An autonomous DG controller using artificial intelligence approach for voltage control

Most recently, solar photovoltaics (PVs) have gained the significant attention due to the considerable reduction in their manufacturing costs as well as the substantial advancements in power electronic converters. However, the widespread integration of rooftop PVs may arise several challenges, such as over-voltages and frequent tap operations. A proper control strategy is required to mitigate these issues. In this paper, a machine learning-based autonomous distributed generator (DG) control is proposed. The controller takes local measurements such as nodal voltage and its sensitivity to changes in load and/or power, and determines the power cap. The controller is trained on different loading conditions to incorporate daily, monthly, and yearly load variations. Simulation results show th at the proposed controller effectively regulates the system voltages as defined by the ANSI C84.1-2016 standard. Moreover, it ensures the fairness among the DGs available at different locations in the distribution system without the need of any communication infrastructure.