Title: A Novel Linear Power Flow Model

Abstract: Linear power flow (LPF) models are important for the solution of large-scale problems in power system analysis. This paper proposes a novel LPF method named data-based LPF (DB-LPF). The DB-LPF is distinct from the data-driven LPF (DD-LPF) model because the DB-LPF formulates the definition set first and then discretizes the set into representative data samples, while the DD-LPF directly mines the variable mappings from historical or measurement data. In this paper, the concept of LPF definition set (i.e., the set where the LPF performs well) is proposed and an analytical algorithm is provided to get the set. Meanwhile, a novel form of AC-PF models is provided, which is helpful in deriving the analytical algorithm and directing the formulations of LPF models. The definition set is discretized by a grid sampling approach and the obtained samples are processed by the least-squares method to formulate the DB-LPF model. Moreover, the model for obtaining the error bound of the DB-LPF is proposed, and the network losses of the DB-LPF is also analyzed. Finally, the DB-LPF model is tested on enormous cases, whose branch parameters are also anal-yzed. The test results verify the effectiveness and superiority of the proposed method.