SafePowerGraph-HIL: Real-Time HIL Validation of Heterogeneous GNNs for Bridging Sim-to-Real Gap in Power Grids
Aoxiang Ma, Salah Ghamizi, Jun Cao, Pedro Rodriguez·January 21, 2025
Summary
The SafePowerGraph-HIL framework validates heterogeneous GNNs for power grid state estimation and analysis using real-time HIL simulations on the IEEE 9-bus system. This dataset, generated by Hypersim, enhances HGNN training and accuracy, demonstrating potential for intelligent, adaptive control strategies in evolving power grids. The framework uses a heterogeneous GNN model for state estimation and dynamic analysis, integrated with AWS for scalable storage and processing. It fine-tunes a GNN with 500 sets of data, predicting system parameters like active and reactive powers, bus voltages, and angles, with an MSE loss for training. The model's effectiveness is validated through fine-tuning with real-time data, reducing validation loss for bus voltages and external grid by 13% and 75%, respectively. Future work aims to extend the framework to more complex networks with renewable energy sources.
Introduction
Background
Overview of power grid state estimation and analysis
Importance of real-time simulations in power grid management
Role of Hypersim in generating realistic power grid scenarios
Objective
To present the SafePowerGraph-HIL framework for validating heterogeneous GNNs in power grid applications
To demonstrate the framework's capability in enhancing GNN training and accuracy through real-time HIL simulations
Method
Data Collection
Description of the IEEE 9-bus system used for simulations
Role of Hypersim in generating the dataset
Data Preprocessing
Preprocessing steps for the dataset
Data format and structure for GNN model input
Model Architecture
Overview of the heterogeneous GNN model
Integration of AWS for scalable storage and processing
Training Process
Fine-tuning of the GNN model with 500 sets of data
Objective of the training: predicting system parameters like active and reactive powers, bus voltages, and angles
Use of MSE loss for training
Validation
Validation process using real-time data
Reduction in validation loss for bus voltages and external grid by 13% and 75%, respectively
Results
Performance Metrics
Evaluation of the model's accuracy in state estimation and dynamic analysis
Comparison with baseline models
Scalability and Efficiency
Discussion on the framework's scalability with AWS
Efficiency in handling large datasets and complex power grid networks
Future Work
Extension to Complex Networks
Plans to incorporate more complex networks with renewable energy sources
Challenges and potential solutions
Integration with Real-Time Systems
Potential for integrating the framework with existing power grid control systems
Benefits of real-time state estimation and analysis
Conclusion
Summary of the SafePowerGraph-HIL framework
Implications for power grid management
Call for further research and development
Basic info
papers
machine learning
artificial intelligence
Advanced features
Insights
What are the future plans for expanding the SafePowerGraph-HIL framework?
What improvements does the framework demonstrate in terms of validation loss for bus voltages and external grid?
How does the framework utilize real-time HIL simulations for power grid state estimation and analysis?
What is the main purpose of the SafePowerGraph-HIL framework?