|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
Chen, Cunliang (Tianjin Branch of CNOOC, China Co., Ltd) | Yang, Ming (Tianjin Branch of CNOOC, China Co., Ltd) | Han, Xiaodong (Tianjin Branch of CNOOC, China Co., Ltd) | Zhang, Jianbo (China University of Petroleum, East China)
Managing oil production from reservoirs to maximize the future economic return of the asset is an important issue in petroleum engineering. One of the most important problems is the prediction of water flooding performance. Traditional strategies have been widely used with a long run time and too much information to solve this problem. Therefore, it is urgent to form a fast intelligent prediction method, especially with the development of large data processing and artificial intelligence methods.
This paper proposed a new method to predict water flooding performance using big data and artificial intelligence algorithms. The method regards layered reservoir as a vertical superposition of a series of single layer reservoirs. An injection-production analysis model is established in each single layer reservoir respectively. And then a superposition model is established only by production data and logging tools data. Finally, the least square principle and the particle swarm optimization algorithm are used to optimize the model and predict water flooding performance.
This method has been tested for different synthetic reservoir case studies. The results are in good agreement in comparison with the numerical simulation results. The average relative error is 4.59%, but the calculation time is only 1/10 of that of numerical simulation by using artificial intelligence method. It showed that this technique has capability to predict water flooding performance. These examples showed that the use of artificial intelligence method not only greatly shortens the working time, but also has a higher accuracy.
By this paper, it is possible to predict the water flooding performance easily and accurately in reservoirs. It has an important role in the field development, increasing or decreasing investment, drilling new wells and future injection schedule.
Time-lapse seismic monitoring is a powerful technique for reservoir management and the optimization of hydrocarbon recovery. In time-lapse seismic datasets, the difference in seismic properties across different vintages enables the detection of spatio-temporal changes in saturated properties and structure induced by production. The main objectives are (1) to identify bypass pay zones in time-lapse seismic data for the deepwater Amberjack field, located in the Gulf of Mexico, (2) confirm the identified bypass pay zones in the results of reservoir simulation, and (3) recommend well planning strategies to exploit these bypassed resources.
A high-fidelity seismic-to-simulation 4D workflow that incorporates seismic, petrophysics, petrophysical property modeling, and reservoir simulation was employed, which leveraged cross-discipline interaction, interpretation, and integration to extend asset management capabilities. The workflow addresses geology (well log interpretation and framework development), geophysics (seismic interpretation, velocity modeling, and seismic inversion), and petrophysical property modeling (earth models and co-located co-simulation of petrophysical properties with P-impedance from seismic inversion). An embedded petro-elastic model (PEM) in the reservoir simulator is then used to affiliate spatial dry rock properties with saturation properties to compute dynamic elastic properties, which can be related to multi-vintage P-impedance from time-lapse seismic inversion. In the absence of the requisite dry rock properties for the PEM, a small data engine is used to determine these absent properties using metaheuristic optimization techniques. Specifically, two particle swarm optimization (PSO) applications, including an exterior penalty function (EPF), are modified resulting in the development of nested and average methods, respectively. These methods simultaneously calculate the missing rock parameters (dry rock bulk modulus, shear modulus, and density) necessary for dynamic, embedded P-impedance calculation in the history-constrained reservoir simulation results. Afterward, a graphic-enabled method was devised to appropriately classify the threshold to discriminate non-reservoir (including bypassed pay) and reservoir from the P-impedance difference. Its results are compared to unsupervised learning (k-means clustering and hierarchical clustering). From seismic data, one can identify bypassed pay locations, which are confirmed from reservoir simulation after conducting a seismic-driven history match. Finally, infill wells are planned, and then modeled in the reservoir simulator.
