Flowing Bottomhole Pressure Research Articles

Calculation Method for Bottomhole Flow Pressure in Co-layer Development of Multilayer Sandstone Reservoir

Abstract Accurately calculating the bottom hole flow pressure of a multi-layer stacked sandstone reservoir (hereinafter referred to as a multi-layer reservoir) is the basis for production splitting, oil well dynamic analysis, and reasonable work system determination in the development of multi-layer reservoirs. The bottomhole flow pressure model for single-layer oil reservoirs does not consider interlayer interference and the pressure drop caused by fluid flow in the wellbore, which doesn’t comply with the characteristics of multi-layer combined production, and the use of the maximum bottomhole flow pressure of the oil layer as the bottomhole flow pressure of each oil layer to adjust the production system of each oil layer doesn’t meet the requirements of other oil layer regulation. Firstly, starting from fluid mechanics, the frictional effect generated by fluid flow in the wellbore is systematically considered, and the relationship between wellbore frictional pressure drop and flow distance is established. The pressure drop formula generated by fluid flow in the wellbore is derived; Secondly, from the perspective of seepage mechanics, a formula for the bottom hole flow pressure of a single-layer reservoir was obtained. Combined with the expression for the difference in bottom hole flow pressure of each oil layer in the development of a multi-layer reservoir and the pressure drop expression for fluid flow in the wellbore, a bottom hole flow pressure model for each oil layer in the development of a multi-layer reservoir was established (hereinafter referred to as the multi-layer reservoir bottom hole flow pressure model). Substitute various parameters into the bottomhole flow pressure model of multi-layer oil reservoirs, and the error between the calculated values and the actual test values in the field is 0.10%~1.30%, with an average error of 0.82%, meeting the actual needs of the site. There is a significant difference in the bottom hole flow pressure values of each oil layer in the co production reservoir. The multi layer reservoir bottom hole flow pressure model can accurately calculate the bottom hole flow pressure of each oil layer, avoiding the inability to accurately adjust the production system of multiple oil layers for the same bottom hole flow pressure. It can provide technical support for the development of a reasonable drainage system for on-site co production oil wells and the realization of high and stable production.

A New Method of Using Production Data to Predict the Effect of Oil Well Liquid Lifting Based on Deep Learning

Abstract Liquid lifting is a low-cost, straightforward, and efficient measure for stabilizing production, achieved by adjusting liquid production to reduce bottom-hole flow pressure, thereby optimizing reservoir development and increasing recovery rates. However, traditional research methods for liquid lifting no longer meet the demands of contemporary oilfield big data applications. Particularly in the selection of target wells for liquid lifting, manual screening based on chart methods faces challenges such as low efficiency, high workload, and poor generalization capabilities. This study focuses on the oil wells in the high water cut phase of the SL oilfield in China and proposes a novel data-driven approach to predict the liquid lifting effects of oil wells. This method forecasts the effect of liquid lifting measures based on dynamic and static production data before the implementation of liquid lifting measures, facilitating intelligent selection of oil wells for liquid lifting measures. Firstly, each evaluation metric is assigned weights by utilizing the Coefficient of Variation-G1 Hybrid Cross-Weighting method. The comprehensive evaluation score is obtained through a weighted sum. Subsequently, the self-organizing map (SOM) clustering method is applied to categorize the oil well liquid lifting effects into four classes (A, B, C, D), representing different levels of effect. Additionally, a learning sample dataset is constructed by selecting multidimensional time-series features before liquid lifting. The oil well liquid lifting effect classification model is established using the Bi-LSTM sequence-to-label deep learning algorithm. Comprehensive testing validates the superior performance of the model, especially with classification accuracies of 0.96 and 0.83 on the training and testing sets, respectively. These results outperform other classification models such as LSTM, XGBoost, and SVM, providing an effective tool for intelligent oil well selection in liquid lifting measures.

