Hierarchical Distributed Robust Optimization for Demand Response Services

This paper presents a hierarchical robust distributed optimization for day-ahead and intra-day scheduling of the operation of flexible devices (electro-thermal heating units) within a city district. An aggregation service provider, which acts as an aggregator, performs this distributed optimization to maximize the flexibility potential of its customers to provide services to other actors, such as to a balance responsible party or the distribution system operator. Our optimization algorithm is based upon the alternating direction method of multipliers and prioritizes each individual customer and its own private objective. A model predictive control and robust design guarantee that uncertainty, e.g., electrical or thermal demand, is managed within the optimization process. The work includes a customer versus system level objective (aggregator) analysis under uncertainty.