Automating Battery Storage Deployment through AI-enabled Design

The clean energy transition is accelerating, with renewable sources such as solar and wind energy leading the charge. Due to its intermittent nature, renewable generation is increasingly coupled with Battery Energy Storage Systems (BESS), which not only increases utilization and corresponding financial revenues but also contributes to grid stability and resiliency. Energy systems researchers estimate that decarbonizing the electric grid will likely include a massive deployment of BESS, as the cost and performance of lithium-ion batteries continue to improve and alternatives such as sodium-ion and iron-air batteries gain ground. The National Renewable Energy Laboratory (NREL), for instance, foresees that the installed capacity of energy storage in the United States will scale from 26 GW in 2024 to 175 GW by 2050, highlighting the critical need for efficient deployment methods.

For the Engineering, Procurement, and Construction (EPC) firms responsible for deploying BESS in a competitive market, speed and efficiency are paramount. These firms are, therefore, increasingly turning to automation to streamline the complex process of designing and sizing BESS projects.

This shift to automation is profoundly impacting the vital early stages of project development, especially in bidding and initial design. Previously, crafting competitive bids was a tedious process that required extensive manual calculations. Nowadays, advanced automation tools can quickly analyze market data, operational parameters, and potential revenue from energy arbitrage to generate optimised, competitive bids. A study in Germany demonstrated algorithms that can determine optimal pricing and operational strategies, enabling EPC firms to craft competitive bids that maximize value for both developers and operators. This capability allows EPC firms to respond to opportunities with increased speed and accuracy, providing a significant advantage as utilities and major corporations increase their demand for clean electricity to power the growth of data centers and other commercial facilities.

Beyond the bidding process, automation is revolutionizing the engineering and design workflow. Web-based platforms and digital twins allow engineers to create dynamic, data-rich models of BESS projects. These virtual environments facilitate real-time simulations and performance analysis under various conditions, reducing the need for costly and time-intensive manual generation of design alternatives. Engineers and researchers frequently use such simulations to test optimal component sizing and dispatch strategies for various applications, from industrial peak shaving to residential use, refining system designs long before deployment and ensuring that the final layout is optimised for performance, reliability, and cost.

The value of a well-designed BESS is ultimately proven through its market performance. An optimally sized and configured system can more effectively participate in energy markets, providing opportunity for ancillary services such as frequency regulation and energy arbitrage. Artificial intelligence (AI) methods, particularly deep reinforcement learning, have emerged as a state-of-the-art approach for optimizing energy arbitrage, allowing BESS to learn the best trading strategies from market data, while accounting for physical constraints such as battery degradation. Benchmark results from a 2020 study, based on historical UK wholesale electricity prices, show that a proposed deep reinforcement learning method improves profits from energy arbitrage by 58.5% compared with the standard mixed integer linear programming method.

Ultimately, these digital capabilities have powerful implications for the broader energy transition. By streamlining workflows and de-risking projects, AI-enabled automation tools significantly accelerate the deployment of BESS solutions. Faster, more efficient installation of energy storage means more renewable energy can be integrated into the grid without compromising its reliability. This creates a virtuous cycle where the growth of renewables and BESS mutually reinforce each other.

While challenges in data integration and model accuracy remain, the benefits of automated BESS design are clear. For EPC firms, automation provides a critical competitive advantage. For developers and operators, it unlocks greater value from renewable energy assets. For policymakers and the broader public, these digital tools provide an easily overlooked but powerful avenue to accelerate the transition to a cleaner, more resilient energy system.

Note: All opinions expressed in this paper belong to the authors and do not necessarily reflect the policies and views of the U.S. Department of Energy.

Zakaria Hsain

Ph.D. Mechanical Engineering and Applied Mechanics

Zakaria Hsain holds a Ph.D. in mechanical engineering and applied mechanics from Penn and is a former Kleinman Center research assistant. He is a fellow at the Hydrogen and Fuel Cell Technologies Office at the U.S. DOE.

Houssame E. Hsain

Houssame E. Hsain is a Ph.D. Candidate at the ARC Centre for Next-Gen Architectural Manufacturing, University of New South Wales.