Two-Stage Decoupled Estimation Approach of Aggregated Baseline Load Under High Penetration of Behind-the-Meter PV System
Kangping Li, Jichuan Yan, Lin Hu, Fei Wang, Ning Zhang
Abstract
Accurate aggregated baseline load (ABL) estimation is critical for the implementation of incentive-based demand response (DR). The increasing penetration of invisible behind-the-meter photovoltaic (BTM-PV) systems makes the net load less predictable, thus posing a significant challenge to ABL estimation. Current direct estimation approaches will produce large errors when the weather changes suddenly on DR event day, which is very likely to happen especially under high PV penetration scenarios. This paper proposes a two-stage decoupled estimation approach to improve the accuracy of ABL estimation under BTM-PV penetration. In <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Stage I</i> , an optimal PV-Load decoupling method is proposed to decouple the PV power and the actual customer load from the net load only using a small number of observable PVs power and smart meter data. In <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Stage II</i> , PV power and customer load within the DR event are individually estimated. Particularly, considering ABL is an ex-post estimation, a weather classification-based PV power estimation method is proposed to fully utilize the information not only before but also after the DR event to further improve the accuracy. The effectiveness and superiority of the proposed approach have been verified using a real smart meter dataset.