Zhou, Xingyuan (China University of Petroleum-Beijing) | Liang, Yongtu (China University of Petroleum-Beijing) | Di, Pengwei (China University of Petroleum-Beijing) | Xiang, Chengcheng (University of Southern California) | Xin, Shengchao (China University of Petroleum-Beijing) | Zhang, Haoran (China University of Petroleum-Beijing)
As the most commonly used process for the secondary development of oilfields, waterflooding plays a significant role in maintaining reservoir pressure, enhancing oil recovery and achieving high and stable oil production. The previous waterflooding optimization studies usually worked out the optimal injection/production rates but didn't take into account the energy consumed by the surface waterflooding pipeline network system which transfers water from the waterflooding stations to the waterflooding wells. Taking the maximum waterflooding development profit as the objective function, this paper proposes an integrated methodology for the unified optimization of injection/production rates and the operation control of surface waterflooding pipeline network system. The objective function is defined as the oil production income minus the operation cost of the pipeline network. With a given set of injection rates of waterflooding wells, the reservoir numerical simulation is employed to obtain the oil production rates and a mixed integer nonlinear programming (MINLP) model is established for the optimal operation control of the surface waterflooding pipeline network, including the pump schedule of waterflooding stations, flowrate of pipe segments and pressure at each node. A hybrid solving strategy incorporating particle swarm optimization (PSO), linear approximation method, and branch-and-bound algorithm, is proposed for solving the results. The PSO algorithm is adopted to search for the optimal injection rates of waterflooding wells, while the linear approximation method and branch-and-bound algorithm are used for the MINLP model solving. In this study, we took the Daqing waterflooding Oilfield in China as an example. The applicability of the methodology and the stability of the solving strategy are illustrated in detail. It is proved that the proposed methodology could provide the engineers with significant guidelines for the unified optimization of waterflooding process incorporating the reservoir and surface pipeline network.
After Fan introduced the concept of synthetic aperture (SA) to marine controlled-source electromagnetics (CSEM), numerous of optimization methods are applied to SA weights selection for improving the detectability of deeply buried targets under seabed, but few are suitable for the seriously non-linear EM problems. This study presents an application of particle swarm optimization (PSO) to optimization of the phase shift and the amplitude compensation coefficient of SA for marine CSEM. Eigenstate analysis (EA), another non-linear SA weights optimization method, is also carried out in marine CSEM data processing for comparison. The 3D synthetic model reconstructed from Fan (2010) is used to better demonstrate the effects of the detectability with and without two optimization algorithms. In order to validate the effectiveness of PSO, we scan all the confined weights. The results of PSO converge to the global maximum in a robust and fast way.
Presentation Date: Monday, October 15, 2018
Start Time: 1:50:00 PM
Location: Poster Station 13
Presentation Type: Poster
Results are presented on the application of an optimization solver for full-waveform inversion (FWI) in a synthetic microseismic monitoring scenario. The optimization solver is based on a heuristic algorithm that does not require knowledge about the gradient of the cost function. The optimized variables are the sources origin times, locations and moment tensors, and the earth model velocities, Thomsen parameters and depths of interfaces. The application to noise-free data offers encouraging results to continue the assessment of the algorithm in more realistic scenarios of microseismic monitoring.
Presentation Date: Tuesday, October 16, 2018
Start Time: 8:30:00 AM
Location: 208A (Anaheim Convention Center)
Presentation Type: Oral
Ibiam, Emmanuel (Heriot-Watt University) | Geiger, Sebastian (Heriot-Watt University) | Almaqbali, Adnan (Heriot-Watt University) | Demyanov, Vasily (Heriot-Watt University) | Arnold, Dan (Heriot-Watt University)
The average global recovery factor of a typical oil and gas field is approximately 40% at after secondary recovery processes such as gas lifting and water flooding. The low recovery factor is often a result of by passing a considerable amount of oil in the reservoir due to unknown reservoir heterogeneity and incomplete understanding of the geology. Enhanced oil recovery (EOR) methods will play a key role in increasing recovery factors from existing reservoirs. Heterogeneous reservoir sands often show permeability contrast between layers and therefore lead to early water breakthrough when water flooding. In such situations, polymer flooding can potentially be a suitable EOR technique that helps to lower water cut and increase recovery. The addition of polymers to the injected water lowers the mobility ratio, thereby reducing viscous fingering and delaying water breakthrough.
