基于正态分布理论的模糊综合评判法识别岩性

不同的测井参数对岩性的识别功能各不相同,只有综合评判各个测井参数曲线才能大致反映地层主要岩性特征,因此传统的定量的数学关系式很难正确地描述地层岩性。模糊综合评判法对多因素、多层次的复杂问题具有较好的评判效果。采用模糊综合评判法识别岩性过程中,合理地确定评判因素及其隶属度函数尤其重要。通过对大量测井和录井资料统计分析,发现对于同一岩性层段,部分测井响应特征参数基本服从正态分布规律。因此利用正态分布理论确定各评判因素隶属度函数,并通过模糊变换,得出用模糊集合表示的评价结果。将评判结果与实例比较证明,该种方法能够克服传统岩性识别中存在的缺点和不足,评判结果更加科学可靠。

LITHOLOGY IDENTIFICATION ANALYSIS BASED ON NORMAL DISTRIBUTION THEORY AND FUZZY SYNTHETIC EVALUATION METHOD

Different log parameters have different functions on lithology identification, only combining all the log parameters can you identify the main lithologic features, so it is difficult to describe formation lithology accurately by the traditional quantitative mathematical relationship. Fuzzy synthetic evaluation method has a good judge effect in multi-factor and multi-level complex problems. It is very important to decide evaluation factors and the degree of membership when using fuzzy synthetic evaluation method to identify lithology. In this paper, we analyzed a large number of logging data and found some logging response characteristics parameters of the same lithological layers subject to the normal distribution. So this paper used the normal distribution theory to determine the membership function of the evaluation factors , and obtained evaluation result by fuzzy transforming. Practice proved that, this method has overcome the disadvantages and insufficient existing in the traditional method of lithology identification, and made the judge results more scientific and reliable.