Applied Sciences (May 2022)

Blockchain Based Smart-Grid Stackelberg Model for Electricity Trading and Price Forecasting Using Reinforcement Learning

  • Md Mahraj Murshalin Al Moti,
  • Rafsan Shartaj Uddin,
  • Md. Abdul Hai,
  • Tanzim Bin Saleh,
  • Md. Golam Rabiul Alam,
  • Mohammad Mehedi Hassan,
  • Md. Rafiul Hassan

Journal volume & issue
Vol. 12, no. 10
p. 5144


Read online

A smart grid is an intelligent electricity network that allows efficient electricity distribution from the source to consumers through telecommunication technology. The legacy smart grid follows the centralized oligopoly marketplace for electricity trading. This research proposes a blockchain-based electricity marketplace for the smart grid environment to introduce a decentralized ledger in the electricity market for enabling trust and traceability among the stakeholders. The electricity prices in the smart grid are dynamic in nature. Therefore, price forecasting in smart grids has paramount importance for the service providers to ensure service level agreement and also to maximize profit. This research introduced a Stackelberg model-based dynamic retail price forecasting of electricity in a smart grid. The Stackelberg model considered two-stage pricing between electricity producers to retailers and retailers to customers. To enable adaptive and dynamic price forecasting, reinforcement learning is used. Reinforcement learning provides an optimal price forecasting strategy through the online learning process. The use of blockchain will connect the service providers and consumers in a more secure transaction environment. It will help tackle the centralized system’s vulnerability by performing transactions through customers’ smart contracts. Thus, the integration of blockchain will not only make the smart grid system more secure, but also price forecasting with reinforcement learning will make it more optimized and scalable.