Abstract
Understanding the feasible power flow region is of central importance to power system analysis. In this paper, we propose a geometric view of the power system loadability problem. By using rectangular coordinates for complex voltages, we provide an integrated geometric understanding of active and reactive power flow equations on loadability boundaries. Based on such an understanding, we develop a linear programming framework to 1) verify if an operating point is on the loadability boundary, 2) compute the margin of an operating point to the loadability boundary, and 3) calculate a loadability boundary point of any direction. The proposed method is computationally more efficient than existing methods since it does not require solving nonlinear optimization problems or calculating the eigenvalues of the power flow Jacobian. Standard IEEE test cases demonstrate the capability of the new method compared to the current state-of-the-art methods.
Introduction
In power systems, "loadability" defines the safe operating envelope of the grid. If power demand exceeds certain physical and mathematical limits, the system faces voltage collapse or the power flow equations simply fail to have a solution. Traditionally, engineers identify these limits using "continuation power flow" (CPF)—gradually increasing load until numerical stability is lost—or by checking the power flow Jacobian matrix for singularity.
However, these traditional methods face significant challenges. CPF requires solving successive nonlinear optimization problems, which is computationally prohibitive for large-scale grids. Furthermore, a singular Jacobian is not always a definitive indicator of the loadability limit; it can sometimes signal "cusp bifurcation" points that reside safely inside the feasible region, leading to false alarms and overly conservative grid operations.
The Geometric Circle Insight
This research introduces a novel geometric framework by shifting from standard polar coordinates (magnitude and angle) to rectangular coordinates (real and imaginary parts) for complex voltages. This transformation turns complex, nonlinear power flow equations into quadratic forms.
A key insight of this work is that for any given bus, the active ($P$) and reactive ($Q$) power equations describe two intersecting circles in the voltage plane. As shown in Figure 1, the stability limit—often referred to as the "nose" of the curve—corresponds to the geometric condition where these circles become tangent or fail to intersect. This unified view allows for a clearer understanding of how both active and reactive power contribute to voltage instability.
Loadability Boundary and the Pareto-Front
The authors redefine the loadability boundary as a Pareto-Front: an operating point is on the boundary if it is impossible to increase the load at one bus without decreasing it at another. This definition highlights why traditional Jacobian singularity tests can be misleading.
As illustrated in Figure 3, points where the Jacobian is singular (represented by red lines) can exist both on the boundary and deep within the feasible region (such as point D). Relying solely on the Jacobian can lead to "false positives" for instability, whereas the proposed geometric method accurately identifies the true boundary of the safe operating zone (the blue region).
Linear Programming Framework
Because the gradients of these "power circles" are linear when expressed in rectangular coordinates, the authors developed a Linear Programming (LP) framework. This allows the system to verify boundary conditions, calculate stability margins, and locate specific boundary points using robust, high-speed LP solvers. This approach linearizes the test for stability limits perfectly without losing the inherent nonlinearity of the underlying power flow problem.
Key Results and Performance
The LP-based method was validated using standard IEEE benchmarks and large-scale MATPOWER test cases. The computational speed is remarkable: even for a massive 13,659-bus system, the boundary condition can be checked in approximately 281 seconds on a standard laptop. Smaller systems, such as the 118-bus network, are analyzed in under 4 seconds.
Beyond speed, the method reduces operational conservatism. Compared to widely used "Thevenin-equivalent" margin methods, the geometric approach provides a more accurate and often larger stability margin when the system is far from its limit. This accuracy could allow grid operators to utilize existing infrastructure more efficiently while maintaining safety. The framework also easily incorporates practical constraints, such as generator reactive power limits, by adding linear inequalities to the LP.
Limitations and Future Work
While the current method is highly effective for distribution contexts where generators are modeled as PQ or slack buses, future work will explore more complex "PV bus" modeling for traditional large-scale generators. Additionally, this analysis focuses on static loadability (steady-state solutions) and does not yet account for fast-acting transient dynamics. The authors suggest that this rectangular coordinate geometric perspective could be extended to other critical areas, including optimal power flow (OPF) and state estimation.
Summary updated