Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Abstract A deliverability test is a flow test performed to generate data utilized in the characterization of the short and long-term production performance of a well for various conditions of reservoir and bottomhole pressure. The generated data includes bottomhole pressure and temperature, flowing wellhead pressure and temperature measurements as well as the associated flow rates. An analysis of the acquired data ultimately yields a well model capable of predicting a tested well's productivity under stipulated conditions. This paper presents a methodology by which the data output from a flow-after-flow test may be generated without ever performing the actual test. A machine learning (ML) model developed and trained on historic production data is used to simulate a producing wells performance at several rates as is typical for an actual flow-after-flow test. The methodology utilizes a ML model trained on recent historical production data comprising of flow rates, flowing pressures and flowing temperatures, measured at surface and/or downhole. The ML model is then used to predict the flowing pressures and temperatures for a user specified flow rate. In its current form, the model is designed to utilize a continuous data stream that may be from a real time or archived source. The model may also be adapted to utilize data with a batch type configuration. By minimizing the need for actual deliverability tests, or altogether eliminating the need for testing, operators stand to benefit immensely. The benefits include minimization or reduction of costs associated with testing requirements, curtailed production, as well as the minimization of emissions.
At first glance, well testing may not come across as high-tech. At a time of rampant digitalization, with catchy phrasing such as "machine learning" and "Internet of things," how can we compete with "slip joints" and "Bourdet derivatives" from the '80s? Further down Memory Lane, bubble buckets and semilog graph paper look even more as if they are from a bygone era. Yet the former are still relied upon in specific onshore applications (I used one 2 years ago), and petroleum engineering students continue to learn the fundamentals of pressure transient analysis with the latter. Most of today's equipment and interpretation methods are indeed not new.
Summary Pressure measurement from permanent downhole gauges (PDHGs) during extended shut ins (SIs) is a key piece of information that is often used for model calibration and reserve estimation in deepwater gas reservoirs. A key challenge in practical operation has been the failure of PDHGs within the first few years of operation. In this work, a physics-based data-driven (PBDD) model and machine learning (ML) models are developed to predict PDHG pressure and temperature measurement from the wellhead and other measurements during well SI events for deepwater dry-gas wells. During SI events, the wellbore cools down, resulting in increased gas density and bottomhole pressure (BHP). In the PBDD model, the temperature profile in the well is modeled with a piece-wise linear model as derived from wellbore simulations. The temperature decline during cooldown is captured using a decline-curve model, with the decline-curve parameters dependent on the location. The dependency of the cooldown effect on past production is captured with a linear model. Model parameters in the PBDD model are calibrated with data. In the ML models, multiple methods are tested, and the best performing method is picked based on cross-validation results. Two use cases are considered in this work. The first case (single well) involves predicting future SI BHP and temperature based on past PDHG measurement of the same well. Both the PBDD model and the ML model show good accuracy in blind tests for this use case. The second case involves predicting SI BHP and temperature of a well based on PDHG measurement of other wells. The PBDD model sees reduced accuracy in temperature prediction but is still reasonably accurate, while unphysical behavior is observed for the ML model even though the cross-validation score is high. It is concluded that, comparing the two types of models, the PBDD model is constrained by physics and thus the result is more interpretable and reasonable even when extrapolating. It also can provide the entire temperature and pressure profile during SIs. However, it does come with a series of assumptions (such as dry gas with no liquid content) and needs to be modified when the problem changes. On the other hand, the ML model is easier to construct and extend to other cases but is not bounded by physics so the result could be unphysical when extrapolation occurs.
During well-testing operations, incorrect burner combustion could have an adverse effect on the environment or personal safety. Combustion efficiency is assessed by personnel who observe the flame. This practice lacks consistency and poses challenges, including environmental and safety considerations and issues with data. In the complete paper, the authors propose a solution that uses a deep neural network that learns from flame videos to define the quality of the combustion.
ABSTRACT: Many techniques were provided for the evaluation of the productivity index of the producing wells. Most of these techniques use the information of the pressure buildup tests. This means that the process of productivity index estimation is accompanied by a shut - in period of the producer. This study provides a new approach based on one technique of those used for time series analysis. In the present work a simple idea about the dependence between the bottom hole flowing pressure and the production rate was developed through the well-known concepts of reservoir and production engineering. By using the dynamic change of both parameters auto correlation function is switched on to pick out the mutual independence of the two variables, which is, simply, the productivity index. The suggested method was conducted over two producing wells from the Gulf of Suez and the Western Desert. The estimated productivity index by the new approach does match that calculated through the buildup process. However, as the technique requires simultaneous mesurements of these two factors, it can be used wherever we have downhole multisensors. This situation encourages us to develop similar approach for replacing bottomhole flowing pressure by the wellhead pressure, through sensitivities to vertical flow correlations to simulate wells.
- Asia > Middle East > Saudi Arabia (0.24)
- Africa > Middle East > Egypt (0.24)