A Statistical-Based Approach for Decentralized Demand Response Toward Primary Frequency Support: A Case Study of Large-Scale 5G Base Stations With Adaptive Droop Control

Abstract

Due to the rapidity required for primary frequency control (PFC), the decentralized responses of distributed energy resources (DERs) are often uncoordinated and cannot be aggregated into a desired outcome. This paper proposes a novel approach to consolidate the decentralized responses of ON/OFF DERs as an organized droop control. For a DER population, a dynamic response threshold can be calculated based on its statistical distribution, enabling DERs to autonomously respond to PFC according to local states. The aggregation of their responses can align with any preset droop curve (linear or non-linear). The approach is demonstrated by a novel DER, 5G base stations (gNBs) and their backup energy storage systems (BESSs), marking the first attempt of gNBs for PFC support, to the authors’ knowledge. In this regard, gNB dormancy model, based on user equipment (UE) access, and dynamic BESS response model, derived from experimental data, are developed. An adaptive droop control scheme is designed to enable gNBs within multiple radio access networks (RANs) to provide differentiated responses based on disturbances degree and RANs busyness level. Simulations based on real device data show that the proposed approach enables decentralized gNBs to perform accurate response, providing PFC support without affecting communication network operation.