Analysis and techniques for non-intrusive appliance load monitoring.
| dc.contributor.advisor | Morsi, Walid | |
| dc.contributor.author | Alshareef, Sami | |
| dc.date.accessioned | 2014-12-23T17:40:44Z | |
| dc.date.accessioned | 2022-03-30T17:04:18Z | |
| dc.date.available | 2014-12-23T17:40:44Z | |
| dc.date.available | 2022-03-30T17:04:18Z | |
| dc.date.issued | 2014-11-01 | |
| dc.degree.discipline | Electrical and Computer Engineering | en |
| dc.degree.level | Master of Applied Science (MASc) | en |
| dc.description.abstract | The increased public awareness of energy conservation and the demand for smart metering system have created interests in home energy monitoring. Load disaggregation using a single sensing point is considered a cost-effective way to sense individual appliance operation as opposed to using dedicated sensors for appliance monitoring. The aim of this thesis is to investigate the effectiveness of the analysis methods and techniques used in load disaggregation using a single point sensing. Time-frequency analysis methods such as Wavelet transforms are carefully examined and machine learning classifiers are used to develop the appropriate prediction models. The results have shown that the use of different Wavelet functions can significantly affect the classification accuracy. Among the four wavelets investigated in this thesis, two wavelets (Daubechies and Symlets) are able to provide the highest mean classification accuracy. | en |
| dc.description.sponsorship | University of Ontario Institute of Technology | en |
| dc.identifier.uri | https://hdl.handle.net/10155/485 | |
| dc.language.iso | en | en |
| dc.subject | Non-intrusive load monitoring | en |
| dc.subject | Power components | en |
| dc.subject | Decision tree classification | en |
| dc.subject | Edge detection | en |
| dc.subject | Transient feature analysis | en |
| dc.title | Analysis and techniques for non-intrusive appliance load monitoring. | en |
| dc.type | Thesis | en |