使用活動電器關聯模型偵測活動非相關電器

Abstract

Providing energy conservation services becomes a hot research topic because more and more people attach importance to environmental protection. This research proposes a framework that consists of two main elements, including activity-appliance model and irrelevant appliance detection. First of all, an activity-appliance model has been built to associate activities with appliances according to the data of common sense. For instance, oven is used for cook, draught fan is used for hair drying, etc. Then, using the relationships can help to recognize the resident’s current activity and detect irrelevant appliance, which are consuming electric power but not take part in the current activities. Finally, the analysis of activity-appliance models shows most appliances are strong activity-related. Such a property offers benefit to identify activity and detect irrelevant appliance. The experimental results show that the work can achieve higher than 74% accuracy for activity recognition, and receive a low false alarm for detection of irrelevant appliance in a real home environment.