Big Data Analytical Techniques for Electrical Energy Forecasting in Smart Grid Paradigm

• Published in: Applications of Big Data and Artificial Intelligence in Smart Energy Systems Vol. 1; pp. 101 - 126
• Main Authors: Rawal, Keerti, Ahmad, Aijaz
• Format: Book Chapter
• Language: English
• Published: River Publishers 2023
• Edition: 1
• Subjects: General Engineering & Project Administration; General References; Software Engineering; RR; Electrical Demand Data; PR; DTR; Demand Forecasting; LPWAN; Ml Algorithm; DE Algorithm; Lasso Algorithm; Big data; ANFIS; High Variance Data; Peak Load Forecasting; feature selection; RFR; Supervise Ml Algorithm; smart grid; Wrapper Methods; Absolute Percentage Error; Supervise Ml; machine learning; Hybrid Renewable Energy System; Attribute Selection; Attribute Selection Method; Input Features; renewable energy forecasting; MedAE; electrical demand forecasting; Time Varying Electricity Pricing
• ISBN: 9788770228251; 9788770229944; 8770228256; 8770229945
• DOI: 10.1201/9781003440710-5

Electrical energy demand forecasting is fundamental for stable operation of an electrical grid to maintain a continuous balance between supply from generating stations and consumers' demand. It facilitates optimum and economical utilization of resources. This fundamental problem is gaining attention again with the increasing integration of renewable energy sources. As real-time electricity market operation is becoming inevitable as a way to mitigate volatility of weather-dependent renewable sources, the complexity of electricity demand prediction has been observed to increase. Big data analytics is an area that has shown promising results in handling complex problems such as electricity demand forecasting. This book chapter explores the advancements in forecasting techniques of electrical demand. These advancements are explored via survey of existing literature, demonstration of techniques, and a comparative analysis of performance of machine learning techniques. Machine learning techniques, viz. linear regression (LR), polynomial regression (PR), ridge regression (RR), least absolute shrinkage and selection operator (LASSO) regression, decision tree regression (DTR), random forest regression (RFR), and K nearest neighbor regression (KNNR) are demonstrated and discussed. Application of machine learning techniques are presented on the one-year electrical demand data of New South Wales area of the Australian electricity market. A brief review of renewable energy forecast techniques is also presented. Electrical energy demand forecasting is fundamental for stable operation of an electrical grid to maintain a continuous balance between supply from generating stations and consumers' demand. Big data analytics is an area that has shown promising results in handling complex problems such as electricity demand forecasting. This chapter explores the advancements in forecasting techniques of electrical demand. These advancements are explored via survey of existing literature, demonstration of techniques, and a comparative analysis of performance of machine learning techniques. The chapter demonstrates machine learning techniques, viz. linear regression, polynomial regression, ridge regression, least absolute shrinkage and selection operator regression, decision tree regression, random forest regression, and K nearest neighbor regression. It presents application of machine learning techniques on the one-year electrical demand data of New South Wales area of the Australian electricity market. The chapter provides a brief review of renewable energy forecast techniques.