Loading...

A Reinforcement Learning Approach for Dynamic Pricing (Case Study: Iran Electrical Power Grid)

Nakhaei, Hamid | 2023

53 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56394 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Shadrokh Sikari, Shahram
  7. Abstract:
  8. Population growth and the escalating use of electric devices have led to a surge in electricity demand. However, the construction of adequate infrastructure to meet this additional demand is highly expensive. To address this issue that sometimes causes power outages, dynamic pricing for electricity, a demand management technique, has gained significant importance. By employing this method, prices can be adjusted on an hourly basis, enabling consumers to regulate their consumption in response to these price fluctuations. The prices can be determined based on different objective functions. This research aims to develop a comprehensive framework for determining the optimal hourly electricity retail prices. The primary objectives of this framework are to minimize household costs and maximize the profit of electricity retailers. First, the wholesale electricity price for each day of the year is predicted using historical data and the SARIMA method; the selection of the model is such that the residuals are as close to the white noise as possible. Prior to prediction, the non-stationarity of the data was fixed. Subsequently, the Q-learning algorithm is employed to determine the optimal hourly retail prices for the following day. This process is then applied to Iran's electricity data in 1400 as a case study, and the results are analyzed. Additionally, a sensitivity analysis is conducted on the balancing factor for retailers' profit and household costs. The results demonstrate the effectiveness of the proposed pricing framework in reducing annual peak demand and total consumption by 17.1% and 19.4%, respectively. Furthermore, the increase in household costs not only safeguards the profitability of retailers but also enables them to achieve substantial gains
  9. Keywords:
  10. Dynamic Pricing ; Reinforcement Learning ; Demand Response ; Forecasting ; Nonstationarity ; Demand Management ; Iran Electrical Power Grid

 Digital Object List

 Bookmark

No TOC