Multi-Source and Multi-Powered Smart Grid Prediction Using Deep Learning

Aim: This study's main goal is to precisely locate the sources of power supplies, forecast power rates, and alert consumers to high energy flow in order to reduce dangers to electronic devices and improve grid efficiency and stability. Methods and Materials: The proposed work included four groups. Group 1 refers to Deep Neural Network (DNN) algorithm used to, Group 2 refers to Long Short- Term Memory algorithm which is used to be more accurate when using datasets with longer sequences, Group 3 Gated Recurrent Unit (GRU) algorithm used for multi-source and multi-powered smart grid prediction with an astounding accuracy. Results: Achieving 99.03% accuracy using GUR algorithm in multi-source and multi-powered smart grid prediction. Conclusion: Outperforming LSTM in source and grid identification, the GRU algorithm shows promise for improving grid prediction.