Physics-Informed Neural Network-Based Control for Grid-Forming Converter's Stability Under Overload Conditions

TL;DR

The paper tackles the stability of grid-forming converters (GFCs) under overload, where DC source saturation and AC current limits compromise switch safety and system frequency/voltage. It introduces a physics-informed neural network (PINN) that replaces the AC voltage control, AC current limitation, and AC current control while preserving the droop-based frequency response, enabling robust islanding and resynchronization in inertia-less microgrids. The PINN is trained with a physics-informed loss that enforces voltage dynamics, current constraints, and ROCOF/ROCOV limits, achieving improved performance: post-disturbance frequency gains of up to $0.245\ \text{Hz}$ and active-power gains of up to $0.03\ \text{pu}$, a peak transient voltage deviation reduction of $24.14\%$, and ROCOF reduction from $0.02$ to $0.005\ \text{Hz/s}$. Tested on a modified IEEE 13-Bus microgrid with four $500\ \text{kVA}$ GFCs, the approach allows stable operation up to $6.197\ \text{MVA}$ loads, significantly enhancing resilience under sudden overloads compared to conventional droop and current-limiting strategies.

Abstract

Grid-forming converters (GFCs) are crucial for frequency and voltage stability in modern power systems. However, their performance under overload conditions remains a challenge. This paper highlights the limitations of existing approaches in managing DC source saturation and AC current limits, emphasizing the need for improved control strategies to ensure system stability. This paper proposes a control strategy based on a physics-informed neural network (PINN) to improve GFC performance under overloaded conditions, effectively preventing switch failures and mitigating DC source saturation. This approach outperforms conventional methods by maintaining stable voltage and frequency, even under significant load increase where traditional droop control alone proves inadequate. The post-disturbance operating point of GFCs remains unchanged using PINN-based control with an improvement of 0.245 Hz in frequency and 0.03 p.u. in active power when compared to an already existing current limitation strategy. Additionally, it reduces peak voltage deviations during transients by 24.14\%, lowers the rate of change of frequency (ROCOF) from 0.02 Hz/s to 0.005 Hz/s, and improves the rate of change of voltage (ROCOV), keeping both within acceptable limits. These improvements significantly enhance system resilience, especially in inertia-less power networks.

Figures (20)

• Figure 1: Per phase equivalent model of the converter in $\alpha\beta$ coordinates
• Figure 2: Block Diagram of GFC Considered
• Figure 3: Droop Control Block Diagram
• Figure 4: Block diagram of proposed PINN-based GFC control
• Figure 5: Modified IEEE 13-Bus Test Feeder