Air particulate classification

1. A method comprising:

receiving air sensor data from an air sensor, wherein the air sensor data comprises a particulate vector;

generating a machine learning model output from the air sensor data using a machine learning model, wherein the machine learning model output comprises one or more of a filter classification vector and air quality data and is generated using a processor that:

generates calibrated data from filtered sensor data and unfiltered sensor data to a calibration model, of the machine learning model, to output the calibrated data,

generates the filter classification vector from the calibrated data using a filter model of the machine learning model, and

generates the air quality data from the calibrated data using an air quality model of the machine learning model; and

transmitting, by a control system, a notification to a heating, ventilation, and air conditioning (HVAC) control to adjust settings for at least one of a damper, a fan, or combinations thereof, using the machine learning model output.

2. The method of claim 1 , further comprising:

training the machine learning model to generate the machine learning model output from the air sensor data using training data and an update function of the machine learning model.

3. The method of claim 1 , further comprising:

periodically receiving, by a server application from the control system, the air sensor data and the machine learning model output.

4. The method of claim 1 , wherein the air sensor data comprises temperature data, humidity data, pressure data, particulate matter data, and molecular compound data.

5. The method of claim 1 , wherein generating the machine learning model output further comprises:

generating an aerosol classification vector as the machine learning model output using an aerosol detection model as the machine learning model.

6. The method of claim 5 , wherein the aerosol classification vector comprises a vector of prediction confidence scores for a set of aerosol identifiers.

7. The method of claim 5 , further comprising:

training the aerosol detection model to generate the aerosol classification vector from the air sensor data using training data and an update function of the aerosol detection model.

8. The method of claim 5 , further comprising:

transmitting, by an air sensor system comprising the air sensor that generated the air sensor data, the aerosol classification vector to the control system; and

transmitting, by the control system, an aerosol detection notification as the notification in response to receiving the aerosol classification vector from the air sensor system.

9. The method of claim 1 , further comprising:

receiving the filtered sensor data from a filtered sensor positioned after an air filter, wherein the air sensor is an unfiltered sensor positioned before the air filter and the air sensor data is the unfiltered sensor data.

10. The method of claim 9 , further comprising:

wherein the filter classification vector comprises a vector of prediction confidence scores for a set of filter exhaustion levels, and

wherein the air quality data comprises an air quality index value mapped to a plurality of strings describing air quality based on the air quality index value.

11. The method of claim 9 , further comprising:

training the calibration model to generate the calibrated data using training data and an update function for the calibration model;

training the filter model to generate the filter classification vector using the training data and an update function for the filter model; and

training the air quality model to generate the air quality data using the training data and an update function for the air quality model.

12. The method of claim 9 , further comprising:

transmitting, by a filtered sensor system comprising the filtered sensor, the filtered sensor data to the control system;

transmitting, by an unfiltered sensor system comprising the unfiltered sensor, the unfiltered sensor data to the control system;

transmitting, by the control system to a heating, ventilation, and air conditioning (HVAC) control, a filter quality notification as the notification responsive to the filter classification vector; and

transmitting, by the control system to the HVAC control, an air quality notification responsive to the air quality data.

13. A system comprising:

a control system comprising one or more processors and one or more memories;

a control application, executing on the one or more processors of the control system, wherein the control application:

receives air sensor data from an air sensor, wherein the air sensor data comprises a particulate vector;

generates a machine learning model output from the air sensor data using a machine learning model, wherein the machine learning model output comprises one or more of a filter classification vector and air quality data and is generated using the one or more processors of the control system to:

generate calibrated data using a processor that inputs filtered sensor data and unfiltered sensor data to a calibration model, of the machine learning model, to output the calibrated data,

generate the filter classification vector from the calibrated data using a filter model of the machine learning model, and

generate the air quality data generated from the calibrated data using an air quality model of the machine learning model; and

transmits, by the control system, a notification to a heating, ventilation, and air conditioning (HVAC) control to adjust settings for at least one of a damper, a fan, or combinations thereof, using the machine learning model output.

14. The system of claim 13 , further comprising:

a server comprising one or more processors and one or more memories; and

a server application, executing on the one or more processors of the server, wherein the server application:

trains the machine learning model to generate the machine learning model output from the air sensor data using training data and an update function of the machine learning model.

15. The system of claim 13 , further comprising:

a server comprising one or more processors and one or more memories; and

a server application, executing on the one or more processors of the server, wherein the server application:

periodically receives, by the server application from the control system, the air sensor data and the machine learning model output.

16. The system of claim 13 , wherein the air sensor data comprises temperature data, humidity data, pressure data, particulate matter data, and molecular compound data.

17. The system of claim 13 , wherein the control application further:

generates generating an aerosol classification vector as the machine learning model output using an aerosol detection model as the machine learning model.

18. The system of claim 17 , wherein the aerosol classification vector comprises a vector of prediction confidence scores for a set of aerosol identifiers.

19. The system of claim 15 , wherein the control application further:

receives the filtered sensor data from a filtered sensor positioned after an air filter, wherein the air sensor is an unfiltered sensor positioned before the air filter and the air sensor data is the unfiltered sensor data.

20. A method comprising:

training a machine learning model to generate a machine learning model output from air sensor data using training data and an update function of the machine learning model;

receiving the air sensor data from an air sensor, wherein the air sensor data comprises a particulate vector; and

generating the machine learning model output from the air sensor data using the machine learning model, wherein the machine learning model output comprises one or more of a filter classification vector and air quality data and is generated using a processor that:

generates calibrated data from filtered sensor data and unfiltered sensor data to a calibration model, of the machine learning model, to output the calibrated data,

generates the filter classification vector from the calibrated data using a filter model of the machine learning model, and

generates the air quality data from the calibrated data using an air quality model of the machine learning model; and

transmitting a notification to a heating, ventilation, and air conditioning (HVAC) control to adjust settings for at least one of a damper, a fan, or combinations thereof, using the machine learning model output.