Error correction for predictive schedules for a thermostat

Patent metadata

Field	Value
Publication number	US-2022042695-A1
Application number	US-202117451627-A
Country	US
Kind code	A1
Filing date	Oct 20, 2021
Priority date	Dec 31, 2019
Publication date	Feb 10, 2022
Grant date	—

Abstract

A heating, ventilation, and air conditioning (HVAC) control device is configured to record a plurality of actual occupancy statuses, to determine a plurality of corresponding predicted occupancy statuses, and to compare the plurality of predicted occupancy statuses to the plurality of actual occupancy statuses. The device is further configured to identify conflicting occupancy statuses based on the comparison. A conflicting occupancy status indicates a difference between an actual occupancy status and a corresponding predicted occupancy status. The device is further configured to identify timestamps corresponding with the conflicting occupancy statuses, to identify historical occupancy statuses corresponding with the identified timestamps, and to update the conflicting occupancy statuses in the predicted occupancy schedule with the historical occupancy statuses.

First claim

1 . A heating, ventilation, and air conditioning (HVAC) control device, comprising:
a memory operable to store:
a predicted occupancy schedule comprising a plurality of time entries, wherein:
each time entry corresponds with a day of a week and an hour of a day; and
each time entry is associated with a predicted occupancy status; and

an occupancy history log comprising a plurality of historical occupancy statuses, wherein each historical occupancy status is associated with a timestamp corresponding with a day and an hour of a day over a first period of time; and
a processor operably coupled to the memory, and configured to:
record a plurality of actual occupancy statuses, wherein each actual occupancy status is associated with a timestamp corresponding with a day and an hour of a day over a second period of time;
determine a plurality of predicted occupancy statuses, wherein each predicted occupancy status corresponds with a timestamp of an actual occupancy status;
compare the plurality of predicted occupancy statuses to the plurality of actual occupancy statuses;
identify one or more conflicting occupancy statuses based on the comparison, wherein a conflicting occupancy status indicates a difference between an actual occupancy status and a corresponding predicted occupancy status;
identify one or more timestamps corresponding with the one or more conflicting occupancy statuses;
identify one or more historical occupancy statuses corresponding with the identified one or more timestamps;
update the one or more conflicting occupancy statuses in the predicted occupancy schedule with the one or more historical occupancy statuses;
determine a number of conflicting occupancy statuses; compare the number of conflicting occupancy statuses to a predetermined threshold value, wherein the predetermined threshold value corresponds with a maximum number of conflicting occupancy statuses for the predicted occupancy schedule; and
determine the number of conflicting occupancy statuses exceeds the predetermined threshold value before updating the one or more conflicting occupancy statuses in the predetermined occupancy schedule.

2 . The device of claim 1 , wherein updating the one or more conflicting occupancy statuses in the predicted occupancy schedule with the one or more historical occupancy statuses comprises:
identifying a plurality of historical occupancy statuses corresponding with a timestamp for a conflicting occupancy status; determining a percentage of present statuses within the plurality of historical occupancy statuses; setting the conflicting occupancy status to a present status in response to determining that the percentage of present statuses within the plurality of historical occupancy statuses is greater than fifty percent; and setting the conflicting occupancy status to an away status in response to determining that the percentage of present statuses within the plurality of historical occupancy statuses is less than fifty percent.

3 . The device of claim 1 , wherein the processor is further configured to generate a heat map that identifies the one or more conflicting occupancy statuses overlaid with the predicted occupancy schedule.

4 . (canceled)

5 . The device of claim 1 , wherein the processor is further configured to:
obtain forecasted weather information for a location associated with a space, wherein the forecasted weather information identifies forecasted temperatures for a plurality of days; obtain historical weather information for the location associated with the space, wherein the historical weather information identifies a previous temperature for the plurality of days; compare the forecasted weather information and the historical weather information for the plurality of days; and determine that a temperature difference between the forecasted weather information and the historical weather information is within a predefined range for a majority of the plurality of days before recording the plurality of actual occupancy statuses.

6 . The device of claim 1 , wherein comparing the plurality of predicted occupancy statuses to the plurality of actual occupancy statuses comprises ignoring timestamps corresponding with a weekend day.

7 . The device of claim 1 , wherein the processor is further configured to:
determine a first set of weights for a machine learning model based on the updated predicted occupancy schedule, wherein each of the first plurality of weights is associated with a day of the week; determine a second set of weights for the machine learning model based on the updated predicted occupancy schedule, wherein each of the second plurality of weights is associated with a time of a day; and generate a machine learning model using the first set of weights and the second set of weights, wherein the machine learning model is configured to output a probability corresponding within an input that identifies a day of the week and a time of a day.

8 . A predictive schedule error correction method, comprising:
recording a plurality of actual occupancy statuses, wherein each actual occupancy status is associated with a timestamp corresponding with a day and an hour of a day over a second period of time; determining a plurality of predicted occupancy statuses, wherein each predicted occupancy status corresponds with a timestamp of an actual occupancy status; comparing the plurality of predicted occupancy statuses to the plurality of actual occupancy statuses; identifying one or more conflicting occupancy statuses based on the comparison, wherein a conflicting occupancy status indicates a difference between an actual occupancy status and a corresponding predicted occupancy status; identifying one or more timestamps corresponding with the one or more conflicting occupancy statuses; identifying one or more historical occupancy statuses corresponding with the identified one or more timestamps; updating the one or more conflicting occupancy statuses in a predicted occupancy schedule with the one or more historical occupancy statuses, wherein the predicted occupancy schedule comprises a plurality of time entries, wherein:
each time entry corresponds with a day of a week and an hour of a day; and
each time entry is associated with a predicted occupancy status;
determining a number of conflicting occupancy statuses; comparing the number of conflicting occupancy statuses to a predetermined threshold value, wherein the predetermined threshold value corresponds with a maximum number of conflicting occupancy statuses for the predicted occupancy schedule; and determining the number of conflicting occupancy statuses exceeds the predetermined threshold value before updating the one or more conflicting occupancy statuses in the predetermined occupancy schedule.

9 . The method of claim 8 , wherein updating the one or more conflicting occupancy statuses in the predicted occupancy schedule with the one or more historical occupancy statuses comprises:
identifying a plurality of historical occupancy statuses corresponding with a timestamp for a conflicting occupancy status; determining a percentage of present statuses within the plurality of historical occupancy statuses; setting the conflicting occupancy status to a present status in response to determining that the percentage of present statuses within the plurality of historical occupancy statuses is greater than fifty percent; and setting the conflicting occupancy status to an away status in response to determining that the percentage of present statuses within the plurality of historical occupancy statuses is less than fifty percent.

10 . The method of claim 8 , further comprising generating a heat map that id

Assignees

• Lennox Ind Inc

Inventors

• Venkatesh Sridhar
• Delgoshaei Payam
• Manohararaj Janathkumar

Classifications

• G05B13/048

  using a predictor · CPC title

• F24F2130/10

  Weather information or forecasts · CPC title

• F24F2140/50

  Load · CPC title

• G05B15/02

  electric · CPC title

• F24F11/64

  using pre-stored data · CPC title