期刊名称：IOP Conference Series: Earth and Environmental Science
印刷版ISSN：1755-1307
电子版ISSN：1755-1315
出版年度：2019
卷号：304
期号：3
页码：1-11
DOI：10.1088/1755-1315/304/3/032058
出版社：IOP Publishing
摘要：Temperature in an arc dam is a critical quantity of interest in safety monitoring. How to estimate a temper-ature field for an arc dam has been a challenging task, particularly when only a limited number of ther-mometers are arranged. This paper proposes to use the ordinary Kriging to estimate the two-dimensional temperature fields based on limited monitored temperature data from thermometers in an arc dam. The Kriging method interpolates the temperature field between spatially distributed values of temperature. The effects of scale of fluctuation (δ) and correlation structure on the estimated temperature fields are examined. The results indicate that the Kriging method can estimate the temperatures at locations where the thermometers are not arranged. The estimated temperatures near the measuring points have higher accuracy. The estimated variances near the thermometers are lower than the variances far away from the thermometers regardless of the δ. When the δ is high, the data from conditioning (known) points can reduce the uncertainties of estimated temperatures near the known points. The distributions of tempera-tures using different correlation structures are quite similar. The δ has obvious effects on the smoothness of the temperature fields, contours and variances.
