摘要
针对基于Am-Be中子源测量中子俘获谱确定地层元素的测井方法，用蒙特卡罗数值模拟方法模拟测量响应谱，研究一种简化的解谱方法，直接求解地层元素的重量百分含量。模型参数设计包含砂岩、石灰岩和白云岩等3种典型的地层，再参入适量的Al、K、Fe、S、Ti、Gd等元素。解谱计算设计了含有H、Si、ca、Fe、Al、S、K、Mg、Ti、Gd等10种元素的地层；地层孔隙度从0．1％～34％。H、Si、Ca、Fe模拟解析结果与理论值有很好的相关性，ca的相关性最好（R2—0．984），白云岩地层Mg元素含量也有较好反映；A1、S、K的测井灵敏度较低，解出的含量相关性不好；Ti、Gd的中子俘获伽马截面积虽然很大，不影响主要元素的含量求解。去除H含量后，也可以以骨架元素含量的表现方式求解元素含量。通过与氧化物闭合模型求解元素含量的方法相比，该方法相对简便，对于实际刻度过程的要求相对容易实现。
Monte Carlo numerical simulation method is used to calculate some typical formations mixed spectrum, and then, the absolute content of each formation element is obtained through spectrum deconvolution. Model design. Monte Carlo simulation geometric model is given （see Fig1）, in which the detector is BGO crystal, and the model well consists of sandstones, lime- stones and dolomites with a proper quantity of AI, K, Fe, S, Ti, and Gd, etc. Simulation algo- rithm. The algorithm uses F8 counting factor, and Gaussian stretch is made according to BGO detector＇s energy resolution. Fig 2 shows that the capture GR spectrum peaks of Ca element in limestone and Si in sandstone are very obvious. Spectrum deconvolution. Weight percentage of each element is derived from equation given in the paper. Ten more elements such as H, Si, Ca, Fe, A1, S, K, Mg, Ti and Gd are given in the spectrum decovolution algorithm. From correla- tion plot between each simulated element content and theoretical value, we can see the contents of H, Si, Ca and Fe have much better correlation with the theoretical values; Content of Mg in dol- omite has better correlation, but other element don＇t have good correlation. The method using multiple regression method to figure out the elements may directly acquire the absolute content of each element and its standard spectrum.
出处
《测井技术》
CAS
CSCD
北大核心
2011年第B12期642-645,共4页
Well Logging Technology
关键词
测井方法
地层元素
俘获伽马
蒙特卡罗数值模拟
解谱
logging methodology, formation element, capture GR, Monte Carlo numerical simulation, spectrum deconvolution
