An Efficient Probabilistic Assessment Method for Electricity Market Risk Management

Abstract

Managing electricity market risks is crucial for market participants. For electricity price risk management, expectation and standard deviation of price, along with possible occurrence of price spike, need to be assessed in order to support further risk control. In this paper, a hybrid probabilistic assessment method based on adaptive importance sampling (AIS) and sequential importance sampling (SIS) is developed. Improvements on AIS and SIS make the method better suit electricity market problems. Case studies are conducted on an equivalent Australian National Electricity Market (NEM) system. Uncertainties considered include system load, renewable energy output, generator bidding strategy, and outage rate. The proposed method provides much faster estimation of both normal price and price spike probability, meanwhile achieving comparable accuracy as Monte Carlo (MC) simulation results. Sensitivity of its estimation efficiency against different load level is also analyzed, which shows the robustness of the proposed method.