Research of productivity and interlayer interference mechanism of multilayer co-production in reservoirs with multi-pressure systems

Abstract Due to the differences of physical and fluid properties in vertical oil layers, the interlayer contradictions are prominent and affect the reservoir utilization degree seriously. Therefore, the coordination between reasonable bottom-hole flow pressure and optimal production, as well as the reduction of interlayer interference are the primary concerns during the co-production of multi-layer oil reservoirs. Based on the reservoir engineering and fluid mechanics in porous media theory, this work adopt the comprehensive pressure system and consider the interlayer interference to establish a multi-branch horizontal well productivity model and interlayer interference mathematical model. By analyzing the main controlling factors and interference mechanisms, this article establishes a pattern of interlayer interference, and improves the quantitative characterization interlayer interference theory in multi-layer combined mining. The study has shown that: (1) The interlayer interference is beneficial for balancing the production of different layers and improving development efficiency, it is greatly affected by interlayer heterogeneity; (2) When the number of layers exceeds 2 layers, the interference coefficient increases; With the increase of the layer thickness, the thicker oil layers have higher productivity, and the thickness of layer has a significant effect on the production of low-pressure layer; When the production pressure difference increases, the interlayer interference can be reduced; (3) The interlayer interference mathematical model constructed in this paper has high prediction accuracy and strong practicality, which has theoretical guidance significance for the division of strata in comprehensive adjustment of reservoirs.

Fast Optimization of the Net Present Value of Unconventional Wells Using Rapid Rate-Transient Analysis

Summary Decline-curve analysis (DCA) is the industry standard to predict hydrocarbon production to then estimate the net present value (NPV) of unconventional wells. Conventional DCA assumes the flowing bottomhole pressure (BHP) is constant. This is an unrealistic assumption for many unconventional wells that can lead to incorrect estimates of ultimate recovery and thus erroneous NPV calculations. This work illustrates the application of a novel technique that combines variable BHP conditions with decline-curve models [Rapid rate-transient analysis (RTA)] to estimate the NPV and select the optimal cluster spacing and number of fractures for a given well. We compare the results of DCA and Rapid RTA to estimate and optimize the cluster spacing for a tight oil and shale gas well. The Rapid RTA results show marked differences in terms of the shape and the optimal value of the NPV function from the simpler DCA method. For the two cases illustrated, DCA results are a direct consequence of the rate-time analysis failing to correctly detect the onset of boundary-dominated flow (BDF). In contrast, Rapid RTA correctly detects the onset of BDF and presents a defined maximum value of the NPV vs. cluster spacing (number of fractures) function. This optimal value is a trade-off between the fracture treatment cost, the speed of recovery of hydrocarbons, and the value of money with time (discount rate). The major contribution of this work is the implementation of the Rapid RTA technique allowing fast computation of pressure-rate-time analysis and NPV calculations with computational times comparable with DCA for optimizing fracture cluster spacing and the total number of fractures of unconventional wells. We have developed a web-based application to give readers a hands-on experience of this new technique and to analyze and optimize the NPV of unconventional wells.

Evaluation of Key Development Factors of a Buried Hill Reservoir in the Eastern South China Sea: Nonlinear Component Seepage Model Coupled with EDFM

The HZ 26-B buried hill reservoir is located in the eastern part of the South China Sea. This reservoir is characterized by the development of natural fractures, a high density, and a complex geological structure, featuring an upper condensate gas layer and a lower volatile oil layer. These characteristics present significant challenges for oilfield exploration. To address these challenges, this study employed advanced embedded discrete fracture methods to conduct comprehensive numerical simulations of the fractured buried hill reservoirs. By meticulously characterizing the flow mechanisms within these reservoirs, the study not only reveals their unique characteristics but also establishes an embedded discrete fracture numerical model at the oilfield scale. Furthermore, a combination of single-factor sensitivity analysis and the Pearson correlation coefficient method was used to identify the primary controlling factors affecting the development of complex condensate reservoirs in ancient buried hills. The results indicate that the main factors influencing the production capacity are the matrix permeability, geomechanical effects, and natural fracture length. In contrast, the impact of the threshold pressure gradient and bottomhole flow pressure is relatively weak. This study’s findings provide a scientific basis for the efficient development of the HZ 26-B oilfield and offer valuable references and insights for the exploration and development of similar fractured buried hill reservoirs.

