物探异常的间接模式识别

INDIRECT PATTERN RECOGNITION OF GEOPHYSICAL ANOMALIES

利用模糊数学知识，对物探工作已知区和未知区的各异常，选择描述其特征的数量化指标，构成各异常的模糊集，然后采用ＩＳＯＤＡ－ＴＡ模糊聚类方法，将巳知区异常分成地质、物探特征各异的若干类，作为已知模式。进而对未知区各异常作间接模式识别，将其归入和已知模式中贴近度最大的那个模糊集，从而将“从已知到未知”这一解释推断原则定量化、客观化。运用该方法对大足一壁山一带航磁剩余异常的间接模式识别取得了较为满意的结果。

By using fuzzy sets theory, the authors have not only overcame the defects in the traditional method of cluster analysis, but also gained the method of indirect pattern recognitionwhich makes the interpretation principle of 'interpreting the unknown by the known' quantitative. It has been used to recognize the patterns of residual airborne magnetic anomalies in Dazearea and proved to be effective by the resea-reh results.