FIELD: smoking accessories.
SUBSTANCE: invention relates to aerosol delivery devices, such as smoking articles which generate aerosol. Aerosol delivery device comprises a housing having a structure that enables to retain the aerosol precursor composition, at least one sensor configured to generate property measurement values during use of the aerosol delivery device, leads configured to connect the power supply to the aerosol delivery device, and an aerosol generating component or second leads for connecting the aerosol generating component to the aerosol delivery device. Aerosol generating component is configured to generate an aerosol from the aerosol precursor composition. Aerosol delivery device also comprises a control component including a processing circuit configured for switchable connection of the power supply to the load including the aerosol generation component to provide power to the aerosol generation component. Processing circuitry is configured to record data for multiple uses of the aerosol delivery device, for each use of which the data include property measurement values. Processing circuit is also configured to build a machine learning model for predicting the target variable. Construction of a machine learning model is provided using a machine learning algorithm, at least one feature selected from said properties, and a training sample obtained from measurement values of said properties. Processing circuit is further configured to deploy a machine learning model for predicting said target variable and controlling at least one functional element of the aerosol delivery device based thereon. Plurality of uses of the aerosol delivery device include corresponding puffs of the user, each of which causes air flow through at least part of the casing and is intended for inhalation of the aerosol by the user. Performing a processing circuit with the possibility of recording data for a plurality of uses includes its execution with the possibility of recording measurement values with an indication of the time and duration of the corresponding puffs of the user, and the target variable is a user profile, which depends on at least one of said properties, as well as on time and duration of corresponding puffs of the user. User profile includes information indicating the forecast period of non-use of the aerosol delivery device. Execution of the processing circuit with the possibility of controlling at least one functional element includes its implementation with the possibility of causing the aerosol delivery device to go into sleep mode during the predicted period of non-use.
EFFECT: high efficiency of using the device owing to the use of machine learning.
22 cl, 10 dwg
Title | Year | Author | Number |
---|---|---|---|
TEMPERATURE CONTROL IN AEROSOL DELIVERY DEVICE | 2019 |
|
RU2827795C2 |
POWER CONTROL FOR AEROSOL DELIVERY DEVICE | 2019 |
|
RU2816312C2 |
DIAGRAM OF BUCK-BOOST REGULATOR FOR AEROSOL DELIVERY DEVICE | 2020 |
|
RU2832066C2 |
CHARGING CONTROL FOR AEROSOL DELIVERY DEVICE | 2019 |
|
RU2812684C2 |
AUXILIARY BATTERY FOR AEROSOL DELIVERY DEVICE | 2018 |
|
RU2768296C2 |
TEMPERATURE CONTROL SCHEME FOR AEROSOL DELIVERY DEVICE | 2020 |
|
RU2828683C2 |
RECHARGEABLE LITHIUM-ION CAPACITOR FOR AEROSOL DELIVERY DEVICE | 2017 |
|
RU2753552C2 |
AEROSOL DELIVERY DEVICE AND CONTROL CASE CONNECTED OR MADE WITH POSSIBILITY OF CONNECTION TO CARTRIDGE TO FORM AEROSOL DELIVERY DEVICE | 2018 |
|
RU2798956C2 |
RECHARGEABLE LITHIUM-ION BATTERY FOR AEROSOL DELIVERY DEVICE | 2017 |
|
RU2753553C2 |
PHOTO SENSOR FOR MEASURING THE AEROSOL PRECURSOR COMPOSITION IN THE AEROSOL DELIVERY DEVICE | 2017 |
|
RU2753087C2 |
Authors
Dates
2025-03-04—Published
2020-04-22—Filed