A Second Paper in IEEE Transactions to End The Year! 🥳

Can we combine Reinforcement Learning and Optimization to make smarter and more reliable decisions in uncertain environments? Yes, we can!

In our latest paper — published in IEEE Transactions on Energy Markets, Policy and Regulation — we introduce a novel Safe RL–Optimization framework. It optimizes bidding strategies in the highly volatile and near real-time Imbalance Settlement. Our method separates learning unknown economics from enforcing known asset physical constraint (e.g. a battery energy storage system in our case). But both are integrated within a single, end-to-end, trainable pipeline. Our framework ensures feasible decisions, increases profits, and significantly accelerates convergence during training.

📄 Read the full paper here.