, preferably calibrated to detect sounds in the surrounding environment, for example in the same room. It is also conceivable that the sensor could be part of a separate mobile device and thus transmit information to the processor.

[0016] The purifier will also include a means for transmitting information from the acoustic sensor to the processor. This may be wired or wireless, depending on the location of the processor. In one embodiment, the processor may be located with the air purifier and wired to the acoustic sensor. However, it is also possible that the processor is remote, for example part of a mobile electronic device such as a phone or tablet, or even completely separate and wirelessly connected. It is contemplated that the sensor and the processor may be part of a separate mobile device, such that the only modification to the air purifier is the ability to be operated as a result of calculations and determinations made separately, for example in the mobile device.

[0017] The processor processes the information from the acoustic sensor and determines whether an increase in the air cleaning speed is necessary. The processor accesses a database of sounds to determine whether a sound occurred in the environment or within the purifier and whether it correlates with a cough or sneeze. If the processor determines that such an event has occurred, the processor increases the air cleaning speed by accelerating the fan.

[0018] This occurs by accelerating the fan, for example the fan speed can be preferably increased to a predetermined speed consistent with cleaning the air in the room within a set time period, in such a way that the medium sized droplets are quickly removed from the surrounding air, reducing the possibility of infecting another person.

[0019] The speed at which a purifier operates is determined by the Clean Air Delivery Rate (CADR) required to purify a room of medium-sized droplets based on the purifier performance. For example, 700 m 3 / h CADR purifiers can remove 30m3 of water in less than 15 minutes. 3 can remove >99% of airborne microorganisms from a room.

[0020] If the purifier determines that the air is again clean, the air purifier returns to the operating performance parameters prior to the sneeze or cough event. Alternatively, the purifier remains at the appropriate performance setting depending on the environment and programming by the consumer. For example, if the consumer is concerned about allergies or the flu, the purifier may be programmed to run at an increased CADR to achieve improved baseline effectiveness.

[0021] It is also preferred that the processor is able to distinguish between a cough or sneeze event and background noise so that accuracy of detection is achieved, and more preferably the processor is able to calibrate itself based on the background noise caused by the air purifier in operation. In either case, such information will be received by the sensor, but additional information, such as fan speed or basic operating noise, may also be transmitted by other sensors if desired.

[0022] If the processor is remote from the device, the air purifier includes means for receiving information from the remote processor and acting based on the determination, e.g., increasing fan speed if a sneeze is identified, In such an example, the processor determines if a sneeze has occurred and electronically communicates with the fan to increase its speed.

[0023] If the processor is part of an air cleaner, the processor is in electronic communication with the fan.

[0024] In a preferred embodiment, there is provided a combination of a processor as described above and an air purifier, where the processor is remote from the purifier, for example the processor is part of a remote electronic device such as a mobile phone.

[0025] An example of an acoustic sensor is the C-800G available from Sony Corporation.

[0026] Preferably, the removable filter media can be a particle filter or a gas filter, although it is also possible that a particle filter and a gas filter are employed simultaneously.

[0027] The airflow velocity measured at the removable filter is known in the art as the media velocity. Media velocity is the speed at which air moves through the filter. The media velocity must be perfectly controlled to ensure the maximum amount of particles are captured. Too fast and many of the contaminants will fly straight through unfiltered. Too slow and the purifier will not reach the farthest corners of your room fast enough to be effective.

[0028] The air cleaner preferably includes an ionizer.

[0029] The ionizer used comprises an emitter electrode and a ground electrode, which upon application of a suitable voltage, preferably around the range of -10 kV to 10 kV, more preferably -8 kV to 8 kV, ejects an ion cloud between the emitter and the receiver.

[0030] The emitter electrode is preferably a point, a tip, or a number of tips or points, e.g., a brush, and is in electrical communication with a power source. The emitter electrode is preferably within 20 cm, more preferably 15 cm, particularly preferably 10 cm, most preferably up to 5 cm, of the nearest filter medium, and is preferably sterilized, if necessary, by placing it in the ion cloud during start-up of the ionizer. This distance is the distance from the tip of the emitter electrode to the nearest part of the nearest filter medium, more specifically the part of the filter medium to be filtered, not, for example, the structural frame.

[0031] The receiving electrode is preferably a metal part in the shape of a ring, so that the resulting ion cloud is three-dimensional, as the ions are expended away from the emitter and then attracted towards the receiver. The receiving electrode is preferably in the form of a cage extending away from the emitter electrode in the airflow direction. More preferably, the receiving electrode is in the form of a mesh device extending towards the filter. Such an extension may present a hemispherical or partially cylindrical shape, so that the receiving electrode is downstream in the airflow direction from the emitter electrode.

