A Deep Learning Model For Intelligent Energy Management
Lalitha Krishnasamy, K. Somasundaram, Munleef Quadir, Rajesh Kumar Dhanaraj, C. Roopa, P Kowshika
Abstract
Green energy management is a cost-effective method of reducing energy use, but current research overlooks the edge intelligence potential of managed IoT. As a result, it focuses on today's smart grid, residential, and commercial demands to develop a deep learning-based paradigm for successful energy management. In our system, we want to forecast or analyze household power usage. The dataset on home electricity usage can be found in the dataset repository. The pre-processing phase is then implemented to avoid inaccurate predictions. We use a range of pre-processing approaches to deal with various types of power data, then create an effective decision-making system for short-term forecasting on reliable resource-constrained devices. The system employs deep learning techniques such as Long Short Term Memory (LSTM) and Gate Recurrent Unit to analyze energy use (GRU).