Indonesian Journal of Electrical Engineering and Computer Science Vol. 34, No. 2, May 2024, pp. 697∼710 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v34.i2.pp697-710 ❒ 697 Support vector regression-based state of charge estimation for batteries: cloud vs non-cloud Mohamed Ben Youssef 1 , Imen Jarraya 2 , Mohamed Ali Zdiri 1 , Fatma Ben Salem 3 1 Control and Energy Management Laboratory, Sfax Engineering School, University of Sfax, Sfax, Tunisia 2 College of Computer and Information Sciences, Prince Sultan University, Riyadh, Saudi Arabia 3 Department of Electrical, Engineering School of Prince Sattam Bin Abdulaziz, Elkahrj, Saudi Arabia Article Info Article history: Received Nov 3, 2023 Revised Feb 14, 2024 Accepted Feb 16, 2024 Keywords: Basic cloud AWS EC2 Experimental results Lithium-ion batteries MATLAB production server Non-cloud MATLAB SOC estimation Support vector regression ABSTRACT Embracing the potential of cloud technology in the field of electric vehicle advancements, this paper explores the application of support vector regression (SVR) for accurate state of charge (SOC) estimation of lithium-ion batteries in various computational landscapes. This study aims to scrutinize and compare the performance of SOC estimation, with a specific focus on precision, computational efficiency, and execution speed. The investigation is conducted across diverse environments, including a traditional non-cloud setup and two cloud-based platforms-a standard cloud environment employing Amazon web services (AWS) EC2 servers and an enhanced configuration utilizing the MATLAB production server. The investigation not only emphasizes the effectiveness of cloud integration but also provides valuable insights into the strengths and weaknesses of the proposed methodology. The experimental results contribute to a nuanced understanding of the methodology’s performance, shedding light on its potential implications for advancing electric vehicle technologies. This study thus extends its significance beyond technical considerations, providing a broader perspective on its relevance to global electrification initiatives. This is an open access article under the CC BY-SA license. Corresponding Author: Mohamed Ali Zdiri Control and Energy Management Laboratory, Sfax Engineering School, University of Sfax Sfax, Tunisia Email: mohamed-ali.zdiri@enis.tn 1. INTRODUCTION In recent years, the dynamic expansion of cloud computing has drastically revolutionized a variety of businesses, offering new levels of efficiency, scalability, and data management capabilities [1], [2]. This technological innovation is critical as the global energy landscape evolves toward sustainable sources. Cloud technology is pivotal in renewable energy, driven by environmental concerns and climate change urgency. It manages complex data from sources like solar and wind power, enabling real-time information exchange essential for energy output regulation [3]. The use of batteries offers significant advantages in energy management strategies. Batteries play a crucial role in storing excess energy during periods of low demand and releasing it when demand is high, contributing to load balancing and grid stability [4]–[8]. They serve as a valuable tool in optimizing the integration of renewable energy sources, allowing for better management of intermittent energy production [9]–[12]. In electric vehicles, cloud technology enhances connectivity and intelligent systems, facilitating battery management, vehicle-to-grid services, and predictive maintenance [13], and [14]. Additionally, cloud integration in unmanned aerial vehicles (UAVs) improves data processing, flight Journal homepage: http://ijeecs.iaescore.com