Setpoint Tracking with Partially Observed Loads
This work addresses setpoint tracking for energy management systems, but it is incremental as it applies known OCO methods to specific feedback scenarios in load control.
The paper tackled the problem of setpoint tracking with uncertain, flexible loads using online convex optimization, achieving sublinear regret bounds under various feedback types including full, bandit, partial bandit, and Bernoulli feedback.
We use online convex optimization (OCO) for setpoint tracking with uncertain, flexible loads. We consider full feedback from the loads, bandit feedback, and two intermediate types of feedback: partial bandit where a subset of the loads are individually observed and the rest are observed in aggregate, and Bernoulli feedback where in each round the aggregator receives either full or bandit feedback according to a known probability. We give sublinear regret bounds in all cases. We numerically evaluate our algorithms on examples with thermostatically controlled loads and electric vehicles.