Optimization Design of Deep-Coalbed Methane Deliquification in the Linxing Block, China

The production of deep-coalbed methane (CBM) wells undergoes four stages sequentially: drainage depressurization, unstable gas production, stable gas production, and gas production decline. Upon entering the stable production stage, the recovery rate of deep CBM wells is constrained by bottom hole flowing pressure (BHFP). Reducing BHFP can further optimize CBM productivity, significantly increasing the production and recovery rate of CBM wells. This paper optimizes the deliquification process for deep CBM in the Linxing Block. By analyzing the production of deep CBM wells, an improved sucker rod pump deliquification process is proposed, and a method considering the flow in the tubing, annulus, and reservoir is established. Using the production data of Well GK-25D in the Linxing CBM field as an example, an optimized design of the improved rod pump deliquification process was undertaken, with design parameters including the depth of the sucker rod pump, the stroke length, and stroke rate. The results show that the improved process significantly lowers the pressure at the coalbed, enhancing CBM well production by 12.24%. The improved sucker rod pump process enriches deliquification technology for deep CBM, offering a new approach for its development and helping to maximize CBM well productivity.

Study of the effect of salt deposition on production capacity and storage capacity in underground gas storage

Underground gas storage (UGS) is the most economical and effective means to guarantee stable gas supply. During gas production process, the evaporation of formation water leads to the increase of water content in the gas, and the salinity of the remaining formation water increases. This work applied numerical simulation to analyze the effect of salt deposition on flowing bottomhole pressure, production capacity and storage capacity. The simulation results show that the minimum and maximum pressure of UGS is more likely to be reached during multi-cycle production under the conditions of salt deposition. Under the initial water condition, reservoir drying can improve the gas storage capacity. At the end of the tenth cycle, the storage capacity increases by 1.4%. It is concluded that the study on the impact of formation water evaporation on storage capacity is helpful for the prevention and control of salt formation water in UGS with high salinity.

Study on evaporation drainage of deep coal seam gas wells

Targeting the problem of a small amount of fluid accumulation in deep coal seam gas (CSG) wells during flowing production stage, the evaporation drainage method is proposed to discharge the liquid accumulation. Based on the Dalton evaporation model and wind speed function, a calculation model of evaporation drainage was established for deep CSG wells, which was verified by laboratory experiments. Taking a CSG well in the western Ordos Basin as an example to analyze the evaporation drainage capacity, the influence of temperature, daily gas production, bottomhole flowing pressure (BHFP), formation gas water saturation on the evaporation drainage capacity was investigated. The results show that the maximum evaporation water production is 2,533.8 kg/d at a bottomhole temperature of 80°C and a gas production rate of 30 × 103 m3/d. It is found that the temperature and pressure have a marked influence on the evaporation drainage. By improving the gas production and bottomhole temperature, and reducing the BHFP can effectively promote the evaporation drainage capacity. The initial moisture content of CSG in the reservoir are inversely proportional to the evaporation drainage capacity. By adjusting the BHFP and daily gas production, the evaporation drainage capacity can match the liquid production rate of the formation. Evaporation drainage can effectively extend the flowing production time of deep CSG wells and reduce the costs of production.

Numerical Simulation and Parameter Optimization for Water-to-CO2 Flooding in a Strongly Water-Sensitive Reservoir.

