AI for Energy Storage

We will discuss how new computational approaches enabled by high-performance computing and machine learning algorithms are accelerating the traditional materials design and commercialization process for battery materials

In this talk, I will share insights from the Center for Ageing, Reliability, and Lifetime Prediction of Electrochemical and Power Electronic Systems (CARL) at RWTH Aachen University, Germany, with a focus on Europe's approach to the digitalization of battery technology with AI. By integrating physics-based models with machine learning, our research is advancing intelligent, data-driven strategies for battery testing and real-world field applications. This talk will offer a unique perspective on how AI and digitalization are shaping the future of energy storage systems, with an emphasis on reliability, performance, and sustainability.

The development of next-generation batteries is bottlenecked by slow, manual scientific workflows. This presentation explores how AI can fundamentally transform how we characterize, design, and optimize battery materials. Drawing on real industrial case studies, we’ll demonstrate how Polaron’s platform combines automated characterization, AI reconstruction, and intelligent design tools to deliver faster insights and higher-performing products—closing the loop between idea, lab, and factory.

This presentation will explore the current state of battery intelligence in large-scale energy storage systems. As transmission-connection energy storage systems regularly top 1 GWh, grids worldwide depend more than ever on reliable performance. Battery intelligence is key: from controls that adapt to state-of-health, to dynamic degradation prediction, to multifaceted balance, and state-of-charge management, to mitigating cyber threats.

BESS sites are rich in time-series telemetry, but many fleets still rely on static thresholds and reactive maintenance. This talk presents an asset-agnostic approach to evolve from rule-based alerts to probabilistic forecasting that estimates the likelihood of abnormal behavior over future horizons and probability of survival.

Saft is pleased to present, the Intensium Sight (I-Sight) technology: a new, powerful hardware and software tool. I-Sight is an AI-powered cloud platform for energy storage system monitoring and optimization. I-Sight integrates with Saft’s CUBE Battery Management System to deliver predictive maintenance, real-time analytics, and secure data management. With seamless integration, digital twin capabilities, and automated performance tracking, I-Sight enhances system availability, safeguards asset economics, and supports scalable, future-proof energy storage architectures.

Increasing electrification—data centers, e-mobility, industrial electrification—are straining traditional electric grid access approaches. Growing interconnection queues, both for loads and generators, drive the need for energy hubs, which optimize asset use with business models for managing increasing amounts of data. Advanced energy hub analytics, paired with AI models on both sides of the utility meter, can reduce costs across project lifecycle, from planning to building to operations to maintenance.

As an MIT study recently highlighted, the hype around AI has failed to materialize measurable returns, with a 95% failure rate. We’ll review a range of the battery industry's ambitious claims and aspirations around AI applications, including proof-of-concepts that are failing to deliver. We’ll then lay out the data building blocks needed to enable scalable AI. Finally we’ll show real-world battery AI delivering measurable returns in production with customers.

Electric vehicles are emerging as a pivotal solution for decarbonizing the transportation sector, the largest contributor to US emissions. Over 70% of EV owners charge immediately after reaching home, missing times when electricity is cheaper and the grid is cleaner, even when their vehicles are already plugged in during these optimal hours. This represents a significant opportunity for OEMs like Rivian to unlock substantial cost savings and emission reductions from EV charging at scale. This presentation explores how Rivian is leveraging AI to transform EVs from potential grid liabilities into intelligent energy assets that benefit owners, utilities, and the environment.

The automotive industry is experiencing unprecedented recalls for problems with its lithium-ion batteries for electric and hybrid vehicles. Automakers are spending billions of dollars to transition to cleaner and greener battery-powered vehicles, but the new technology has come with an even steeper cost: reputation-damaging vehicle fires, recalls, sudden power loss and problems starting the cars.  The fire incidents are due to the infrequent presence of manufacturing defects and impurities in singular battery cells. Manufacturers cannot find the defects and contaminations during production without conducting exhaustive tests requiring specialized equipment and time-consuming experiments.  EECOMOBILITY has created a Rapid Battery Test system for production lines - EECOPower - that is extremely sensitive and can detect defective cells quickly. Test results from the application of EECOPower for defect detection in battery cell and module manufacturing are presented.

In this lecture, I will present our research on AI-driven virtual and mixed-reality platforms to enhance decision-making and training for battery manufacturing operators and students. Examples showcase a battery manufacturing metaverse, where geographically dispersed students collaboratively troubleshoot problems, and specific mixed-reality applications for real-time guidance. This work highlights how immersive AI technologies can optimize both operational efficiency and educational outcomes in the battery sector.

We show how Solid Power integrates multiscale machine learning (ML), informatics, and physics-guided surrogates to design efficient solid-state electrolytes, linking atomistic structure to process conditions and cell metrics. We demonstrate ultra-early predictions, intelligent design of experiments, and robust optimization pipelines that speed materials design, streamline cell integration, and improve manufacturing scale-up to deliver a faster, more effective path to market for solid-state electrolytes.

Next-generation batteries require intelligent, adaptive manufacturing systems to scale ceramic-based architectures and meet demands for high energy density and safety. Innovative developers use AI to optimize processes, enabling high-throughput and predictive analytics. The session will detail how image-based deep learning models detect product defects in ceramic separators. These robust machine learning pipelines are scaled to optimize yields, ensure safety and reliability, and accelerate defect-free solid-state battery manufacturing.