基于横波速度差异的裂缝分布识别方法——以准噶尔盆地火山岩地层为应用实例

A fracture identification method based on S-wave velocity difference: A case study form the volcanic strata in the Junggar Basin

在火成岩等裂缝和微裂缝发育地层的钻井过程中,井壁掉块甚至井壁坍塌引起的卡钻事故时有发生,严重影响了钻井安全和施工进度,要解决上述问题,需要在钻前了解地层的裂缝分布情况,进而设计合理的井身结构并采用有针对性的钻井工艺。为此,在详细分析声波速度与岩石力学特征参数关系的基础上,给出了以声波测井数据为依据的裂缝概率分布的定量计算方法,并且通过分析井径对声波时差影响的机理,提出了井径对声波时差影响的校正方法,进而建立了基于声波测井数据的裂缝识别模型。将该模型应用于准噶尔盆地2口深探井,应用效果表明:1裂缝识别结果与井径分布情况吻合较好,从而反映出该方法能够较准确地对裂缝进行识别;2井径变化对于声波时差会产生较大的影响,在数据处理过程中必须进行校正;3该方法利用常规的声波测井数据进行裂缝识别,也为一些仅有常规测井资料的老井进行裂缝识别提供了一条新的途径。

When well drilling is performed in the strata with developed fractures and micro-fractures （e.g. igneous rocks）, pipe sticking happens now and then due to wellbore sloughing and collapse, and consequently the drilling safety and construction schedule are affected badly. In order to solve these problems, it is necessary to figure out the distribution of fractures in strata so as to design casing program rationally and make use of drilling technologies properly. In this paper, the quantitative fracture probability distribution calculation meth- od based on the acoustic logging data was developed after the relationship between the acoustic velocity and rock mechanical character- istic parameters was analyzed in detail. Then, the correction method for the influence of well diameter on acoustic time was put forward after its influential mechanisms were investigated. Finally, the fracture identification model based on the acoustic logging data was es- tablished and applied to two deep exploration wells in the Junggar Basin. It is shown that fractures can be identified more accurately by using this new method, for its fracture identification result is in good consistence with well diameter distribution. The effect of well di- ameter change on acoustic time is so strong that it should be corrected during data processing. Based on this new method, fractures can be identified by using conventional acoustic logging data. Obviously, it provides a new approach to fracture identification for some old wells with conventional logging data only.