Research on non-intrusive load disaggregation

Non-intrusive load decomposition of residential electricity consumption is a technology that decomposes the total load information of users into the energy consumption information of each electrical equipment. The obtained energy consumption information can be used to analyze users’ electricity consumption behavior, and guide users to manage energy consumption reasonably, thus saving electricity bills and reducing unnecessary power consumption. At the same time, according to the energy consumption information of electrical equipment, we can further carry out refined distribution work, effectively enhance the inter- active response ability between power grid and users. In this thesis, We propose a new model inspired by YOLO architecture. Its goal is to detect the activation of the appliance in a forward pass. The advantage of this model is that we can predict the electrical appliance through the total table without the data of sub table. The model was applied to public household energy data, and its result can be compared with the most advanced performance at present.