2021 International Conference on System, Computation, Automation and Networking (ICSCAN) | 2021
Implementation of Deep Learning Neural Network In Forecasting of Solar Power
Abstract
Solar power is a non-conventional cleanest form of energy which is abundant and available gratuitously. Mostly the generated solar PV power is fed to the grid. The power generated from solar depends upon various dynamically changing environmental factors such as irradiation, temperature etc which demands an high spinning reserve over the grid side. In this article, a Deep learning Neural Network based model is implemented to predict the solar power generation of a solar plant by using various environmental parameters by which spinning reserve stress towards the grid can be adhered. The proposed Deep learning model is compared with the conventional neural network model of single & two hidden layers and the results are compared. The results shows promising efficiency of deep learning neural network based system over the conventional neural network for the power predictions on various environmental scenarios.