Powering data centers with renewable energy and storage presents a range of technical and economic challenges that independent power producers, EPC contractors, and investors have only recently begun to address. In this context, large-scale batteries are expected to play a key role, supplying the active power needed during demand spikes and coordinating energy flows between generation, storage, and data center loads.
Yet it remains unclear whether current battery technologies can perform these tasks efficiently, raising doubts about the short-term ability of the solar-plus-storage industry to meet the growing electricity demand from data centers.
To explore this, an international research team investigated how batteries could effectively support data center power and found that the development of advanced battery management systems will be crucial. “Energy storage, battery technologies, artificial intelligence (AI) integration, and thermal regulations can be considered as the top four sectors requiring more investigation and research to have reliable data centers, with the lowest costs and highest revenues,” the group stated.
Join us on April 22 for the 3rd SunRise Arabia Clean Energy Conference in Riyadh. The event will spotlight how solar and energy storage solutions are driving sustainable and reliable infrastructure, with a particular focus on powering the country’s rapidly growing data center sector. In the paper “A research-industry perspective of battery systems technology for next-generation data centers,” published in the Journal of Energy Storage, the researchers noted that most existing reviews focus on demand response, load modeling, energy management systems, airflow control, resource allocation, and energy-efficient architecture, but rarely address batteries or their integration with standards, industry white papers, financed projects, and practical applications. “Powering data centers entirely with renewables and storage alone is not immediately feasible everywhere,” the research's corresponding author, Ashkan Safari, told pv magazine. “Renewable sources, such as solar and wind, are intermittent, and the amount of energy storage needed to ensure 24 hours, 7 days of reliable operation remains very large and expensive. Economically, the upfront costs of large-scale renewables and long-duration storage are often higher than using the grid, especially when grid electricity is cheap. Following this, most data centers today rely on a hybrid approach – renewables and storage supported by the grid.” “As renewable generation becomes cheaper, storage technologies mature, and grids become cleaner and more flexible, most types of data centers could be powered primarily by renewables,” he went on to say. “However, differences in location, size, reliability requirements, and workload criticality mean that some data centers, such as large-scale ones, will adopt this faster than smaller ones.” According to Safari, the feasibility of powering data centers with renewable energy and storage largely depends on grid features. Grids with high renewable penetration, strong transmission capacity, low congestion, and fast balancing services can effectively manage renewable variability, reducing the need for on-site generation, storage, and lowering costs for data centers. In contrast, limited grid capabilities force data centers to oversize on-site renewables and storage, making fully renewable operation more technically challenging and expensive. The study fills this gap by providing a comprehensive overview of battery technologies, their components, and control methodologies, while linking them to broader energy efficiency strategies in data centers. This holistic approach covers both batteries for uninterrupted power supply (UPS) inside data centers and battery energy storage systems (BESSs) deployed outside the facilities. The researchers reviewed all battery technology systems for data centers, outlining power architectures and recommended topologies, examining batteries and battery management systems (BMSs) integrated with AI control strategies, summarizing net-zero initiatives and industry standards, and analyzing recent industry white papers, proofs-of-concept, and financed projects. By detailing UPS topologies and the types of BESS in use, they highlighted that advanced BMS technologies will be essential for reliable, efficient, and sustainable data center operations. “BMS technology is critical for ensuring the safety, remaining useful life (RUL), and efficiency of lithium-ion batteries by monitoring parameters such as temperature, voltage, current, and battery states,” the team explained. “It prevents overcharging, overheating, and deep discharging, and enables strategies like cell balancing, thermal management, and predictive maintenance, which optimize battery performance.” The academics explained that, in modern data centers, BMSs continuously monitor key states including state of charge (SoC), state of health (SoH), state of energy (SoE), and state of temperature (SoT). Cell monitoring collects voltage, temperature, and current data from individual cells, enabling real-time adjustments to optimize performance and extend battery life. The hierarchical BMS architecture comprises three layers – data, computation, and application – allowing the integration of measurement, modeling, and real-time control. The computation layer estimates internal battery states, while the application layer manages security, aging, thermal control, balancing, and fault diagnosis. SoC and SoH estimation uses conventional, AI, observer-based, adaptive filter, and hybrid methods, often integrated into multi-state estimation frameworks that predict RUL and performance across real-time, intermediate, and long-term scales. Moreover, the team emphasized that advanced AI-based models such as neural networks, neuro-fuzzy systems, reinforcement learning, BiLSTM, LSTM, random forests, and generative adversarial networks (GANs) further enhance state estimation, fault detection, and predictive control. Activation functions support these AI models by mapping inputs effectively, enabling accurate SoC and SoH predictions and optimizing BMS performance for high-performance data centers. The researchers concluded that AI-driven BMSs will be able to provide real-time monitoring, predictive control, and optimized energy management, laying the foundation for intelligent, reliable, and sustainable battery operation in data centers. “The integration of AI into BMS improves state estimations, reduces failures, and extends RUL,” they stated. “However, challenges remain in scaling AI-driven solutions, achieving net-zero data centers, and meeting diverse industry-specific requirements.” This content is protected by copyright and may not be reused. If you want to cooperate with us and would like to reuse some of our content, please contact: editors@pv-magazine.com.
By submitting this form you agree to pv magazine using your data for the purposes of publishing your comment.
Your personal data will only be disclosed or otherwise transmitted to third parties for the purposes of spam filtering or if this is necessary for technical maintenance of the website. Any other transfer to third parties will not take place unless this is justified on the basis of applicable data protection regulations or if pv magazine is legally obliged to do so.
You may revoke this consent at any time with effect for the future, in which case your personal data will be deleted immediately. Otherwise, your data will be deleted if pv magazine has processed your request or the purpose of data storage is fulfilled.
Further information on data privacy can be found in our Data Protection Policy.