- Title
- Mining the big data of residential appliances in the smart grid environment
- Creator
- Yang, Jiajia; Zhao, Junhua; Wen, Fushuan; Kong, Weicong; Dong, Zhaoyang
- Relation
- IEEE Power and Energy Society General Meeting (PESGM 2016). Proceedings of the IEEE Power and Energy Society General Meeting ( 17-21 July, 2016)
- Publisher Link
- http://dx.doi.org/10.1109/PESGM.2016.7741444
- Publisher
- Institute of Electrical and Electronics Engineers (IEEE)
- Resource Type
- conference paper
- Date
- 2016
- Description
- Based on the dynamic time warping (DTW) matching method, a novel appliance identification algorithm for low frequency sampling load data is proposed. First, residential load sequences are segmented into subsequences composed of single appliance load profiles and multi-appliance load profiles. Then, reference load sequences of all candidate appliances, which have identical lengths, are generated before conducting DTW matching. Next, the reference sequence that has the minimum DTW distance to the query subsequence is assigned as the identification result. For a multi-appliance subsequence, the best-matched reference sequence is subtracted from it after each DTW matching, and this process is repeated until all the appliances are identified. The proposed algorithm is implemented in R language and tested with a load dataset, which contains 1000 residential load profiles which all have a length of 24h and are sampled every minute. The result shows that the proposed algorithm is fast and accurate. It only takes 12.67 minutes to identify 1000 load profiles and could identify single appliance load subsequence at an accuracy of 95%. While for multi-appliance subsequences which is much more difficult to identify, it could also achieve an accuracy of about 75%.
- Subject
- smart grid; data mining; R programming language; dynamic time warping; appliance identification
- Identifier
- http://hdl.handle.net/1959.13/1341105
- Identifier
- uon:28657
- Identifier
- ISBN:9781509041688
- Language
- eng
- Reviewed
- Hits: 1732
- Visitors: 1666
- Downloads: 0
| Thumbnail | File | Description | Size | Format |
|---|