本研究以熱探針法進行橫向等向性岩石的熱傳導係數量測，同時進行超音波波速的量測，為了提高熱探針法的精準度，並比較其關係。故於量測熱傳導係數前先以誤差傳播理論對熱探針法進行標準差的評估，並針對原始量測資料，對量測結果進行改良加權移動平均，減少量測雜訊對結果產生的影響。 實驗結果顯示，誤差傳播的理論可以求得熱傳導係數標準差。此標準差可應用於設計實驗參數，對實驗參數進行最佳化配置。對熱傳導係數影響最大的誤差為量測雜訊，可影響熱傳導係數達59 %以上。將數據透過改良加權移動平均法的處理後，使雜訊對熱傳導係數的影響降至9 %。於改良的過程中，發現若是熱傳導係數大的樣本，量測雜訊對熱傳導係數的影響越大。故若樣本的熱傳導係數越大，其量測雜訊的影響必需要減少，否則量測所得的熱傳導係數會有偏差。 於橫向等向性岩石的量測結果發現，本研究中的橫向等向性岩石，其平行片理方向的熱傳導係數約為垂直片理方向的熱傳導係數的兩倍，最快超音波波速為最慢超音波波速的兩倍以上。正交方向量測得的熱傳導係數比例，與最快超音波波速及最慢超音波波速比例相接近。 The subject of this research is using thermal probe method to measure the thermal conductivity of transversely isotropic rock while P-wave velocity at different angles is also measured. In order to improve the precision of thermal probe method, error propagation theory is used to estimate the standard deviation of measured results. A refined weighted moving average algorithm is applied to reduce the effect of noise from measuring. The value of standard deviation can be used to optimize experimental parameter for thermal probe method. For thermal conductivity, measurement noise affects the most, up to 59% rising of the standard deviation. Using a refined weighted moving average algorithm to process experimental data, the affection on the noise from measuring will reduce to 9%. The result of refined weighted moving average algorithm shows the effect of measurement noise is larger for a sample with higher thermal conductivity than a sample with lower thermal conductivity. Therefore, if a sample has higher thermal conductivity, the affection on noise must be reduced in order to maintain accuracy. For the measurement on transversely isotropic rock in the study, the thermal conductivity of parallel isotropic plane is about twice of vertical isotropic plane in transversely isotropic rock. The maximum P-wave velocity is more than twice of the minimum P-wave velocity. On measurements of a set of orthogonal directions, the ratio of thermal conductivity is almost the same with the ratio of P-wave velocity.
