Recording of Tuesday, June 23, 2026 | The smarter E Europe Conferences 2026 | Language: English | Duration: 13:34 .
The document emphasizes the integration of adaptive electrothermal control to improve the performance and availability of battery systems. It connects energy management system (EMS) strategies with operational controls at the battery management system (BMS) level. A variety of control techniques, including group control and advanced models like mixed-integer linear programming (MILP) and nonlinear programs, address inverter losses and thermal dynamics. The proposed HVAC control methodology shifts towards model predictive control (MPC) to enhance system responsiveness. Optimization objectives are clarified through a multi-objective framework, addressing the interplay between performance metrics such as round-trip efficiency and thermal derating which affects system availability. Historical simulations show that using mixed-integer nonlinear programming (MINLP) results in significant efficiency gains, and a reinforcement learning (RL) controller demonstrates marked improvements in round-trip efficiency. The RL agent selectively activates battery strings based on operational conditions, optimizing performance by responding to price signals. This indicates a shift toward an energy-aware, data-driven approach in thermal management and battery optimization.
Automated summarization by AI Conver
Speaker
Vivek Tanjavooru
Research Associate
Kempten University of Applied Sciences
Germany
While predictive maintenance has become a well-known concept in battery energy storage, the next step is predictive operation. This approach aims not only to prevent failures, but also to continuously optimize system operation in real time. The goal is to maximize the performance, efficiency and lifetime of the existing hardware in each specific use case. Predictive operation extends beyond the battery cells themselves: Adaptive thermal management, intelligent inverter control and optimal coordination of power electronics all play a crucial role. By anticipating system behavior under different operating scenarios, predictive operation enables asset owners and operators to extract greater value from their storage systems while improving reliability and integration with renewable generation. This session will highlight the latest methods, control strategies and demonstration results that bring predictive operation from research into practice.
Further Talks of this session:
Speaker
Aleksej Krükov
General Manager Overseas Service EMEA
Contemporary Amperex Technology Limited
Speaker
Gaëlle Ryckebusch
Head of Digital Products Strategy - Energy Storage Systems Division
SAFT
France
Speaker
Dr. Raphael Hollinger
Managing Director
The Mobility House Energy GmbH
Germany
Speaker
Dr. Stephan Rohr
CEO
TWAICE
Germany