Optimisation of gas lift performance using artificial neural network

Abstract

Gas lift is one of the most widespread methods of artificial lift technologies used when wells' production rate declines. Gas is employed to maintain the production by injecting gas into the tubing through a gas lift orifice. Lifting costs are generally low. However, capital costs of compression are very high, so it is necessary to optimise gas lift wells. In this paper, conventional nodal analysis models were used to predict the optimisation parameters based on wells system parameters. Artificial neural network (ANN) models were also used based on gas lift databases. ANN models were trained then tested against nodal analysis models. Also, this paper presents a new theory about the relative importance of gas lift system inputs on output parameters of gas lift system. [Received: June 9, 2020; Accepted: August 7, 2020]