Automated Secure Computing for Next-Generation Systems

16Machine Learning-Based Intelligent Power Systems

Kusumika Krori Dutta¹, S. Poornima¹*, R. Subha¹, Lipika Deka² and Archit Kamath³

¹Department of Electrical and Electronics Engineering, M S Ramaiah Institute of Technology, Bangalore, India

²School of Computer Science and Informatics, De Montfort University, Leicester, UK ³Nanyang Technological University, Singapore, Singapore

Abstract

Machine learning (ML) plays a crucial role in power systems by providing advanced tools for data analysis, pattern recognition, and decision making. Some of the key applications of machine learning in power systems include load forecasting, predictive maintenance, load scheduling, state estimation, optimization, fault detection, energy management, power quality monitoring, etc. The researchers have used many classification and regression algorithms of ML towards developing a smart power system. Among all machine and deep learning methods, convolutional neural networks, support vector machines, recurrent neural network, K nearest neighbor, decision tree, etc., are widely used in various aspects of power systems. In this book chapter, various machine learning techniques are explained along with its implementations towards intelligent power systems. A case study on the fault detection of IEEE five-bus systems using different machine learning techniques is explained in greater detail.

Keywords: ML, DL, power systems, fault classification, KNN, RNN, SVM

16.1 Introduction

An intelligent power system ...