شناسایی لا یه های کربناته با کیفیت مخزنی با استفاده از مدل های هوشمند

Evaluation, development and management of reservoirs have a strong relationship with knowing the reservoir properties. Indeed, the porosity of a reservoir plays an important and basic role in evaluation of other petrophysical parameters. As the majority of the reservoirs in Iran are of carbonate type, in which high heterogeneity exists. Studying this type of reservoirs, is more important than the other ones as this type of reservoirs forms more hydrocarbon reservoirs in the world. To date, the petroleum industry has tried to determine porosity by injecting helium gas to core samples and to determine their textures by examining the obtained thin sections from the samples under the microscope. Laboratory methods are usually time consuming and costly, and are not possible in all the circumstances. In today’s world, oil industry is dealing with a large number of difficult problems, which do not meet all the needs of engineers and experts. In recent years, with advances in computer hardware and software, the use of artificial intelligent techniques and image analysis in the petroleum industry are expanded. Thus, in order to reduce costs and the time in reservoir studies, this study uses artificial neural network, fuzzy logic and neuro-fuzzy techniques to calculate the porosity of the collected core samples. The results show that a neuro-fuzzy model is more accurate, thus, it is a good model to estimate porosity.

کلیدواژه ها:

Intelligent methods ، Porosity ، Carbonate layers ، Log ، Core