Comparing Object Detection Models for Electrical Substation Component Mapping

Abstract

Electrical substations are a significant component of an electrical grid. Indeed, the assets at these substations (e.g., transformers) are prone to disruption from many hazards, including hurricanes, flooding, earthquakes, and geomagnetically induced currents (GICs). As electrical grids are considered critical national infrastructure, any failure can have significant economic and public safety implications. To help prevent and mitigate these failures, it is thus essential that we identify key substation components to quantify vulnerability. Unfortunately, traditional manual mapping of substation infrastructure is time-consuming and labor-intensive. Therefore, an autonomous solution utilizing computer vision models is preferable, as it offers greater convenience and efficiency. In this research paper, we train and compare 16 models on a manually labeled dataset of US substation images. These models include 12 from the You Only Look Once, or YOLO, family, 2 from the RF-DETR family, and 2 from the Cascade R-CNN family. Each model is evaluated for detection accuracy, precision, and efficiency. We present the key strengths and limitations of each model and identify which model provides reliable, large-scale substation component mapping. RFDETR-large achieved the highest overall detection performance, with an mAP@50 of 0.881 and a mAP@50:95 of 0.632, outperforming both YOLO and Cascade R-CNN families. Across all models, alternate energy systems were the most accurately detected, while transformers and reactors were the most difficult to identify due to their smaller size and greater visual variability. Applying our best-performing model to nationwide imagery, it identified approximately 22,591 substation components across the United States across 11,083 unique substations within the United States, with MISO, PJM, and SERTP FERC Order 1000 transmission planning regions containing the largest amount.