Data Science Approaches in Machine Learning for Analytics in Power Systems

EasyChair Preprint 12278

Abstract

The integration of data science approaches into machine learning applications has emerged as a transformative paradigm in the field of power systems analytics. This study investigates the synergies between data science techniques and machine learning algorithms, aiming to enhance the efficiency, reliability, and sustainability of power systems. The application of advanced analytics in power systems is pivotal for handling the increasing complexity and volume of data generated by modern energy infrastructures. This research explores various data science methodologies such as data preprocessing, feature engineering, and exploratory data analysis, laying the foundation for robust machine learning models. Emphasis is placed on leveraging supervised learning techniques for predictive maintenance, fault detection, and load forecasting. Unsupervised learning methods are employed for anomaly detection and clustering analysis, contributing to the identification of hidden patterns within power system data. The integration of reinforcement learning techniques facilitates optimal decision-making in dynamic and complex power grid scenarios. Additionally, this study delves into the utilization of deep learning models, particularly neural networks, for their ability to capture intricate relationships in large-scale power system datasets.

Keyphrases: Data Science, Predictive Maintenance, Reinforcement Learning, anomaly detection, clustering analysis, deep learning, fault detection, load forecasting, machine learning, power systems

@booklet{EasyChair:12278,
  author    = {Deep Himmatbhai Ajabani},
  title     = {Data Science Approaches in Machine Learning for Analytics in Power Systems},
  howpublished = {EasyChair Preprint 12278},
  year      = {EasyChair, 2024}}