Real-time probabilistic backfill thermal property estimation method enabling estimation convergence judgment

Abstract

This study proposes a real-time probabilistic estimation method for determining the thermal properties of backfill used in underground power transmission lines and ground heat exchangers. The thermal properties of backfill, serving as a heat conductor between the heat source and surrounding soil, are crucial design parameters for underground transmission lines and ground heat exchangers because temperature limits their maximum operational output. However, an unpredictable environment and human fieldwork during on-site backfill construction lead to considerable performance uncertainty, requiring in situ thermal property estimation. Furthermore, when transmission lines are installed over long distances, on-site assessments are necessary at multiple locations, making it essential to reduce the experimental time. The proposed estimation method enables real-time monitoring of temporal changes in the estimated probability distribution of the thermal property and quantifies the estimation uncertainty. The proposed method also allows immediate decision-making regarding experiment termination by evaluating real-time convergence by calculating the Jensen-Shannon divergence of sequentially estimated probability distributions. The effectiveness of the proposed method was tested on two backfill materials, demonstrating its ability to determine the estimation convergence by capturing contextual information affecting the estimation uncertainty.