An Enhanced Power Management and Prediction for Smart Grid Using Machine Learning

In recent years, managing power generation and utilisation to stay on track has played a major role. Apart from this, analysis of energy bill generation and smoothing the peak demand curve to an optimal level are very difficult with the existing system. More power loss has occurred due to the fault detection in the smart grid; it has created manipulation data and obstacles for the training of the data. Hence, to overcome this, the proposed machine learning-based power management and prediction system (ML-PMPS) for the smart grid has been operated, which is essential for providing an uninterrupted power supply to consumers. As per simulation analysis, the proposed method performs better as compared to the conventional methods such as Multilayer Perceptron (MLP) and Naive Bayesian (NB) with respect to accuracy, prediction rate, and power management.