Improving the Accuracy of Virtual Flow Metering and Back-Allocation through Machine Learning

Abstract

In this study we have investigated a fully data-driven approach (artificial neural networks) for real-time back-allocation and virtual flow metering in oil and gas production wells. The main goal of this study is to develop computationally efficient data-driven models to determine the multiphase production rates of individual phases (gas and liquid) in wells using existing measured data in fields. The developed approach was tested on simulated and field data from several gas wells. Two different type of artificial neural networks (ANNs) were tested on simulated and field data to assess the accuracy of estimations for steady-state, transients and dynamics in productions due to cyclic operation (shut-ins and restart). The results showed that ANN was capable of accurately estimate the multiphase flow rates in both simulated and field data. The accuracy of the production rates estimation depends on the type of neural networks employed, production behavior (steady-state or transients) and uncertainties in data.

Data-Driven Virtual Flow Rate Sensor Development for Leakage Monitoring at the Cradle Bearing in an Axial Piston Pump

“…The average absolute percentage error during the test was only about 4%. In addition to these, several other studies have been carried out to investigate the feed-forward neural network-based virtual sensor for flow rate [16][17][18][19][20][21][22][23]. In these studies different activation functions such as the sigmoid function, radial basic function, etc., as well as network structures were investigated.…”

“…All of them were able to provide excellent prediction performance in the system's steady state. In addition to the simple feed-forward neural networks (NN), long short-term memory (LSTM) algorithm [17,24,25] and neural networks combined with novel ensemble learning [26] have also been investigated. LSTM outperforms feed-forward neural networks in system transient operation.…”

Productivity Prediction of Fractured Horizontal Well in Shale Gas Reservoirs with Machine Learning Algorithms

“…Therefore, the numerical simulation of shale gas production is a time-consuming method. Recently, machine learning (ML), especially deep learning, has developed rapidly, providing an effective means for shale gas production forecast [10,11]. ML is good at dealing with nonlinear problems.…”

Combining design of experiments, machine learning, and principal component analysis for predicting energy consumption and product quality of a natural gas processing plant

“…18,19 Moreover, Kamari et al developed an ANN with particle swarm optimization to predict the natural gas well performance, 15 Ghasemi et al developed a particle swarm optimization procedure to optimize the parameters of two neuro-fuzzy models, for power density and percentage of chemical oxygen demand to be maximized, in order to improve the microbial fuel cell performance, 20 Tahmasebi et al combined fuzzy neural network with genetic algorithm (GA) to model the relationship of the parameters in a shale reservoir, 10 Xiao et al proposed a combination of ANN models and sequence quadratic program algorithm to obtain optimized performance of hydrogen purification. 21 Furthermore, Al-Qutami et al 22 proposed radial basis function-neural network (RBF-NN) to predict gas flow rate in multiphase flow in petroleum industry, Omrani et al developed ANN protocols to improve metering accuracy in oil and gas production wells, 23 Lee et al utilized combined NN and fuzzy procedure to develop an adaptive control methodology and then considered model predictive control (MPC) for energy saving in buildings and finally proposed a mixed GA and SVM techniques to tune the MPC parameters, 24 Putcha and Ertekin modeled an oil production system using ANN coupled with a numerical compositional reservoir simulation model. 25 However, none of the above articles considered modeling of two critical items of a gas processing plant: (1) total consumed energy and (2) final water content of the processed gas.…”