US 11,906,958 B2
Systems and methods for determining occurrence of pattern of interest in time series data
Tanushyam Chattopadhyay, Kolkata (IN); Abhisek Das, Kolkata (IN); Suvra Dutta, Kolkata (IN); Shubhrangshu Ghosh, Kolkata (IN); and Prateep Misra, Kolkata (IN)
Assigned to Tata Consultancy Services Limited, Mumbai (IN)
Filed by Tata Consultancy Services Limited, Mumbai (IN)
Filed on Jul. 2, 2021, as Appl. No. 17/366,777.
Claims priority of application No. 202121001728 (IN), filed on Jan. 13, 2021.
Prior Publication US 2022/0221847 A1, Jul. 14, 2022
Int. Cl. G05B 23/02 (2006.01); G06F 18/20 (2023.01); G06F 18/2415 (2023.01)
CPC G05B 23/0227 (2013.01) [G06F 18/2415 (2023.01); G06F 18/285 (2023.01)] 18 Claims
OG exemplary drawing
 
1. A processor implemented method, comprising:
obtaining, via one or more hardware processors, an input time series data corresponding to one or more sensors attached to at least one computing device;
computing, via the one or more hardware processors, a first order derivative over time using the obtained input time series data;
computing, via the one or more hardware processors, a gradient of change in value of the obtained input time series data of the one or more sensors over time based on the first order derivative;
deriving, via the one or more hardware processors, an angle of change in direction based on the gradient of change in value of the one or more sensors over time, and converting the derived angle to a measurement unit;
quantizing, via the one or more hardware processors, the input time series data into a plurality of bins based on the measurement unit;
obtaining a weighted finite state transducers diagram (WFSTD) based on a domain knowledge;
converting, via the one or more hardware processors, the WFSTD into a specific pattern;
determining, via the one or more hardware processors, a number of occurrences of the specific pattern in one or more bins of the plurality of bins within a specific time range; and
detecting, via the one or more hardware processors, one or more anomalies based on the number of occurrences of the specific pattern within the specific time range.