We have conducted a comparison between three types of recurrent neural networks and their ability to predict anomalies occurring in oil wells using publicly available dataset. included two well-known state-of-the-art new type with neurons evolved specifically for the dataset automatic programming. show that neuron offers massive improvement over state-of-the-art. The overall test accuracy netwo...

The main purpose of this paper is two-fold: (i) to generalize an existence result for a compressible gas-liquid model with a friction term recently published by Friis and Evje [SIAM J. Appl. Math., 71 (2011), pp. 2014–2047]; (ii) to derive a uniqueness result for the same model. A main ingredient in the existence part is the observation that we can consider weaker assumptions on the initial liq...

There are numerous oil fields that approaching the end of their lifetime and have great geothermal potential considering temperature water cut. On other hand, industry is facing challenges due to increasingly stringent environmental regulations. An example this case France where extraction will be forbidden starting from year 2035. Therefore, some companies switching business investing in proje...