灰色关联度与改进的贝叶斯判别法在油气层识别中的应用

The Study on the Gray Relational Degree and the Improved Bayes Discriminant Method

准确识别油气层是测井、录井解释的最终目的,为了提高解释准确率,首先通过灰色关联度分析优选测井、录井参数,继而对优选的测井、录井参数采用改进的Bayes判别法识别油气层。该方法可充分有效的利用测井、录井资料,避免了以往单一的利用测井或录井资料识别油气层符合率偏低的缺点,而采用改进的Bayes判别法识别油气层可以弥补Bayes判别法识别油气层时在距离计算过程中不能将各个自变量的权数选择与判别效率很好的联系起来的不足,从而进一步提高解释准确率。实验结果表明:先用灰色关联度分析优选参数,然后采用改进的Bayes判别法识别油气层的解释符合率平均达到82.38%。

The final purpose of logging interpretation and mud logging interpretation is to identify the hydrocarbon reservoir accurately.To improve the accuracy of interpretation,first,in this paper,logging and mud logging parameters are optimized by the gray relational degree,and then these parameters are further analyzed by the improved Bayes discriminant method so that hydrocarbon reservoir can be identified.Compared with the only using logging or mud logging data to identify hydrocarbon zones,the method proposed in this paper can make full and effective use of logging and mud logging data and the disadvantages of low accuracy can also be avoided.What's more,utilizing the improved Bayes discriminant method to identify hydrocarbon reservoir can help compensate the insufficient that the choice of weight of each independent variable and discriminant efficiency can not be better connected by the Bayes discriminant method in the distance calculation process.Hence,the accuracy of interpretation is further more improved.It shows in the experiment that the interpretation coincidence rate of the method that the gray relational degree is used in the first to optimize parameters,and then the improved Bayes discriminant method is used to identify of hydrocarbon reservoir can reach 82.38% equally.