We modified further our extended dynamic model of a geyser induced by an inflow of gas, by taking into consideration the effects during spouting of an elbow shape, pairs of sudden expansions and contractions, and repeats of this shape in an underground watercourse. Through numerical simulations of this extended dynamic model, we see that a large number of sudden expansions and contractions or a...

Numerous subsurface factors, including geology and fluid properties, can affect the connectivity of storage spaces in depleted reservoirs; hence, flow simulations become more complicated, predicting their deliverability remains challenging. This paper applies Machine Learning (ML) techniques to predict underground natural gas (UNGS) reservoirs. First, three baseline models were developed based ...