Writers : Akhoundi, Habib; Kamali, Mohammad Reza; Kadkhodaei, Ali; Amini, Amin; Abbasi, Amir
Refference : proceeding of the first applied geological congressof iran (volume.2)
Publishing Year : 1386

Abstract :
Determination of accurate porosity is important for petroleum engineers to evaluate hydrocarbon reservoirs. There are two methods for porosity determination including direct (core analysis) and indirect (well log analysis). However, there are limitations for each method mentioned. The well log data are available for almost all wells of a field, but they indicate to some portion of porosity. As well as, due to economic reasons and impossibility of coring in horizontal wells, some wells have core data. In the present study, intelligent soft computing neural networks have been used for porosity estimation. For this purpose, the dataset of three wells form a study field including well logs and cores have been used for designing and developing a neural network with back propagation algorithm. By comparison of measured and network results, the parameters of the NN were adjusted for a desired network. Finally, the third well was used for testing the model and estimation of porosity. The results show that NN predicted porosity is acceptable.
Subject List : porosity