PROJECT TITLE :

Shapley Value Estimation for Compensation of Participants in Demand Response Programs

ABSTRACT:

Planning truthful compensation mechanisms for demand response (DR) is difficult. This paper models the problem in an exceedingly game theoretic setting and designs a payment distribution mechanism based on the Shapley price (SV). As precise computation of the SV is normally intractable, we propose estimating it using a reinforcement learning algorithm that approximates optimal stratified sampling. We have a tendency to apply this algorithm to a DR program that utilizes the SV for payments and quantify the accuracy of the resulting estimates.