An unsupervised electrical appliance modeling framework for non-intrusive load monitoring
| Autor: | Bo Zeng, Bo Liu, Wenpeng Luan, Yinxin Yu |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Structure (mathematical logic)
Measure (data warehouse) business.industry Event (computing) Computer science 020206 networking & telecommunications Control engineering 02 engineering and technology Set (abstract data type) 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Electricity State (computer science) business |
| Zdroj: | 2017 IEEE Power & Energy Society General Meeting. |
| DOI: | 10.1109/pesgm.2017.8273794 |
| Popis: | Non-Intrusive Load Monitoring (NILM) is a novel and cost-effective technique for monitoring the load electricity consumption details. Building the load model of every concerned electrical appliance for modeling its electricity consumption behavior is the basis of implementing NILM. The vast majority of existing NILM approaches need to individually measure the concerned appliance in the targeted scenarios to build its appliance model before they can be put into use. This will constrain the practical application of NILM. In this paper, we propose a fully unsupervised electrical appliance finite state machine (FSM) modeling framework for NILM. Without intruding into the targeted scenarios or requiring any prior knowledge of the unmodeled appliances, it can automatically establish the FSM model for different concerned appliances only based on aggregated load event related signatures. For each appliance, the established model includes the information of the complete state set and topological structure of FSM, and the related model parameters. We believe the proposed framework can significantly improve the applicability of the existing NILM technologies, and provide a basis for the realization of autosetup NILM. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|