Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

On the Potential of Digital Twins for Distribution System State Estimation with Randomly Missing Data in Heterogeneous Measurements

Ying Zhang, Yihao Wang, Yuanshuo Zhang, Eric Larson, Di Shi, Fanping Sui

TL;DR

The paper addresses DSSE under realistic data-absence and multi-modal measurement conditions by introducing a digital twin that jointly harnesses physical knowledge, virtual modeling, and data fusion. It presents an end-to-end interactive DT with physics-informed data augmentation, parallel transfer for heterogeneous measurements, and a cross-interaction fusion scheme based on group-queried attention to achieve robust voltage estimation without imputation. The approach is validated on a real unbalanced 84-node distribution network, showing superior accuracy and stability compared with LSTM and Transformer baselines, and demonstrates resilience to random data missing up to 40% of measurements. This work offers a practical, end-to-end DSSE framework that can adapt to varying observability and data-quality scenarios in real distribution grids, reducing error accumulation and improving situational awareness.

Abstract

Traditional statistical optimization-based state estimation (DSSE) algorithms rely on detailed grid parameters and mathematical assumptions of all possible uncertainties. Furthermore, random data missing due to communication failures, congestion, and cyberattacks, makes these methods easily infeasible. Inspired by recent advances in digital twins (DTs), this paper proposes an interactive attention-based DSSE model for robust grid monitoring by integrating three core components: physical entities, virtual modeling, and data fusion. To enable robustness against various data missing in heterogeneous measurements, we first propose physics-informed data augmentation and transfer. Moreover, a state-of-the-art attention-based spatiotemporal feature learning is proposed, followed by a novel cross-interaction feature fusion for robust voltage estimation. A case study in a real-world unbalanced 84-bus distribution system with raw data validates the accuracy and robustness of the proposed DT model in estimating voltage states, with random locational, arbitrary ratios (up to 40% of total measurements) of data missing.

On the Potential of Digital Twins for Distribution System State Estimation with Randomly Missing Data in Heterogeneous Measurements

TL;DR

The paper addresses DSSE under realistic data-absence and multi-modal measurement conditions by introducing a digital twin that jointly harnesses physical knowledge, virtual modeling, and data fusion. It presents an end-to-end interactive DT with physics-informed data augmentation, parallel transfer for heterogeneous measurements, and a cross-interaction fusion scheme based on group-queried attention to achieve robust voltage estimation without imputation. The approach is validated on a real unbalanced 84-node distribution network, showing superior accuracy and stability compared with LSTM and Transformer baselines, and demonstrates resilience to random data missing up to 40% of measurements. This work offers a practical, end-to-end DSSE framework that can adapt to varying observability and data-quality scenarios in real distribution grids, reducing error accumulation and improving situational awareness.

Abstract

Traditional statistical optimization-based state estimation (DSSE) algorithms rely on detailed grid parameters and mathematical assumptions of all possible uncertainties. Furthermore, random data missing due to communication failures, congestion, and cyberattacks, makes these methods easily infeasible. Inspired by recent advances in digital twins (DTs), this paper proposes an interactive attention-based DSSE model for robust grid monitoring by integrating three core components: physical entities, virtual modeling, and data fusion. To enable robustness against various data missing in heterogeneous measurements, we first propose physics-informed data augmentation and transfer. Moreover, a state-of-the-art attention-based spatiotemporal feature learning is proposed, followed by a novel cross-interaction feature fusion for robust voltage estimation. A case study in a real-world unbalanced 84-bus distribution system with raw data validates the accuracy and robustness of the proposed DT model in estimating voltage states, with random locational, arbitrary ratios (up to 40% of total measurements) of data missing.

Paper Structure

This paper contains 11 sections, 8 equations, 5 figures, 2 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Preliminary
  3. Distribution System State Estimation
  4. Practical Data Challenges
  5. Proposed Interactive Attention-Based DT
  6. Physics-Informed Data Augmentation and Transfer
  7. Cross-Interaction Feature Fusion for Voltage Estimation
  8. Case Study
  9. Estimation Accuracy and Scalability
  10. Robustness against Random Data Missing
  11. Conclusion

Figures (5)

  • Figure 1: The proposed interactive attention-based DT architecture
  • Figure 2: The proposed GQA module in the feature fusion layer
  • Figure 3: Comparison of estimated voltages over multiple time steps by different methods. The ground-truth values are also plotted.
  • Figure 4: Comparison of the statistical distribution of errors for voltage magnitudes and phase angles in all test cases among three models.
  • Figure 5: Error trends of the proposed method in RMSEs and MAEs under different data missing ratios