Electricity Price and Demand Forecasting Under Smart Grid Environment and Demand Response

  • Dina Masri

Student thesis: Master's Thesis


Demand Response (DR) is widely seen as one of the key enabling technologies for smart grids. To enable consumers to change their consumption behavior in an effective manner, electricity price and demand forecasting become a critical task. Various forecasting techniques have been proposed which aim to minimize forecasting errors in both demand and price in a smart grid environment. However, the effect that DR would have on the dynamics and volatility of demand and price time series has not been investigated. In this thesis, the effect of market structure, DR program types (Real-Time vs non Real-Time), and forecasting tools, considering DR, on the demand and price predictability is analyzed. Three different electricity markets; Australian National Electricity Market (NEM), PJM Independent System Operator (ISO), and Ontario Independent Energy System Operator (IESO) which have significant DR participation are considered. Furthermore, three different forecasting techniques, namely; Least Squares-Support Vector Machine (LSSVM), Extreme Learning Machine (ELM), and Random Forests (RF) are implemented to comprehensively study the effect of DR. Peak to off-peak qualitative analysis is performed and the trend in price and demand forecasting errors before and after the introduction of large scale DR programs is quantified.
Date of AwardMay 2015
Original languageAmerican English
SupervisorHatem Zein El din (Supervisor)


  • Smart Grids
  • Electricity Price
  • Demand Forecasting
  • Support Vector Machines.

Cite this