[0032] The emitter is preferably centrally disposed between the rings of receivers. The emitter may be disposed towards or away from the airflow during use, but it is preferred that the emitter is directed towards the surface to be disinfected, e.g. an interior wall or filter media.

[0033] In a preferred embodiment, the further sensor is a particulate sensor.

[0034] The air cleaner preferably includes two further sensors, eg a particulate sensor and a humidity sensor.

[0035] The air cleaner preferably includes three further sensors, eg, a particulate sensor, a humidity sensor, and a carbon dioxide sensor.

[0036] In a second aspect, there is provided a method of operating an air purifier by detecting a sound, comparing the sound to sounds in a database, determining whether a determined sound type is detected, processing data from a further sensor, and operating the air purifier, wherein the further sensor is selected from a carbon dioxide sensor, a particulate sensor, a humidity sensor, and a volatile organic carbon sensor.

[0037] The processor first processes information from the acoustic sensor to determine a possible sneeze incident. The processor then processes information from additional sensors to more accurately determine whether a sneeze incident has occurred. [Brief description of the drawings]

[0038] [Figure 1] FIG. 1 is a diagram showing the process of cough / sneeze detection by a sensor. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0039] Working Example Air purifier Sneeze / Cough Detection System (SCDS) The challenges involved in devising a sensor for sneeze and cough are multi-faceted: a) the sensor needs to accurately distinguish the characteristic sound wave patterns of a sneeze / cough as opposed to speech and more importantly everyday noise. b) It is believed that the vocal characteristics of various ethnicities around the world may differ due to differences in vocal cord physiology of different tribes. Hence, it is believed that the cough / sneezing features may vary across tribes. c) The sensitivity of the sensor needs to accurately detect the acoustic intensity of sneeze / cough across all age groups from babies to elderly people. d) It is believed that the sensor needs to compare the detected voice signal as opposed to such a large and diverse repository of cough / sneezing features as defined above, which may slow down the response time of the sensor to a sneeze or cough event.

[0040] One of the techniques to overcome the above challenges is as follows: Signals from three sensors, namely, an acoustic sensor, a particulate sensor, and a humidity sensor, are assumed to be used by a sneeze / cough detection system (SCDS) to achieve a high degree of detection accuracy. Based on the signals from these three sensors, the SCDS is assumed to determine that an incoming audio signal (IAS) from the acoustic sensor is indeed a sneeze or cough event. To overcome the challenge of discrimination against various speech and environmental noises, the SCDS analyzes the IAS in three stages. 1) In the first stage, the acoustic signal frequencies are analyzed by a local processor located in the air purifier or a separate electronic device using a small database pre-stored in a local memory. The database performs a superficial validation based on frequency ranges that are characteristic of coughs / sneezes. 2) If the results of this analysis indicate enough reason to believe that the sound is a cough or sneeze, in a second step, the acoustic pattern is sent to the cloud, where a server compares the signal against a large database of stored sounds and noises. 3) Once the server has confirmed that the signal is likely a cough / sneeze, in the third stage, the SCDS also verifies the signals from the particle sensor and humidity sensor as to whether any sudden changes in aerosol concentration occurred during the sneeze / cough event in the form of changes in aerosol particle count, CO2 levels, and relative humidity in the surrounding environment.

[0041] Then, based on a specific judgment algorithm, the SCDS would identify a sneeze / cough event and would instruct the air purifier to switch to its maximum performance mode to remove the aerosol and its present bacteria.

[0042] The diverse sneeze / cough feature patterns required for the second stage can be sourced from acoustic sets available online and commercial resources. It is conceivable that these acoustic patterns can be stored in the cloud via a server. For example, Google has built an ontology of 632 acoustic event classes and 2,084,320 audio files, available as freeware (http: / / research.google.com / audioset / ). This database includes sounds from people, animals, musical instruments, and everyday sounds from the environment. Similar datasets are available from institutions and academic societies specializing in speech and acoustic recognition.

[0043] Also, additional sensor checks in the form of signals from particulate and humidity sensors are employed to build higher accuracy in sneeze / cough detection. 1) It is believed that every cough / sneeze event is accompanied by an increase in aerosol concentration in the surrounding environment. The sudden detection of particles / aerosols by the particulate sensor located in the air purifier will serve as an additional confirmation of the cough / sneeze. 2) The increase in humidity during the aerosol release due to the cough / sneeze event will be detected by the humidity sensor. This signal will serve as a second confirmation. 3) The CO2 level in the room will affirm the presence of a person in the room. This positive acknowledgement can help the IAS distinguish whether it is noise or a true cough / sneeze event.

