On the characterization and monitoring of building energy demand using statistical process control methodologies

A statistical process control based system and scheme for monitoring, assessment and tracking of energy consumption is presented. The proposed strategy is set up using a Multichannel Structure to model and estimate statistics profile (average and uncertainty) of the energy consumption of a building during a predetermined cycle, e.g. a week period. These statistics profile is analyzed with statistical process control (SPC) techniques for signaling alarms or reporting interpreted faults of unpredicted or unusual behavior of energy demand. Identification of deviations from predicted or programed consumption profiles modeled by a Multichannel Structure are shown to be easily designed, tuned and implemented with low computation demand of memory and processing speed. Simulated results illustrate the Multichannel Structure programing and parameters tuning. Experimental results are presented to exemplify the proposed strategy. (C) 2013 Elsevier B.V. All rights reserved.