تخمين سرعت موج استونلي از نگارهاي پتروفيزيكي با استفاده از ماشين مشاوره هوشمند در سازند سروك، دشت آبادان

Prediction of Stoneley wave velocity from petrophysical logs using intelligent committee machine in Sarvak Formation, Abadan Plain

Stoneley wave velocity (Vst) which is a type of surface waves provides valuable information from hydrocarbon reservoirs. In this study Vst was predicted from well log data using neuro-fuzzy, Sugeno and Mamdani fuzzy inference systems, and then results of each intelligent system were combined by an intelligent committee machine (CMIS). A genetic algorithm was used for constructing CMIS. For this purpose a total of 3030 data points from two wells in Sarvak carbonate reservoir which have Vst and conventional well log data were utilized. These data were divided into two parts, one part included 2047 data points used for constructing intelligent models and the other part included 983 data points used for models testing. The results show that CMIS technique has been useful method for prediction of Vst in Sarvak Formation with correlation coefficient equals to 0.98 and mean squared error equals to 0.000096.