Over more than 30 years development of Non-intrusive Load Monitoring (NILM), many technologies were developed from sampling frequency 1 Hz to hundreds kHz to extract from simple information such as difference power dP, dQ; shape of transient signatures to complex information such as harmonics or electromagnetic interference (EMI). There are now many real NILM systems in residential applications but there is lack of researches to apply NILM to commercial and industrial application. This site uses so much energy in business and production so that analyzing power consumption to optimize their operation is very important to produce competitive products on the market. Unfortunately, it is very difficult to recognize commercial and industrial machine that are composed of many electric elements. This paper provides a new NILM approach to monitor commercial and industrial machines based on basic electrical appliances inside architecture of machine. All power differences dP, dQ, dS are used to analyze and classify basic electrical appliances. Adaptive filter CUSUM is used to improve removing noise in preprocessing phase of the NILM system. Finally, Field Program Gate Array (FPGA) is discussed to implement our NILM system.
[1]
S.B. Leeb,et al.
Estimation of variable-speed-drive power consumption from harmonic content
,
2005,
IEEE Transactions on Energy Conversion.
[2]
S. R. Shaw,et al.
Transient event detection in spectral envelope estimates for nonintrusive load monitoring
,
1995
.
[3]
S. Drenker,et al.
Nonintrusive monitoring of electric loads
,
1999
.
[4]
Steven B. Leeb,et al.
Power signature analysis
,
2003
.
[5]
Fredrik Gustafsson,et al.
Adaptive filtering and change detection
,
2000
.
[6]
G. W. Hart,et al.
Nonintrusive appliance load monitoring
,
1992,
Proc. IEEE.
[7]
Dong Luo,et al.
Detection and diagnosis of faults and energy monitoring of HVAC systems with least-intrusive power analysis
,
2001
.
[8]
Eric C. Larson,et al.
Disaggregated End-Use Energy Sensing for the Smart Grid
,
2011,
IEEE Pervasive Computing.