US 12,147,224 B2
Systems and methods for artificial intelligence-based maintenance of an air conditioning system
Andrea Casa, Borgarello (IT)
Assigned to Alisea S.r.l., Mezzanino (IT)
Filed by Alisea S.r.l., Mezzanino (IT)
Filed on Oct. 20, 2021, as Appl. No. 17/505,790.
Application 17/505,790 is a continuation of application No. 16/672,651, filed on Nov. 4, 2019, granted, now 11,156,997.
Claims priority of application No. 19425063 (EP), filed on Sep. 9, 2019.
Prior Publication US 2022/0035361 A1, Feb. 3, 2022
Int. Cl. F24F 11/62 (2018.01); F24F 11/39 (2018.01); F24F 11/52 (2018.01); G05B 23/02 (2006.01); G06N 3/08 (2023.01); H04L 12/28 (2006.01); F24F 110/64 (2018.01); F24F 120/00 (2018.01); G06N 3/088 (2023.01); G06T 17/00 (2006.01); H02J 13/00 (2006.01)
CPC G05B 23/0283 (2013.01) [F24F 11/39 (2018.01); F24F 11/52 (2018.01); F24F 11/62 (2018.01); G06N 3/08 (2013.01); H04L 12/2825 (2013.01); F24F 2110/64 (2018.01); F24F 2120/00 (2018.01); G06N 3/088 (2013.01); G06T 17/00 (2013.01); H02J 13/00017 (2020.01)] 29 Claims
OG exemplary drawing
 
21. A computer-implemented method for monitoring the operation of an air conditioning system, comprising:
capturing image data inside of the air conditioning system;
processing the image data to determine an image delta value, wherein the image delta value represents a change in the image data over a period of time;
comparing the image delta value to a threshold value;
transmitting an alert signal over a computer network when the image delta value exceeds the threshold value,
wherein the threshold value is determined using a neural network, wherein the neural network is trained using historic images and collected data,
wherein
a) the threshold value is further determined by the neural network using corresponding qualitative assessments of air conditioning systems, or
b) the method further comprises providing a user interface that identifies a current air conditioning system status based on the image delta value and a predictive analysis status based on the image delta value, wherein the predictive analysis status identifies an anticipated time until degradation of the air conditioning system that will require service using the neural network.