Adaptive Smoothing Optimization Method for Dynamic Load Shedding Based on CVaR under Photovoltaic Uncertainty

The integration of renewable energy like photo-voltaic (PV) introduces uncertainty and variability to power grids, jeopardizing stability. In order to optimize dynamic load shedding and lessen such risks, this research offers a conditional value-at-risk (CVaR) model under PV uncertainty. To improve the computation efficiency, we then propose an adaptive smoothing method (ASM) to solve the CVaR model by smoothing maximal function. Extensive experiments on the IEEE-RTS79 system show that the proposed ASM method solves the CVaR model more efficiently than the traditional linear programming method (LPM) and direction solution method (DSM). In terms of reliability assessment indicators, the ASM method significantly outperforms the LPM method and the DSM method. Overall, the CVaR model can be successfully solved using the proposed ASM method based on smoothing technique to identify the best dynamic load shedding strategy under PV uncertainty.