On the problem of controlling shiftable prosumer devices with price signals

Price-based control approaches can be derived from the Lagrangian relaxation of the unit commitment problem and provide a scalable framework for solving practical sized optimization problems. The framework postulates an iterative gradient-based pricing mechanism, in which the price signal is used to negotiate the energy exchange between individual (small sized) market participants. Within this framework shiftable loads, renewables and storage devices induce discontinuous utility functions, which can cause divergence of the pricing mechanism resulting suboptimal solutions.The authors propose a new simple but efficient method to deal with this problem in large scale market setups by introducing the concept of randomized price offsets.