Carbon dioxide flooding can accelerate the development of low-permeability reservoirs of the Kexia group in the K region of the T oil field, thus resolving the issue of inadequate water drive effects. This study was focused on the well group 80513 in the K region, and based on the reservoir and fluid parameters, a simulation model of water-sensitive post-CO2 flooding was constructed to refine the gas injection strategy gradually. The injection rate of the continuous gas injection stage was preferred based on the degree of recovery. Multiindicator and multifactor injection and extraction schemes were established to optimize and analyze the key controlling factors, including the gas injection rate, gas injection period, gas-to-water ratio, and bottom-hole flow pressure, in the carbon dioxide gas-to-water alternation process. Recovery efficiency, oil exchange rate, formation pressure, and carbon dioxide storage rate were used as indicators. After 5 years of continuous CO2 flooding, the results indicated that switching to CO2 gas-water alternating flooding was more appropriate for the target block's environment. The best development plan was achieved when the gas injection rates were 1.0 and 1.25 × 104 m3·d-1 for continuous gas injection and CO2 gas-water alternating flooding, respectively, with a gas-water ratio of 1:1, a gas injection cycle of 90 days, and a bottom-hole flow pressure of 25 MPa in the production wells. A comparison between the results revealed that the formation pressure and oil recovery efficiency of this well group significantly increased upon CO2 flooding, and the parameter optimization results were well suited for controlling the gas flurry, offering a versatile model for future development of the block.

Flowing Bottomhole Pressure during Gas Lift in Unconventional Oil Wells

Summary We present artificial neural network (ANN) models for predicting the flowing bottomhole pressure (FBHP) of unconventional oil wells under gas lift operations. Well parameters, fluid properties, production/injection data, and bottomhole gauge pressures from 16 shale oil wells in Permian Basin, Texas, USA, are analyzed to determine key parameters affecting FBHP during the gas lift operation. For the reservoir fluid properties, several pressure-volume-temperature (PVT) models, such as Benedict-Webb-Rubin (BWR); Lee, Gonzalez, and Eakin; and Standing, among others, are examined against experimentally tuned fluid properties (i.e., viscosity, formation volume factor, and solution gas-oil ratio) to identify representative fluid (PVT) models for oil and gas properties. Pipe flow models (i.e., Hagedorn and Brown; Gray, Begs and Brill; and Petalas and Aziz) are also examined by comparing calculated FBHP against the bottomhole gauge pressures to identify a representative pipe flow model. Training and test data sets are then generated using the representative PVT and pipe flow models to develop a physics-based ANN model. The physics-based ANN model inputs are hydrocarbon fluid properties, liquid flow rate (qL), gas-liquid ratio (GLR), water-oil ratio (WOR), well true vertical depth (TVD), wellhead pressure (Pwh), wellhead temperature (Twh), and temperature gradient (dT/dh). A data-based ANN model is also developed based on only TVD, Pwh, qL, GLR, and WOR. Both physics- and data-based ANN models are trained through hyperparameter optimization using genetic algorithm and K-fold validation and then tested against the gauge FBHP. The results reveal that both models perform well with the FBHP prediction from field data with a normalized mean absolute error (NMAE) of around 10%. However, a comparison between results from the physics- and data-based ANN models shows that the accuracy of the physics-based model is higher at the later phase of the gas lift operation when the steady-state pipe flow is well established. On the contrary, the data-based model performs better for the early phase of gas lift operation when transient flow behavior is dominant. Developed ANN models and workflows can be applied to optimize gas lift operations under different fluid and well conditions.

Well Production Decline Models Considering Mechanical Skin Factor

Production decline analysis is an essential tool in predicting reservoir performance and in estimating reservoir properties. In its most basic form, decline curve analysis is to a large extent based on Arps’ empirical models that have little theoretical basis. But the mechanical skin factor due to formation damage is not taken into account in the production decline models available in the literature. The conventional models presented in the literature are mostly empirical and not grounded in fundamental in theory. New theoretical models are developed in this paper to forecast the production decline of a well producing in a closed circular or rectangular reservoir at constant bottomhole pressure, and mechanical skin factor due to formation damage is taken into account. Both the transient flow behavior and boundary‐dominated flow behavior are incorporated; the effects of reservoir size, well location, and mechanical skin factor on transient flow rate and pressure drop in the reservoir are studied. Analytical solutions are obtained through combinations of Green’s function and inverse Laplace transform. Stehfest’s algorithm is used to convert the obtained solution from the Laplace transform domain into the real time domain. The proposed models in this paper are based on analytical solutions to diffusivity equations in porous media, and the solutions are fast and practical tools to forecast the production performance of a well producing at constant flowing bottomhole pressure.

