Statisticians from the National University of Singapore and the University of Texas have designed a pattern-recognition algorithm that records the energy consumed by Industrial equipment. The algorithm based framework gathers raw data from various components and identifies abnormal energy patterns. Company engineers can use this data to analyze machine defects, and hopefully reduce the irregular energy patterns through data filtering and machine tweaking.
The model uses recorded machine energy states to organize the data. Sensors implemented within the machines collect the energy consumption data and and feed the data to the mathematical model (the programmers call the model a "finite-state machine," or FSM). The FSM ultimately creates power consumption profiles for each individual machine.
The Agency for Science, Technology and Research (A*STAR). "Real-time energy audit reduces power consumption." ScienceDaily, 24 Oct. 2013. Web. 31 Oct. 2013.
http://www.sciencedaily.com/releases/2013/10/131024114117.htm
No comments:
Post a Comment