Distance characteristics for incremental quantities

1. Problem Statement

Traditional distance protection relays rely on apparent impedance calculations ($Z = V/I$) to detect faults. However, in modern power grids with high penetration of Inverter-Based Resources (IBRs), these schemes face significant challenges:

• Source Uncertainty: IBRs behave differently from traditional Synchronous Generators (SGs). Their current output can change rapidly and unpredictably during faults (e.g., entering current-limiting modes), making the "remote current" (current flowing from the rest of the grid into the fault) highly uncertain.
• Operating Point Dependence: Traditional relay characteristics often depend on the real-time voltage and current injections of the network, which vary with load and generation.
• Goal: The authors aim to derive distance relay characteristics using incremental quantities (changes in voltage and current) that are operating-point independent. This means the relay's decision logic should depend only on the network structure and source types, not on the specific real-time load or generation levels at the moment of the fault.

2. Methodology

The paper develops a mathematical framework in the phasor domain to model faults and derive relay characteristics.

A. Incremental Quantities

The authors define incremental voltage ($\tilde{v}$) and current ($\tilde{i}$) by subtracting phasors from a previous cycle ($t - \delta$) from the current cycle ($t$):
$$\tilde{v}_k = v_k(t) - v_k(t - \delta)$$
$$\tilde{i}_k = i_k(t) - i_k(t - \delta)$$
By assuming sources remain stationary over this short interval (Assumption 1), the pre-fault load components cancel out.

B. Network Modeling

• Sources: Synchronous Generators (SGs) are modeled as voltage sources; IBRs are modeled as current sources in parallel with impedances (Norton equivalents).
• Fault Modeling: The network is represented by a bus admittance matrix $Y(m_T)$, where $m_T$ is the normalized fault location ($0$ to $1$) and $m_F$ is the fault resistance factor.
• Derivation: By subtracting the pre-fault network equations from the post-fault equations, the authors derive an expression for the incremental remote current ($\sigma_\eta$). Crucially, this derivation shows that $\sigma_\eta$ depends on:
  1. Pre-fault measurements ($v_L(t-\delta), i_L(t-\delta)$).
  2. Network structure (admittance matrix).
  3. Fault location and resistance.
     It does not depend on the real-time operating point (post-fault source injections).

C. Apparent Impedance Formulation

The authors express the apparent impedance ($z_A$) seen by the relay in terms of these incremental quantities. For both line-to-ground and line-to-line faults, the apparent impedance is shown to be a function of the fault location ($m_T$), fault resistance ($m_F$), and the calculated incremental remote current.

D. Handling Nonlinearity and Uncertainty

Because the incremental remote current depends nonlinearly on the fault location and resistance, the set of all possible apparent impedances ($Z_\eta$) does not have a simple geometric shape (like a circle or polygon). To make this practical for relay implementation, the authors propose two approximations:

1. Point Estimate Approximation: The remote current is estimated at a nominal fault location/resistance (e.g., $m_T=0.5, m_F=1$). The resulting characteristic is a parallelogram (the Minkowski sum of two line segments).
2. Convex Hull Approximation: The apparent impedance is calculated for a grid of discrete fault locations and resistances. The convex hull of these points forms the relay characteristic. This method is computationally tractable (e.g., using the gift-wrapping algorithm).

3. Key Contributions

• Operating-Point Independence: The paper rigorously proves that distance relay characteristics based on incremental quantities are independent of the network's real-time operating point, eliminating uncertainty regarding remote source injections.
• IBR Compatibility: The methodology is specifically tailored for grids with IBRs, addressing the limitations of traditional protection schemes in inverter-dominated networks.
• Geometric Approximations: The authors provide two practical methods (parallelogram and convex hull) to approximate the complex, non-linear set of possible fault impedances, making the theory implementable in real-world relays.
• Unified Framework: It extends previous time-domain incremental protection concepts (referenced from [1]) into the phasor domain, allowing for standard distance relay analysis.

4. Results

The authors validated their approach using a test system from previous work [4]:

• Accuracy: The Convex Hull approximation matched the exact theoretical characteristics ($Z_\eta$) almost perfectly, with only minor non-convexity observed at the top edge (which was negligible).
• Computational Efficiency:
  • The Convex Hull calculation took approximately 3 milliseconds.
  • The Point Estimate (parallelogram) calculation took under 1 millisecond.
• Performance: The point estimate was found to be reasonably close to the convex hull, suggesting that for faster computation, the simpler parallelogram model may be sufficient in many scenarios.

5. Significance and Future Work

• Grid Resilience: As power grids transition toward renewable energy (IBRs), traditional protection schemes are becoming less reliable due to the lack of inertia and changing fault current characteristics. This paper provides a theoretical foundation for robust, adaptive distance protection that remains effective regardless of load or generation levels.
• Scalability: The ability to compute relay characteristics in milliseconds suggests these methods can be integrated into digital relays without significant latency.
• Limitations & Future Work: The authors note that Assumption 1 (sources remain stationary) is less valid for IBRs that rapidly switch to current-limiting modes. Future work will focus on incorporating uncertainty bounds for IBR current changes, which would increase the dimensionality of the problem but further enhance robustness.

In summary, this paper offers a mathematically rigorous and computationally efficient solution for next-generation distance protection, ensuring grid stability in the era of high renewable energy penetration.