Enhancing Distribution System State Estimation Using SRGAN-Based Data Interpolation for Load and DER Measurements

Accurate Distribution System State Estimation (DSSE) is critical for the reliable operation, monitoring, and control of modern distribution networks, ensuring precise estimation of system states essential for maintaining voltage stability and efficient load management. However, the limited temporal resolution of measurement data from smart meters, typically recorded at 30 -minute intervals, poses challenges to achieving high-accuracy DSSE. This paper proposes a novel approach that employs a Super-Resolution Generative Adversarial Network (SRGAN) to upscale low-resolution data for load power consumption and Distributed Energy Resource (DER) measurements into high-resolution data. By capturing finer temporal variations in the measurement data during training process, SRGAN improves the quality of measurements for DSSE, leading to more accurate state estimation. The proposed method substantially improves the quality of high-resolution measurements when compared to linear-interpolated data. By integrating SRGAN-interpolated pseudo-measurements into the state estimation process, the approach achieves a marked improvement in state estimation accuracy over low-resolution data. This demonstrates the strong potential of SRGAN-based interpolation to enhance the precision and reliability of state estimation in distribution networks, offering a promising solution for addressing measurement gaps in modern power systems.