Application of Fuzzy Control Theory in the Study of the Macroscopic Remaining Oil Distribution for a Mature Sandstone Reservoir in the Tarim Oilfield, China

Research on macroscopic remaining oil distribution (MROD) has become a worldwide issue and not yet been satisfactorily solved in oil industry. At present, the commonly used methods for determining MROD are heavy workload, long time consuming and greatly influenced by human factors. In the research of MROD, geological factors are uncontrollable and development factors are controllable. Therefore, under certain development conditions, geological factors determine the MROD. This paper mainly studies MROD in geological factors. The new approach is based on development geological principles. It uses fuzzy control theory to predict MROD. Firstly, according to structural location from high to low and clay content from much to less, the structural location parameters and sedimentary microfacies parameters are fuzzified into 5 grades respectively. Secondly, according to the hydrocarbon potential from good to poor, the oil and gas universe is fuzzified into 5 grades. Finally, fuzzy control rules were established. It can be calculated by characteristic expansion method. Complete governing equations are provided and the method is described in detail to permit readers to replicate all results. The method of using fuzzy control theory to predict MROD has been successfully applied in Tarim oilfield. According to the range of tectonic height and clay content, the membership degrees of structural location parameters and sedimentary microfacies parameters are assigned respectively. LN2-G well in research region is chosen as an example for study. The characteristic coefficients corresponding to LN2-G well under different structure and sedimentary conditions are respectively obtained. Through the fuzzy control statement, and combining the principle of maximum membership degree, the oil level in the target formation of well LN2-G can be calculated, which value is 5, corresponding to the oil layer. Then we use the existing exploration wells in research region, and use the above method to calculate them respectively. Based on the calculated results, we draw the remaining oil distribution level range prediction plan. Finally, we selected 3 wells to verify the accuracy of the prediction. It shows that the reservoirs drilled by LN2-X, LN2-Y and LN2-Z wells are water layer, water layer and poor oil layer, respectively. The actual drilling results of these 3 wells are water layer, water layer and oil-bearing water layer, which are completely consistent with the prediction results. This research can accurately forecast the MROD based on the analysis of the essential factors (geological factors) affecting the MROD. The method is convenient and economical, which provides a reliable basis for the remaining oil development.