Stochastic unit commitment basec on arima scenario generation and reduction

Abstract

Under high penetration of wind energy, the power system is facing the challenge to maintain the balance between generation and load due to the uncertainty and variability of wind. The traditional unit commitment only consider the deterministic load and wind output in next day and use the operating reserve to handle the net load uncertainty. In this paper, the stochastic unit commitment is developed to address the uncertainty of load and wind output. The time series model Autoregressive Integrated Moving Average (ARIMA) is applied to provide the point estimation and prediction interval for next 24 hours based on the historical load and wind data. The Monte-Carlo simulation and scenario reduction is applied to generate the scenarios and reduce the scenario number. The Benders Decomposition is used to solve the large scale stochastic programming problem in a more efficient manner. A 6 bus system is simulated and studied for the proposed model and algorithm.