A New Approach to Apply Decline-Curve Analysis for Tight-Oil Reservoirs Producing Under Variable Pressure Conditions

Summary Decline-curve models inherently assume that the bottomhole flowing pressure (BHP) is constant. This is a poor assumption for many unconventional wells. For this reason, the application of decline-curve models might lead to incorrect flow regime identification and estimated ultimate recovery (EUR). This work presents a novel technique that combines variable BHP conditions with decline-curve models and compares its results with traditional decline-curve analysis (DCA) for both synthetic and tight-oil wells. Using superposition, we generate a synthetic rate example using the constant-pressure solution of the diffusivity equation for a slightly compressible fluid (decline-curve model) along with a BHP history. However, we validate the technique using bottomhole and initial reservoir pressures that contain errors. The algorithm consists of three sequential optimizations. In each optimization, the algorithm estimates (1) the decline-curve model parameters, (2) the BHP, and (3) the initial reservoir pressure. The result of the synthetic example leads to an accurate production history match and corrected estimates of the initial reservoir pressure and the BHP history. Finally, we compare the results of the technique with traditional DCA in terms of (a) the model parameters, (b) flow regime identification, (c) production history matches, and (d) EUR for tight-oil wells using three decline-curve models: 1D single-phase constant-pressure solution of the diffusivity equation for a slightly compressible fluid, logistic growth model, and Arps hyperbolic relation. For the synthetic case, the algorithm estimates the model parameters and the true initial reservoir pressure within a 2% error. In addition, the method regenerates the true BHP history and provides an excellent production history match. The analysis of the tight-oil wells shows that the new approach clearly identifies the flow regimes present in the well, which can be difficult to detect using traditional DCA when the BHP varies. In contrast, the application of traditional DCA shows considerable errors in the estimation of the model’s parameters and a poor history match of the production data. Finally, this work shows that incorporating variable BHP into the decline-curve models leads to more accurate production history matches and EUR values compared to using only rate-time data. This paper illustrates a workflow that incorporates variable BHP conditions for any decline-curve model. Moreover, the approach can handle errors in both the BHP and the initial reservoir pressure and provides corrected estimates of these variables. The technique is computationally fast and history matches and forecasts the production of unconventional wells more accurately than traditional DCA. The major contribution of this work is the remarkable simplicity yet robustness of our solution to variable-pressure DCA. Finally, we developed a web-based application to provide the readers with a hands-on experience of this new technique.

Model and Analysis of Pump-Stopping Pressure Drop with Consideration of Hydraulic Fracture Network in Tight Oil Reservoirs

The existing pump-stopping pressure drop models for the hydraulic fracturing operation of tight oil reservoirs only consider the main hydraulic fracture and the single-phase flow of fracturing fluid. In this paper, a new pump-stopping pressure drop model for fracturing operation based on coupling calculation of the secondary fracture and oil-water two-phase flow is proposed. The physical model includes the horizontal wellbore, the fracture network and the tight oil reservoir. Through the numerical simulation and calculation, the wellbore afterflow performance, the crossflow performance between the main hydraulic fracture and the secondary fracture, the fracturing fluid leakoff and the oil-water replacement after termination of pumping are obtained. The pressure drop characteristic curve is drawn out by the bottom-hole flow pressure calculated through the numerical simulation, and a series of analyses are carried out on the calculated pressure drop curve, which is helpful to diagnose the -oil-water two-phase flow state and the fracture closure performance under the control of the fracture network after hydraulic fracturing pumping. Finally, taking a multi-stage fractured horizontal well in a tight oil reservoir in the Junggar basin, China as an example, the pump-stopping pressure drop data of each stage after hydraulic fracturing are analyzed. Through the history fitting of the pressure drop characteristic curve, the key parameters such as fracture network parameters, which include the half-length of main hydraulic fracture, the conductivity of main hydraulic fracture and the density of secondary fracture, the fracture closure pressure are obtained by inversion, thus, the hydraulic fracturing effect of fractured horizontal well in tight oil reservoirs is further quantified.