[0044] Decision Flow Chart The flow chart in the figure shows the process of cough / sneeze detection by sensors. Five sensors in the air purifier, namely, an acoustic sensor, a particulate sensor, a CO2 sensor, a light sensor, and a humidity sensor, continuously monitor the room. When an IAS is received by the acoustic sensor, as a first stage, the signal is first analyzed by a microcontroller in the air purifier. The microcontroller analyzes the signal for its broad feature pattern and also compares it with the acoustic data stored locally in its ROM. This analysis is to broadly evaluate whether the IAS falls within the acoustic pattern of a cough / sneeze. It is believed that this can be done by verifying the wave frequencies and the specific tonal points that are characteristic of a cough / sneeze. However, since it is believed that many environmental noises may also have similar acoustic characteristics, this may not be sufficient to determine conclusively whether the IAS is that of a cough / sneeze. Therefore, in a second stage of evaluation, it is believed that the IAS is further transmitted to a cloud server by a WiFi module in the air purifier. The server would then analyze the signal and compare it to acoustic features that are part of a large acoustic database, such as that of Google Audioset. The server would then calculate the % match between the input acoustics and the sneeze / cough features from the cloud database. If the % match is greater than 65%, the server would then send the % match value back to the WiFi module of the air purifier. If the value is less than 65%, the server would ignore the IAS as a negative sneeze / cough event.

[0045] Final decision algorithm Once the electronics of the air cleaner (microcontroller) receives the % match value, the electronics will use a "decision algorithm" to conclusively determine whether the acoustic signal is a sneeze / cough event. The microcontroller will also seek signals from a variety of ambient sensors, including particulate sensors, humidity sensors, CO2 sensors, and light sensors. Depending on the exact value of the % match, and whether the particulate sensors and humidity sensors report a sudden change in their respective input signals, and whether the CO2 levels affirm the presence of a person, the microcontroller will determine whether the IAS is a sneeze / cough event. The light sensor will also help make an overall assessment of the accuracy of the inputs from the sensors to assess whether the inmate is asleep and whether the CO2 levels are commensurate with that.

[0046] [Table 1]

[0047] One possible algorithm decision tree that the SCDS could develop is as shown in the previous table. Among the various ambient sensors that aid in the decision, a sudden change in the particulate sensor input is considered with the highest level of certainty, followed by the CO2 sensor and the humidity sensor in that order. At a very low % match, from 65% to 75%, the algorithm seeks positive signals from the particulate sensor, CO2 sensor, and humidity sensor for its assertion as to whether the IAS is actually a cough / sneeze event or not. Whereas, if the % match is from 95% to 100%, a high confidence value is associated with the cough / sneeze event, and therefore, it is asserted as a cough / sneeze event regardless of whether the particulate / CO2 / humidity sensor output is positive or not. At a low to medium level of confidence % match, from 75% to 85%, positive signals from both the particulate sensor and the CO2 sensor are considered as a higher positive source of the cough / sneeze event than a sudden change in the humidity value. If the % match is between 85% and 95%, then just a positive input from the particle sensor is considered sufficient positive for a cough / sneeze event.

[0048] The above is just one such algorithm for making a decision and it is contemplated that the exact % match value, as well as values ​​associated with positive input from the humidity sensor, CO2 sensor, and particulate sensor, may vary and thus result in an entirely different algorithm.

Claims

1. 1. An air purifier comprising a processor, a fan, a motor, an air inlet and an air outlet, a removable particulate or gas filter, an acoustic sensor, a further sensor, means for transmitting information from the acoustic sensor and the further sensor to the processor for processing input from the acoustic sensor and the further sensor, and means for controlling the fan based on calculations by the processor, wherein the further sensor is selected from a carbon dioxide sensor, a particulate sensor, a humidity sensor, and a volatile organic carbon sensor.

2. 10. The air purifier of claim 1, wherein the processor compares input from the acoustic sensor to sounds in a database to identify occurrences of predetermined sounds within a detectable environment.

3. The air purifier of claim 1 or 2, wherein the processor is in a device separate from the air purifier.

4. 4. The air purifier of claim 1, wherein controlling the fan comprises increasing a fan speed.

5. 5. An air purifier according to any one of claims 1 to 4, comprising two further sensors.

6. 5. An air purifier according to any one of claims 1 to 4, comprising three further sensors.

7. 1. A method of operating an air purifier by the steps of detecting a sound, comparing the sound to sounds in a database, determining whether a determined sound type is detected, processing data from a further sensor, and operating the air purifier, wherein the further sensor is selected from a carbon dioxide sensor, a particulate sensor, a humidity sensor, and a volatile organic carbon sensor.

8. The method of claim 7 , wherein the sound is detected by an acoustic detector remote from the air cleaner.

9. A combination of an air purifier and a processor, the air purifier being as claimed in any one of claims 1 to 6, the processor being part of a mobile electronic device.

Citation Information

Patent Citations

  • An air treatment system and method

    EP3578886A1

  • Air cleaner

    JP2011174624A