Energy Disaggregation on Real Time Ground Truth Data

Deep neural networks are regarded as the fundamental axis of the artificial intelligence research in the present day. Recent years have seen an increase in the volume of information, which has led to the creation of sophisticated deep learning models. These models are able to draw inferences from massive data sets that aid in understanding and addressing issues. In contrast, significant progress in hardware and electronics has led to the development of inexpensive IoT data gathering devices for everyday usage. One such intelligent gadget is the electric meter, which is installed in a home's central panel and records electrical data. A contemporary challenge in electricity consumption is the separation of total electrical energy into states that indicate the use of particular home appliances. With the use of deep neural network models and the data obtained by smart meters, it is possible to monitor the power usage of each individual device, resulting in energy savings and a greater concern for the environment. Consequently, the objective of this study is to construct regression-based deep learning models that segregate electricity from actual data gathered by smart meters of the business MEAZON S.A.