Developing an Energy Management System for Optimal Operation of Prosumers Based on a Modified Data-Driven Weather Forecasting Method

Renewable energy sources (RESs) are playing a significant role in the optimal operation of residential prosumers. Uncertainty of weather parameters such as solar irradiance and wind speed can affect the output power of RESs and consequently day-ahead operation of prosumers. Therefore, in this study, a new energy management system (EMS) has been proposed to mitigate fluctuations of RESs by proposing a 2-Level corrective weather forecasting method based on multilayer perceptron artificial neural networks (MLP-ANN). Forecasted data of the proposed method is applied to a peer-to-peer (P2P) prosumer environment. And the effects of weather uncertainty on operation cost and operational decisions of the residential prosumer are investigated. Simulation results indicate the effectiveness of the suggested data-driven method for weather prediction in the optimization of the prosumer.