A Real-Time Inversion Approach for Fluid-Flow Fractures in Unconventional Stimulated Reservoirs

Summary Fluid-flow fractures, through which fluids can move under pressure, make a more significant contribution to increasing production than do microseismic and propagation fractures. An accurate description of the distribution of fluid-flow fractures is the basis for evaluating hydraulic fracturing and oil/gas recovery. In this study, a real-time inversion approach for fluid-flow fractures was proposed, and the complex fluid-flow fracture morphology was obtained in real time by updating the data of the fracturing construction curve. First, a dynamic permeability model was proposed to describe the filtration rate of the fracturing fluid during hydraulic fracturing. Combined with the point source function, the flowing bottomhole pressure (pwf) can be quickly calculated based on the fracture morphology and displacement of the fracturing fluid. The variance of pwf and bottomhole pressure (pwb) obtained by pump pressure were used as an objective function, and the length of fluid-flow fractures and fracture morphology were used as fitting parameters. The length of the fluid-flow fractures was updated with the simultaneous perturbation stochastic approximation (SPSA) to achieve a rough fitting of the bottomhole pressure. On this basis, a probability function was used to constrain the randomness of the fractures, and the fracture morphology with a fixed fracture length was continuously simulated and finely matched. Finally, a complex fluid-flow fracture morphology was obtained. The method was used to analyze the fluid-flow fracture morphology of multifractured horizontal wells in shale reservoirs, and the fitting rate of the fracturing construction curve was more than 95%. The results show that the total length of the fluid-flow fractures in one stage in naturally fractured reservoirs was approximately 629 m, and those in homogeneous reservoirs and high-stress difference reservoirs were 564 m and 532 m, respectively. The length of fluid-flow fractures with “grooves” in the fracturing construction curve was longer than the length of fluid-flow fractures with “bulges.” The effectively stimulated reservoir area with fluid-flow fractures was only approximately 28–51% of the stimulated reservoir area with microseismic fractures.

Reservoir Analysis for A Low Recovery Factor Mature Reservoir with A Gas Cap and Large Aquifer: A Proposed Coning Correlation

With the development of reservoir engineering and management, enhanced oil recovery (EOR) techniques are becoming increasingly sophisticated. Generally, the recovery factor (RF) for a conventional reservoir can range from around 10% to 60% after four decades of production, with an average recovery factor of around 35%. While the RF of the NGJ field with high porosity (0.25) and permeability (300 to 600 md) is 6.11% after producing for over 40 years, which is rare in such good reservoir conditions. In this work, some classic techniques and several advanced methods were utilized to investigate the reason behind the low RF for this reservoir.A dynamic reservoir simulation was constructed on the basis of a statics petrophysical model. Based on the preliminary simulation study, it was found that gas coning is a common problem in the reservoir. Therefore, a concept of coning correlation function (CCF) is proposed, and the procedures for developing the CCF are explained. The critical oil rate was estimated using the Addington correlation and modified through the proposed coning correlation that varies as a function of time. The estimated critical rate through CCF is used to control flowing bottom hole pressure (FBHP) in the prediction scenarios.

