A monitoring-mining-modeling system and its application to the temperature status of the Xiluodu arch dam

Abstract

Knowledge of the concrete temperature of a dam is necessary for the implementation of improved construction methodologies. A monitoring-mining-modeling system is proposed in this article to study the thermal state of a super high arch dam in China. The monitoring-mining-modeling system includes an in situ monitoring network setup, data mining methods, and numerical finite element analysis. Two engineering cases are surveyed in detail using monitoring-mining-modeling system. The first case is the long-term temperature rise phenomenon, whose cause is determined from the data mining result, and the numerical modeling of the external boundary condition is improved using the result. The modeling result then shows that the effect of the TRP is also a dangerous factor for the deformation and stress state of the dam. The second case is the performance of cooling pipes. The energy absorption capability of the cooling pipes is revealed by the data mining result, as well as the performance difference between high-density polyethylene and iron pipes. The numerical modeling result shows good agreement with the monitored result, which demonstrates the closed-loop validity of the monitoring-mining-modeling system.