US 11,656,589 B2
Systems and methods for automatic power topology discovery
Xiaolong Zhang, Shanghai (CN); Benfeng Tang, Shanghai (CN); Linglai Li, Shanghai (CN); David B. Berman, Franklin, WI (US); Phillip R. Bush, New Berlin, WI (US); Steven T. Haensgen, New Berlin, WI (US); Michael L. Gasperi, Racine, WI (US); Sean C. Schmelzer, Colgate, WI (US); and Alex Nicoll, Brookfield, WI (US)
Assigned to Rockwell Automation Technologies, Inc., Mayfield Heights, OH (US)
Filed by Rockwell Automation Technologies Inc., Mayfield Heights, OH (US)
Filed on Sep. 9, 2020, as Appl. No. 17/16,049.
Prior Publication US 2022/0075337 A1, Mar. 10, 2022
Int. Cl. G06N 7/00 (2023.01); G01R 19/25 (2006.01); G05B 19/042 (2006.01); H02J 13/00 (2006.01)
CPC G05B 19/042 (2013.01) [G01R 19/2513 (2013.01); G06N 7/005 (2013.01); H02J 13/00002 (2020.01); G01D 2204/47 (2021.05); G05B 2219/2639 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method of identifying power network topology, the method comprising:
collecting power data from a plurality of devices within a power network;
detecting a power change event in at least a first power device of the plurality of devices and a second power device of the plurality of devices;
identifying, in at least a portion of the power data associated with the power change event, a power signature of the first power device and a power signature of the second power device;
updating a probability that the first power device is connected to the second power device based on the power signature of the first power device and the power signature of the second power device; and
generating a network topology diagram, wherein the network topology diagram reflects the probability that the first power device is connected to the second power device.