TY - JOUR AV - none ID - scholars18822 SP - 151 TI - Prediction of Electromagnetic Properties Using Artificial Neural Networks for Oil Recovery Factors KW - Computer system recovery; Enhanced recovery; Forecasting; Gas industry; Multilayer neural networks; Petroleum reservoir engineering; Recovery; Secondary recovery KW - Electromagnetic properties; Enhanced-oil recoveries; Oil and Gas Industry; Oil consumption; Oil Prices; Oil recoveries; Operational conditions; Original oil in places; Recovery factors; Water saturations KW - Deep neural networks IS - 1 N2 - Abstract: The most important parameter in the oil and gas industry is the recovery factor (RF). Higher oil consumption has resulted in a rise in global oil prices. Several research have been carried out in order to create, analyze, and optimize operational conditions. However, several reservoir factors such as viscosity, porosity, permeability, water saturation, original oil-in-place, and API gravity have been employed in the development of RF to boost oil recovery techniques without taking electromagnetic properties into account. The recovery factor was predicted using core flooding tests and a deep neural network using electromagnetic parameters. We offer a deep neural network (DNN) method with 256 nodes, seven hidden layers, and a single output. According to the acquired results, the DNN algorithms' coefficient correlation is R2 = 0.98478 for training and R2 = 0.91679 for testing, which was subsequently evaluated and confirmed for RF prediction. In terms of cost and production effectiveness, the results of this study reveal a good forecast of RF with reservoir rock and fluid parameters. © 2023, Pleiades Publishing, Ltd. N1 - cited By 0 Y1 - 2023/// EP - 165 VL - 85 A1 - Sikiru, Surajudeen A1 - Soleimani, H. A1 - Shafie, A. A1 - Olayemi, R.I. A1 - Hassan, Y.M. JF - Colloid Journal UR - https://www.scopus.com/inward/record.uri?eid=2-s2.0-85152886016&doi=10.1134%2fS1061933X2260004X&partnerID=40&md5=07ad91d8a631d8136b03b2e262b63521 ER -