The current research is focused on the Asmari reservoir in one of Iran's oil fields. Integration of well logs and seismic data is a main goal of geophysicists. Likewise, one of the most significant petrophysical parameters is effective porosity prediction, which plays a very important role in the oil and gas industries. Effective porosity is predominantly significant from the point of view of petroleum geology and exploitation engineers; consequently, in this study, the researchers examine effective porosity. In the research method used in this project, first, via seismic data inversion, the acoustic impedance attribute is extracted. By applying mathematical relations to it and other seismic characteristics with well logs, the parameters of the reservoir porosity at the well site are estimated and then extended to the seismic data range. Lastly, their lateral and vertical changes were checked. The current study was done in a hydrocarbon anticline, and the effective porosity of the source rock was estimated using common seismic attributes. The examined methods for porosity estimation are single and multi-attribute methods and neural networks. The PNN algorithm uses well logs in the training phase to estimate reservoir properties. The results of the porosity estimation in the single and multiple attributes and neural network methods have been compared. Compared to the other two methods, the use of the neural network method has resulted in less error in estimating the effective porosity. This study has revealed that the neural network application effectively predicts porosity.
