With the help of the optical sensor described in this paper, a NIALM system can be integrated with little expense in the existing electrical installation of households for analysing and optimising the active electric power consumption. The monitoring system requires as input device the electromechanical three-phase electricity meter (Ferraris meter) installed in the most of the German households. The active power consumption can be measured with time resolution down to second. The measured data constitute the basis for an autonomous NIALM approach which can without manual initialisation phase extract the switching sequence structures of the chief consumer load devices in the household, from the total load trace. The focus of the analysis lies on the automatic controlled on-off consumer loads. The analytical procedure consists of an approach based on certain rules for recognising repetitive patterns. In the decomposition of the total load as a function of time into switching events, simple on-off consumer loads and combinations of these patterns are investigated. The recognised patterns can be stored in a neural network. The network is updated daily and weekly. The results of the analysis are evaluated with regard to the potential for demand optimisation or for controlling the energy consumption of individual electric consumer devices.
[1]
Hans Bandemer,et al.
Fuzzy Data Analysis
,
1992
.
[2]
Dr. Zbigniew Michalewicz,et al.
How to Solve It: Modern Heuristics
,
2004
.
[3]
Sankar K. Pal,et al.
Fuzzy models for pattern recognition
,
1992
.
[4]
Teuvo Kohonen,et al.
Self-Organizing Maps
,
2010
.
[5]
Hans Bandemer.
Fuzzy Analysis of Fuzzy Data
,
1993
.
[6]
Steven B. Leeb,et al.
Non-intrusive electrical load monitoring in commercial buildings based on steady-state and transient load-detection algorithms
,
1996
.
[7]
Radu Zmeureanu,et al.
Using a pattern recognition approach to disaggregate the total electricity consumption in a house into the major end-uses
,
1999
.
[8]
Radu Zmeureanu,et al.
Nonintrusive load disaggregation computer program to estimate the energy consumption of major end uses in residential buildings
,
2000
.
[9]
Clifford Lau,et al.
Neural Networks: Theoretical Foundations and Analysis
,
1991
.
[10]
G. W. Hart,et al.
Nonintrusive appliance load monitoring
,
1992,
Proc. IEEE.