Chance constrained dynamic optimisation method for AGC units dispatch considering uncertainties of the offshore wind farm

Abstract

The continuing growth of an offshore wind farm integrated into a power grid via high-voltage DC has posed great challenges for automatic generation control (AGC) of the power system. To address these challenges, a new concept of ‘dynamic dispatch of AGC units (DDA)’ under control performance standards from the view of economic dispatch (ED) has been proposed in the previous work, and proved to be an effective technique to co-operate the AGC units with different ramping rates and to fill the gap between ED and AGC. However, the existing DDA model is deterministic in nature, which can hardly deal with the uncertain forecasting error of the offshore wind power output. A novel stochastic DDA model based on chance-constrained programming is proposed considering a random offshore wind power forecasting error, and a hybrid algorithm combining the evolutionary programming algorithm and point estimate method is then developed to solve the stochastic model. Numerical results from a two-area test system with additional offshore wind power generation demonstrate the accuracy and computation efficiency of the hybrid algorithm and the benefits offered by the stochastic AGC dispatch method.