The Application of Improved 3D_Apriori Three-Dimensional Association Rules Algorithm in Reservoir Data Mining

Logging curve data plays a key role in oil and gas exploration and development owning to the ability to provide plentiful data and large amount of useful information, so the logging curves interpretation methods are also of importance. With the rapid increase of log data, our human is out of the ability to understand such numerous and complicated data, therefore conflict between the increasing data and the limited comprehend capability occurs. The thesis attempts to introduce association rules into logging data interpretation and provides a novel method. The classical Apriori algorithm is improved in the paper that named 3D_Apriori to interpret the logging attribute data and enhance efficiency of mining association rules behind the logging data transformation and the inherent information. Logging data acquired from Jingbian gas field of CNPC is used to verify the algorithm. Two strong spatial association rules are resulted from the computation. Applying these rules to interpret the test logging data, 78.6% coincidence validate the methodology.