Loading...

Optimal Decision-Making of Household Prosumer and Renewable Producer in the Pool-Based Electricity Market Using Reinforcement Learning Methods

Bagherzadeh, Amir Hossein | 2024

0 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 57619 (46)
  4. University: Sharif University of Technology
  5. Department: Energy Engineering
  6. Advisor(s): Boroushaki, Mehrdad; Talaei, Mohammad
  7. Abstract:
  8. Today, with the expansion of the use of renewable energy and the increase in the production of electrical products, which has resulted in the increase of active consumers in the electricity market, making an optimal decision about the production and consumption of electrical energy has become particularly important. This research examines a sequential decision-making problem in an environment with high uncertainties in which an active household consumer and a renewable producer are simultaneously present in the electricity market. The main goal of this research is to provide an optimal model for making decisions regarding the consumption and production of electric power in the power market and ancillary services market, using reinforcement learning algorithms. The proposed model is developed using sequential decision-making methods and is designed to optimize two parallel objectives, i.e. reducing the costs caused by forecast mismatch and using the flexibility of the household consumer without disturbing the thermal comfort in the building. Since the fluctuations of renewable energy production and sudden changes in demand cause many complications in electricity market decisions, reinforcement learning methods have been used as a powerful tool to manage uncertainties and update decisions according to new conditions. The results of simulations based on real data show that the proposed model is effectively able to improve decision. Also, this model can help to reduce the fines caused by non-compliance with obligations in the electricity market by creating a balance between power production and consumption and using side markets such as the power regulation market. This leads to improving the stability and reliability of the power grid and increasing the economic efficiency of market players
  9. Keywords:
  10. Sequential Decision Making ; Reinforcement Learning ; Demand Response ; Ancillary Service Markets ; Time Series Prediction ; Power Market

 Digital Object List

 Bookmark

No TOC