Digital energy solutions

We develop digital algorithms that solve constrained optimization problems such as dispatch, aggregation, and optimal control. These approaches automate operational decisions and generate datadriven recommendations while ensuring robustness and scalability, including in real-time environments. 

Our models also account for asset aging, including state of health (SoH) and battery cycling costs, to optimize both operational performance and equipment lifetime.

We systematically integrate uncertainty management through probabilistic forecasting, including quantiles, confidence intervals, and risk indicators. 

These approaches help anticipate uncertainty related to weather, usage variability, and asset availability. They reduce operational violations and rebound effects while supporting more secure decision-making in complex, distributed energy systems.

Our solutions enable the estimation and coordinated activation of distributed energy resource (DER) flexibility. 

We control buildings, EV charging, and storage systems in real time through constrained optimization to reduce local congestion at substations and transformers, lower peak loads, and strengthen grid resilience. 

Deployments can rely on hybrid cloud-edge architectures with local safeguards to ensure operational continuity and simplify IT/OT integration. 

We deploy advanced predictive maintenance approaches based on multi-sensor time-series analysis. 

Our models detect anomalies, diagnose root causes, and anticipate performance drift to increase reliability and reduce unplanned downtime. 

Modern methods, including zero-shot approaches, support multi-site deployment even when data is incomplete or only lightly labeled.