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.

Reliable forecasting of production rates from mature hydrocarbon fields is crucial both in optimizing their operation (via short-term forecasts) and in making reliable reserves estimations (via long-term forecasts). Several approaches may be employed for production forecasting from the industry standard decline curve analysis, to new technologies such as machine learning. The goal of this study is to assess the potential of utilizing deep learning and hybrid modelling approaches for production rate forecasting. Several methods were developed and assessed for both short-term and long-term forecasts, such as: first-principle physics-based approaches, decline curve analysis, deep learning models and hybrid models (which combine first-principle and deep learning models). These methods were tested on data from a variety of gas assets for different forecasting horizons, ranging from 6 weeks to several years. The results suggest that each model can be beneficial for production forecasting, depending on the complexity of the production behavior, the forecasting horizon and the availability and accuracy of the data used. The performances of both hybrid and physical models were dependent on the quality of the calibration (history matching) of the models employed. Deep learning models were found to be more accurate in capturing the dynamic effects observed during production – this was especially true for mature fields with frequent shut-ins and interventions. For long-term production forecasting, in some cases, the hybrid model produced a greater accuracy due to its consideration of the long-term reservoir depletion process provided by the incorporated material balance model.