Estimation of the fracture density in reservoir rock using regression analysis of the petrophysical data
Image logs are presently the main specialized tools for fracture detection in hydrocarbon reservoirs. Where image logs are not available, other less rewarding substitutes such as isolated well tests and type curve analysis, drilling mud loss history, core description and conventional petrophysical logs are used for fracture detection. In this paper a novel method is proposed for fracture density estimation in the fractured zones, using energy of petrophysical logs. Image and petrophysical logs from Asmari reservoir in one well of an oilfield in southwestern Iran were used to investigate the accuracy and applicability of the proposed method. Energy of the petrophysical logs in the fractured zones is calculated and linear and non-linear regressions between them are estimated. Results show that there is strong correlation between the energy of caliper, sonic (DT), density (RHOB) and lithology (PEF) logs with fracture density in well. In order to find a generalized estimator, a unique normalization method are developed, and by using it, a non-linear regression has been found which estimates fracture density with correlation coefficient of higher than 85%. The resultant regression has the capability of generalization in the studied field.