Local Control Strategy for Optimal Power Flow in Low-Voltage Distribution Network Based on Robust Stochastic Optimization

With the promotion of the “dual carbon” goal and the construction of new power systems, the complexity and uncertainty of power systems have increased dramatically, which brings challenges of high proportion of distributed energy and asymmetric loads, such as voltage overstep and three-phase imbalance to distribution network. To cope with these problems, this paper proposes a local control strategy for optimal power flow (OPF) in low voltage distribution network based on robust stochastic optimization (RSO), which uses a 1-norm Wasserstein distance uncertainty set to describe the output uncertainty of distributed energy resource (DER), and builds a robust stochastic optimization model for three-phase four-wire low-voltage distribution network. This model aims to minimize control costs and network losses, while considering the expected adjustments under worst-case scenarios. Local control strategies for distributed energy are obtained without communication infrastructure by convolutional neural networks training. The simulation results verify the effectiveness and economy of the proposed control strategy.