A Deep Learning based Sustainable Energy Scheduling System

Climate change greatly affects human life. According to previous studies, the biggest cause of climate change is the increase of greenhouse gases in the atmosphere. In order to effectively reduce carbon emissions, many countries not only formulate many carbon reduction policies, but also focus on developing green energy. Green energy, e.g., solar energy and wind energy, is a solution to reduce carbon emission; however, it is difficult for green electricity to provide stable energy supply due to climate factors. In this paper, we proposed a long short-term memory (LSTM) deep learning model based sustainable energy scheduling system to control the energy usage between green power and utility power. The proposed system analyzes the data of green power generation and weather conditions, and based on the weather forecast and the results calculated by this system, predicts the power generation status of solar and wind energy in the next few days. According to the experimental results, by using the proposed system, the goals of stable power supply and minimization of carbon emissions from energy consumption can be achieved.