{"type":"ScholarlyArticle","title":"On the Balance of Meter Deployment Cost and NILM Accuracy","authors":[{"name":"Xiaohong Hao","slug":"xiaohong-hao","given":"Xiaohong","family":"Hao","affiliation":""},{"name":"Bangsheng Tang","slug":"bangsheng-tang","given":"Bangsheng","family":"Tang","affiliation":""},{"name":"Yongcai Wang","slug":"yongcai-wang","given":"Yongcai","family":"Wang","affiliation":""}],"abstract":"Non-Intrusive Load Monitoring (NILM) uses one smart meter at the power feed to disaggregate the states of a set of appliances. Multiple NILM meters are deployed to achieve high monitoring accuracy in large-scale power systems. Our work studies the tradeoff between monitoring accuracy and meter deployment, in a quantitative and extensible way. In particular, we introduce a clearness function as an abstract indicator of expected monitoring accuracy given any NILM method, and then showcase two concrete constructions. With the notation of a clearness function, we propose solutions to the smart meter deployment problem (SMDP), that is, the problem of finding a deployment scheme with minimum number of meters while attaining a required monitoring accuracy. Theoretically, SMDP is shown NP-hard and a polynomial-time approximation scheme (PTAS) is proposed in this paper. For evaluation, we show that our proposed scheme is efficient and effective in terms of approximation ratio and running time. On real and simulated datasets, our proposed framework achieves a higher monitoring accuracy at a much lower cost, outperforming common baseline algorithms.","doi":null,"url":"https://ijcai.org/proceedings/2015/369","pdf":"https://ijcai.org/proceedings/2015/0369.pdf","markdown":"https://ijcai.org/proceedings/2015/369.md","bibtex":"https://ijcai.org/proceedings/2015/bibtex/369","volume":{"slug":"2015","title":"Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2015","url":"https://ijcai.org/proceedings/2015/","year":2015,"location":"Buenos Aires, Argentina","dates":"25–31 July 2015","editors":"Qiang Yang and Michael Wooldridge","isbn":"978-1-57735-738-4","publisher":"International Joint Conferences on Artificial Intelligence","doi_prefix":""},"section":"Special Track on Computational Sustainability","track":"Special Track on Computational Sustainability","number":369,"pages":{"first":2603,"last":2608},"keywords":[],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}