Optimization of Fault Current Limiter Reactance Based on Joint Simulation and Penalty Function-Constrained Algorithm

Jun Zhao, Chao Xing, Zhigang Zhang, Boyuan Liang, Lu Sun, Bin Wei, Weiqi Qin, Shuguo GaoEnergies 2025 · DOI

TL;DR

This paper introduces a novel method for optimizing the reactance of fault current limiters (FCLs) in large power grids. By combining joint simulation with MATLAB and ATP-EMTP, a penalty function method, and the Gravitational Search Algorithm (GSA), the approach efficiently finds optimal FCL settings while ensuring system stability and avoiding overvoltage issues. This method significantly reduces computational complexity and improves efficiency compared to traditional approaches.

What Problem Does This Paper Address?

The increasing expansion of power grids, particularly in regions like China, has led to a significant rise in short-circuit currents. These elevated currents pose a serious threat to the stability and reliability of the power system, potentially damaging critical equipment like transformers. While fault current limiters (FCLs) are a solution, traditional methods for determining their optimal reactance values are inefficient and time-consuming. This involves extensive trial-and-error calculations to satisfy multiple constraints, including short-circuit current limits and overvoltage protection, especially in complex, large-scale grids with numerous variables. The paper aims to address this by developing a more efficient and robust optimization method for FCL reactance configuration.

Methods

The proposed method integrates several key components:

  • Joint Simulation (MATLAB and ATP-EMTP): This is a crucial step for accurately deriving the constraint conditions for the optimization problem. MATLAB is used for control system integration and low-frequency dynamics, while ATP-EMTP excels at high-frequency electromagnetic transient analysis, particularly for overvoltage calculations. This dual-tool approach leverages the strengths of each to provide precise data support for the optimization.
  • Penalty Function Method (PFM): To tackle the challenge of constrained optimization, PFM is employed. It transforms the original problem with multiple constraints (short-circuit current and overvoltage) into an unconstrained optimization problem by adding penalty terms to the objective function. This significantly reduces computational complexity and ensures the feasibility of the solutions.
  • Gravitational Search Algorithm (GSA): GSA, a global optimization technique inspired by Newton's law of universal gravitation, is used to find the optimal reactance value. GSA is chosen for its strong global search capabilities, which help prevent entrapment in local optima, making it suitable for complex, multi-modal optimization problems.
  • Adaptive Dynamic Adjustment of Penalty Parameters: To further enhance GSA's performance, the penalty parameters are dynamically adjusted during the iterative process. This helps balance exploration and exploitation, ensuring effective constraint satisfaction without sacrificing convergence speed.

Datasets

Not discussed in the paper. The study utilizes a simulated 220 kV substation in the Hebei power grid for case studies.

Key Insights and Ideas

  • Synergy of Simulation Tools: Combining MATLAB and ATP-EMTP is essential for accurately capturing both low-frequency and high-frequency power system dynamics, which is critical for FCL optimization.
  • PFM for Constraint Handling: Transforming a complex constrained optimization problem into an unconstrained one via PFM is a powerful strategy for reducing computational burden and ensuring feasible solutions.
  • GSA's Global Search Prowess: GSA demonstrates superior performance in finding the global optimum, especially in scenarios with multiple FSFCLs, where the optimization landscape can become multimodal.
  • Adaptive Penalty for Robustness: Dynamically adjusting penalty parameters in GSA leads to a more robust optimization process, balancing constraint satisfaction with convergence efficiency.
  • Medium-Voltage Side Installation: The study suggests that installing FCLs at the medium-voltage side's neutral point is more effective for limiting short-circuit currents and avoiding potential safety hazards compared to the high-voltage side.

Key Results

  • The proposed method successfully optimized the fault current limiter reactance to 15.9 mH for the case study transformer.
  • In a single reactance optimization scenario, GSA achieved convergence in an average of 83 iterations, comparable to GA (33 iterations) but with a more robust global search capability.
  • In a multi-reactance optimization scenario (three FSFCLs at neutral points and five on feeder lines), GSA achieved a 92% global optimal solution hit rate, significantly outperforming GA (78%) and PSO (67%).
  • The optimized FCL settings met both short-circuit current targets and overvoltage constraints, with maximum transient overvoltage at the medium-voltage side neutral point being 74.6 kV and steady-state overvoltage being 49.8 kV, well below the insulation limits.
  • The method significantly improves computational efficiency compared to traditional trial-and-error approaches.

Limitations and Future Work

  • Computational Cost of Joint Simulation: While the PFM and GSA reduce optimization complexity, the joint simulation itself can still be computationally intensive, especially for very large and complex grids.
  • Parameter Tuning: The performance of GSA is sensitive to its parameters (e.g., gravitational constant, population size, maximum iterations). While the paper discusses some parameter settings, further research into adaptive parameter tuning could be beneficial.
  • Real-world Validation: The study relies on simulation-based case studies. Future work could involve validating the proposed method with extensive field testing and real-world grid data.
  • Extension to More Complex Scenarios: The paper focuses on specific fault types. Extending the optimization to include a wider range of fault conditions and system configurations would enhance its applicability.
  • Integration with Other FCL Technologies: Exploring the application of this optimization method to other types of FCLs beyond the fast-switching, reactor-based type could be a valuable future direction.

Summary updated

Abstract

This paper proposes a novel optimization method for fault current limiter (FCL) reactance configuration based on joint simulation and penalty function constraint optimization. By integrating MATLAB and ATP for joint simulation, the method accurately derives the constraint conditions of the objective optimization function, providing critical data support for the optimization process. To address the challenges of high computational complexity and solution difficulties in constrained optimization, the Penalty Function Method (PFM) is employed to transform the original constrained optimization problem into a standard unconstrained optimization problem, significantly reducing computational complexity and ensuring the feasibility of the solution. On this basis, the Gravitational Search Algorithm (GSA) is applied to compute the optimal reactance value. Through comparative analysis of engineering case studies, the superiority of the GSA over the Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) in optimization performance is validated, further confirming the accuracy and efficiency of the proposed method. The results indicate that this method not only achieves precise calculation results but also significantly improves computational efficiency. Moreover, the integration of PFM and GSA demonstrates excellent robustness, providing reliable technical support for the optimized deployment of fast-switching fault current limiters in large-scale power grids.
BibTeX
@inproceedings{zhao2025optimization,
  author = {Jun Zhao and Chao Xing and Zhigang Zhang and Boyuan Liang and Lu Sun and Bin Wei and Weiqi Qin and Shuguo Gao},
  title = {{Optimization of Fault Current Limiter Reactance Based on Joint Simulation and Penalty Function-Constrained Algorithm}},
  booktitle = {Energies},
  year = {2025},
  doi = {10.3390/en18051077},
}
Figure