US 11,941,506 B2
System and method for monitoring via smart devices
Paul Robert Langdon, Wethersfield, CT (US)
Assigned to Hubbell Incorporated, Shelton, CT (US)
Filed by Hubbell Incorporated, Shelton, CT (US)
Filed on Feb. 3, 2021, as Appl. No. 17/166,553.
Claims priority of provisional application 62/969,958, filed on Feb. 4, 2020.
Prior Publication US 2021/0241927 A1, Aug. 5, 2021
Int. Cl. G06N 3/04 (2023.01); G16Y 20/20 (2020.01); G16Y 40/10 (2020.01); G16Y 40/20 (2020.01); G16Y 40/50 (2020.01)
CPC G06N 3/04 (2013.01) [G16Y 20/20 (2020.01); G16Y 40/10 (2020.01); G16Y 40/20 (2020.01); G16Y 40/50 (2020.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
transmitting, via one or more devices, one or more first signals indicative of one or more first statuses of the one or more devices;
receiving, via a controller, the one or more first signals;
determining, via the controller and based on the one or more first statuses, a base model;
transmitting, via one or more devices, one or more second signals indicative of one or more second statuses of the one or more devices;
receiving, via the controller, the one or more second signals;
comparing, via the controller, the one or more second statuses to the base model;
determining, via the controller and based on said comparing, an occurrence of an abnormal condition when the one or more second statuses are different than the base model;
receiving, via the controller and transmitted via the one or more devices, one or more third signals indicative of one or more third statuses of the one or more devices;
determining, via the controller, an occurrence of a recurring abnormal condition when the one or more third statuses are the same as the one or more second statuses;
updating, via the controller, the base model based on the recurring abnormal condition to modify the first status to include the recurring abnormal condition;
determining, based on the comparison, a safe score;
comparing the safe score to a threshold to determine the occurrence of the abnormal condition; and
comparing, via the controller, the safe score to the base model and updating the base model if the safe score is within a similarity range of the base model.