High-Fidelity Model Order Reduction for Microgrids Stability Assessment

TL;DR

This paper introduces a computationally efficient reduced-order model for inverter-based microgrids that accurately assesses stability boundaries, revealing unique stability factors and limits specific to microgrid configurations.

Contribution

A novel reduced-order modeling approach for microgrid stability that improves computational efficiency and provides new insights into microgrid-specific stability mechanisms.

Findings

Stability limits depend on inverter rating to network capacity ratio.

Shorter lines reduce the stability region in microgrids.

The reduced-order model matches detailed models in accuracy and efficiency.

Abstract

Proper modeling of inverter-based microgrids is crucial for accurate assessment of stability boundaries. It has been recently realized that the stability conditions for such microgrids are significantly different from those known for large- scale power systems. While detailed models are available, they are both computationally expensive and can not provide the insight into the instability mechanisms and factors. In this paper, a computationally efficient and accurate reduced-order model is proposed for modeling the inverter-based microgrids. The main factors affecting microgrid stability are analyzed using the developed reduced-order model and are shown to be unique for the microgrid-based network, which has no direct analogy to large-scale power systems. Particularly, it has been discovered that the stability limits for the conventional droop-based system (omega - P/V - Q) are determined by the ratio of inverter rating to network capacity, leading to a smaller stability region for microgrids with shorter lines. The theoretical derivation has been provided to verify the above investigation based on both the simplified and generalized network configurations. More impor- tantly, the proposed reduced-order model not only maintains the modeling accuracy but also enhances the computation efficiency. Finally, the results are verified with the detailed model via both frequency and time domain analyses.