An Efficient Deep Learning-Based Workflow for CO2 Plume Imaging With Distributed Pressure and Temperature Measurements

Summary Geologic carbon dioxide (CO2) sequestration has received significant attention from the scientific community as a response to global warming due to greenhouse gas emissions. Effective monitoring of CO2 plume is critical to CO2 storage safety throughout the life cycle of a geologic CO2 sequestration project. Although simulation-based techniques such as history matching can be used for predicting the evolution of underground CO2 saturation, the computational cost of high-fidelity simulations can be prohibitive. Recent development in data-driven models can provide a viable alternative for rapid CO2 plume imaging. Here, we present a novel deep learning–based workflow that can efficiently visualize CO2 plume in near real time. Our deep learning framework utilizes field measurements, such as downhole pressure, distributed pressure, and temperature, as input to visualize the subsurface CO2 plume images. However, the high output dimension of CO2 plume images makes the training inefficient. We address this challenge in two ways: First, we output a single CO2 onset time map rather than multiple saturation maps at different times; second, we apply an autoencoder-decoder network to identify lower-dimensional latent variables that compress high-dimensional output images. The “onset time” is the calendar time when the CO2 saturation at a given location exceeds a specified threshold value. In our approach, a deep learning–based regression model is trained to predict latent variables of the autoencoder-decoder network. Subsequently, the latent variables are used as inputs of the trained decoder network to generate the 3D onset time image, visualizing the evolving CO2 plume in near real time. The power and efficacy of our approach are demonstrated using both synthetic and field-scale applications. We first validate the deep learning–based CO2 plume imaging workflow using a 2D synthetic example. Next, the visualization workflow is applied to a 3D field-scale reservoir to demonstrate the robustness and efficiency of the workflow. The monitoring data set consists of distributed temperature sensing (DTS) data acquired at a monitoring well, flowing bottomhole pressure (BHP) data at the injection well, and time-lapse pressure measurements at several locations along the monitoring well. Our approach is also extended to efficiently evaluate the uncertainty of predicted CO2 plume images. Additionally, an efficient workflow for optimizing data acquisition and measurement type is demonstrated using our deep learning–based framework. The novelty of this work is the development and application of a unique and efficient deep learning–based subsurface visualization workflow for the spatial and temporal migration of the CO2 plume. The efficiency and flexibility of the data-driven workflow make our approach suitable for field-scale applications.

Development of hybrid computational data-intelligence model for flowing bottom-hole pressure of oil wells: New strategy for oil reservoir management and monitoring

Among several metric parameters concerning the assessment of oil and gas well production, the flowing bottom-hole pressure (FBHP) is considered essential. Accurate prediction of FBHP is crucial for petroleum engineering and management. Several related parameters are associated with the FBHP magnitude influence; thus, proper inspection of those parameters is another vital concern. This research proposes a hybrid modeling framework based on the hybridization of machine learning (ML) models (i.e., Extreme Learning Machine (ELM), Support Vector Machine Regressor (SVR), Extreme Gradient Boosting (XGB), and Multivariate Adaptive Regression Spline (MARS)) and nature-inspired Differential Evolutional (DE) optimization for FBHP prediction. The adjustment of the internal parameters of the ML-based models and the input feature selection is formulated as an incremental learning problem that is solved by the evolutionary algorithm. Problem-specific samples were collected from the open-source literature for this investigation. Modeling results are adaptable, automatically determining the most relevant variables for the context of the ML model. The adaptive polynomial structure of hybridized MARS model attained the best average performance for the FBHP modeling with correlation (R = 0.94) and minimum root mean square (RMSE = 97.88). The proposed modeling framework produces an alternative efficient computer aid model for FBHP prediction, resulting in reliable automated technology to assist oil and gas well management.

