Data Driven Linearized AC Power Flow Model with Regression Analysis

TL;DR

This paper introduces a data-driven linearized AC power flow model that improves upon traditional models by incorporating regression analysis, enabling more accurate and computationally efficient power system analysis especially when reactive power and voltage are important.

Contribution

The paper proposes a novel data-driven linearized AC power flow model using regression analysis, enhancing accuracy over traditional linearized models for power system analysis.

Findings

The DLAC model outperforms the regular LAC model in simulations.

Numerical results on TVA system validate the effectiveness of the proposed approach.

The model improves computational efficiency while maintaining accuracy.

Abstract

Full AC power flow model is an accurate mathematical model for representing the physical power systems. In practice, however, the utilization of this model is limited due to the computational complexity associated with its nonlinear and nonconvexity characteristics. An alternative linearized DC power flow model is widely used in the power system operation and planning tools. However, when reactive power and voltage magnitude are of concern, DC power flow model will be useless. Therefore, a linearized AC (LAC) power flow model is needed to address this issue. This paper first introduces a regular LAC model. Subsequently, with the advance in regression analysis technique, a data driven linearized AC (DLAC) model is proposed to improve the regular LAC model. Numerical simulations conducted on the Tennessee Valley Authority (TVA) system demonstrate the performance and effectiveness of the proposed DLAC model.