A Reinforcement Learning-Based Decision System For Electricity Pricing Plan Selection by Smart Grid End Users

Tianguang Lu, Xinyu Chen, Michael B. Mcelroy, Chris P. Nielsen, Qiuwei Wu, Qian Ai

Research output: Contribution to journalJournal articleResearchpeer-review

24 Downloads (Pure)

Abstract

With the development of deregulated retail power markets, it is possible for end users equipped with smart meters and controllers to optimize their consumption cost portfolios by choosing various pricing plans from different retail electricity companies. This paper proposes a reinforcement learning-based decision system for assisting the selection of electricity pricing plans, which can minimize the electricity payment and consumption dissatisfaction for individual smart grid end user. The decision problem is modeled as a transition probability-free Markov decision process (MDP) with improved state framework. The proposed problem is solved using a Kernel approximator-integrated batch Q-learning algorithm, where some modifications of sampling and data representation are made to improve the computational and prediction performance. The proposed algorithm can extract the hidden features behind the time-varying pricing plans from a continuous high-dimensional state space. Case studies are based on data from real-world historical pricing plans and the optimal decision policy is learned without a priori information about the market environment. Results of several experiments demonstrate that the proposed decision model
can construct a precise predictive policy for individual user, effectively reducing their cost and energy consumption dissatisfaction.
Original languageEnglish
JournalI E E E Transactions on Smart Grid
Number of pages10
ISSN1949-3053
Publication statusAccepted/In press - 2021

Keywords

  • Smart grid end user
  • Decision system
  • Electricity market
  • Value-based Q learning
  • Demand response

Fingerprint Dive into the research topics of 'A Reinforcement Learning-Based Decision System For Electricity Pricing Plan Selection by Smart Grid End Users'. Together they form a unique fingerprint.

Cite this