Estimation of the Shapley Value of a Peer-to-peer Energy Sharing Game Using Multi-Step Coalitional Stratified Sampling

Liyang Han*, Thomas Morstyn, Malcolm Mcculloch

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

One of the main objectives of a peer-to-peer energy market is to efficiently manage distributed energy resources while creating additional financial benefits for the participants. Cooperative game theory offers such a framework, and the Shapley value, a cooperative game payoff allocation based on the participants’ marginal contributions made to the local energy coalition, is shown to be fair and efficient. However, its high computational complexity limits the size of the game. In order to improve this peer-to-peer cooperative scheme’s scalability, this paper investigates and adapts a stratified sampling method for the Shapley value estimation. It then proposes a multi-step sampling strategy to further reduce the computation time by dividing the samples into incremental parts and terminating the sampling process once a certain level of estimation performance is achieved. Finally, selected case studies demonstrate the effectiveness of the proposed method, which is able to scale up the game from 20 players to 100 players.
Original languageEnglish
Pages (from-to)1863–1872
JournalInternational Journal of Control, Automation and Systems
Early online date18 Feb 2021
DOIs
Publication statusPublished - May 2021

Fingerprint

Dive into the research topics of 'Estimation of the Shapley Value of a Peer-to-peer Energy Sharing Game Using Multi-Step Coalitional Stratified Sampling'. Together they form a unique fingerprint.

Cite this