Paper Info
Title
Data Driven Approach for Reduced Value at Risk Forecasts in Renewable Power Supply Systems
Abstract
Electricity production from renewable resources such as wind and solar has increased uncertainty in the electricity supply chains. This uncertainty fluctuates the electricity price and consequently causes a highly volatile electricity market and increases the value at risk (VaR) of electricity price. Therefore, production managers need to have an accurate forecast of electricity price as well as VaR before making any plan for further production. In this work, performance of exponentially weighted moving average (EWMA) and recently introduced generalized-EWMA (G-EWMA) are evaluated for VaR forecasting. Both methods have been applied to electricity price dataset of different Canadian provinces. The results of our data analysis show that G-EWMA perform more accurately than EWMA. In addition, we show that Ontario has the highest electricity VaR among other Canadian provinces and this would be the consequence of participation of wind and solar power plants in electricity production system. The electricity market, which is responsible for scheduling electricity buyers and sellers, should also use forecasting tools for matching supply and demand to avoid any sudden change in electricity price. Double exponential smoothing (DES) and triple exponential smoothing (TES) forecasting methods have been used for electricity supply and demand forecast. Our analyses show that DES forecasts of supply/demand outperform TES.
Year
DOI
Venue
2020
10.1109/CCECE47787.2020.9255789
2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)
Keywords
DocType
ISSN
Electricity market,supply,demand,price,volatility,VaR,EWMA,G-EWMA,DES,TES,Backtesting
Conference
0840-7789
ISBN
Citations 
PageRank 
978-1-7281-5443-5
2
0.42
References 
Authors
0
3
Name
Order
Citations
PageRank
Behrouz Banitalebi182.35
Srimantoorao S. Appadoo220.76
A. Thavaneswaran313021.94