Limitations of rate normalization and material balance time in rate-transient analysis of unconventional reservoirs

Rate-transient analysis (RTA) is routinely used for unconventional production data analysis and forecasting. This technique uses rate normalization along with material balance time to estimate: (a) flow regimes, (b) the time of end of transient flow, and (c) the drainage volume and to thereby predict the estimated ultimate recovery (EUR) of wells. However, rate normalization approximates deconvolution and material balance time is only strictly applicable for constant-pressure or constant-rate systems. This work investigates the validity of rate normalization and material balance time for synthetic and tight-oil well examples producing under variable bottomhole flowing pressure (BHP) conditions.The results of this study show that BHP changes alter the slope of the log-log normalized rate-vs. material balance time plot. In addition, BHP variations introduce error to the behavior of the material balance time vs. time function. Consequently, normalized-pressure rate and material balance time are not always reliable variables to properly identify the flow regime(s) and thus, to correctly estimate the time of end of transient flow and EUR. Alternatively, applying a deconvolution as the one presented in this paper to rigorously account for the pressure variations and generate the unit-pressure-drop rate and the constant-pressure material balance time solves these problems.

Uninterrupted lift of gas, water, and fines in unconventional gas wells using foam-assisted artificial lift

Sustainable and clean production of natural gas is of paramount importance in transitioning to a low-emission economy. Gas production in an unconventional well such as a coal seam gas (CSG) relies on continuous dewatering, commonly performed through a progressive cavity pump (PCP). PCPs are susceptible to failure for different reasons, primarily the interference of solid particles. This results in interruptions, pump replacement, well cleaning and recommissioning which pose long downtimes (i.e., reduced well deliverability), work-over at substantial costs, along with environmental impacts such as increased methane emissions to the atmosphere.This paper aims to demonstrate and assess the performance and suitability of a sustainable dewatering technique, referred to as foam-assisted gas lift (FAGL) which offers uninterrupted dewatering, minimal downtime and minimised environmental impacts. FAGL is capable of maintaining the gas well deliverability by replacing the downhole pump with a gas compressor at the surface and an engineered gas-liquid-solid flow dynamics by injecting foamers and a small amount of surplus gas into the wellbore.The performance of FAGL is evaluated in 45 vertical CSG wells in Australia under different operating conditions and depths ranging between 400 and 1050 m and is compared with well data using PCPs. The flowing bottomhole pressure (FBHP) is calculated to assess the suitability of FAGL under different operating conditions. The results from this study show that FAGL outperforms PCPs in more than 50% of the cases under this study, offering on average 34% lower FBHP. The FAGL's capability in lifting solid particles is demonstrated using solids taken from gas wells in Queensland, showing a reliable and continuous lift of particles. In addition, chemical defoaming of the foam at the surface is demonstrated to avoid re-foaming in the downstream and mitigate the risk of contamination of the surrounding area.

An adaptive neuro-fuzzy inference system white-box model for real-time multiphase flowing bottom-hole pressure prediction in wellbores

The majority of published empirical correlations and mechanistic models are unable to provide accurate flowing bottom-hole pressure (FBHP) predictions when real-time field well data are used. This is because the empirical correlations and the empirical closure correlations for the mechanistic models were developed with experimental datasets. In addition, most machine learning (ML) FBHP prediction models were constructed with real-time well data points and published without any visible mathematical equation. This makes it difficult for other readers to use these ML models since the datasets used in their development are not open-source. This study presents a white-box adaptive neuro-fuzzy inference system (ANFIS) model for real-time prediction of multiphase FBHP in wellbores. 1001 real well data points and 1001 normalized well data points were used in constructing twenty-eight different Takagi–Sugeno fuzzy inference systems (FIS) structures. The dataset was divided into two sets; 80% for training and 20% for testing. Statistical performance analysis showed that a FIS with a 0.3 range of influence and trained with a normalized dataset achieved the best FBHP prediction performance. The optimal ANFIS black-box model was then translated into the ANFIS white-box model with the Gaussian input and the linear output membership functions and the extracted tuned premise and consequence parameter sets. Trend analysis revealed that the novel ANFIS model correctly simulates the anticipated effect of input parameters on FBHP. In addition, graphical and statistical error analyses revealed that the novel ANFIS model performed better than published mechanistic models, empirical correlations, and machine learning models. New training datasets covering wider input parameter ranges should be added to the original training dataset to improve the model's range of applicability and accuracy.