This study investigates how a polymer flood design can be optimized while considering geological uncertainty in the reservoir models as well as modelling decisions. We applied an adjoint based technique to match data from a prolonged waterflood in the Watt Field, a synthetic but realistic clastic reservoir that is based on real data and captures a wide range of geological heterogeneities and uncertainties through a range of different model scenarios and model realizations. We apply Latin hypercube experimental designs with the Particle Swarm Optimization algorithm in CMOST. This was used to build a proxy model employing polynomial regression for the optimization of the engineering parameters to maximize NPV. The optimization were performed for both history-matched models (constrained optimization) and the original, non-history-matched models (unconstrained optimization). The aim of this work is to analyse how geological uncertainties inherent to a heterogeneous clastic reservoir as well as modelling decisions impact the design and performance a polymer flood. We further investigate how the different optimization methods impact the predicted reservoir performance and optimal design of the polymer flood.
Our findings show that both, geological and engineering uncertainties, impact polymer flooding and that designing the right well controls is essential for successful polymer flooding. Shale cut-offs are identified as a key petrophysical uncertainty when optimizing a polymer flood in a heterogeneous clastic reservoir. Furthermore, forecasts using constrained optimization yielded a much narrower range of incremental oil recovery and NPV during polymer flooding and may underestimate both, risk and opportunities for polymer flooding because the history matching of the water flood emphasizes different geological features compared to the way geology interacts with a more viscous polymer solution.
Moussa, Tamer (King Fahd University of Petroleum and Minerals) | Patil, Shirish (King Fahd University of Petroleum and Minerals) | Mahmoud, Mohamed (King Fahd University of Petroleum and Minerals) | Elkatatny, Salaheldin (King Fahd University of Petroleum and Minerals)
Determination of optimal well locations plays an important role in the efficient recovery of hydrocarbon resources. However, it is a challenging and complex task because it relies on reservoir, and fluid and economic variables that are often nonlinearly correlated. Traditionally, well placement optimization (WPO) has been done through experience and use of quality maps. However, reservoir management teams are beginning to appreciate the use of automatic optimization tools for well placement that will yield the largest financial returns or highest net present value (NPV). In addition, the performance of a reservoir is time and process dependent, therefore well placement decisions cannot be based on static properties alone. On the other hand, well placement optimization requires a large number of simulator runs in an iterative process, and thus several runs to reach the maximum achievable NPV. Therefore, there is a real need for automatic well placement approach that uses highly efficient optimization method, which can improve the result quality, speed of the convergence process to optimal result and thus decrease the time required for computation.
The objective of this work is to determine the optimal well locations in a heavy oil reservoir under production using a novel recovery process, in which steam is generated, in-situ, using thermochemical reactions. Self-adaptive differential evolution (SaDE) and particle swarm optimization (PSO) methods are used as the global optimizer to find the optimal configuration of wells that will yield the highest NPV. Comparison analysis between the two proposed optimization techniques is introduced. The CMG STARS Simulator is utilized in this research to simulate reservoir models with different well configurations.
Comparison of results is made between the NPV achieved by the well configuration proposed by the SaDE and PSO methods. The results show that SaDE performed better than PSO in terms of higher NPV after ten years of production while under in-situ steam injection process using thermochemical reactions. This is the first known application where SaDE and PSO methods are used to optimize well locations in a heavy oil reservoir that is recovered by injecting steam generated in-situ using thermo-chemical reactions. This research shows the importance of well placement optimization in a highly promising and novel heavy oil recovery process. This also is a step forward in the direction to eliminate the CO2 emissions related to thermal recovery processes.
In heterogeneous reservoir formations, water tends to have early breakthrough due to the overriding and viscous fingering during secondary recovery. The overall hydrocarbon recovery efficiency remains very low in gas and water flooding projects because of less viscosity and higher mobility of water and gas. Therefore, there is an underlying need for improving recovery through a suitable chemical enhanced oil recovery (EOR) method. After investigating the feasibility of alkaline, polymer, surfactant, surfactant-polymer, alkaline-polymer and alkaline-surfactant-polymer (ASP) flood, ASP was selected as a chemical EOR method in low permeability heterogeneous reservoirs. However, the performance of the ASP flooding was highly dependent on operational parameters. Thus, it was important to select these parameters with extensive care to increase the recovery along with the profitability. The relationship between the ASP flooding operational parameters and profitability (NPV) has not been yet understood fully.
In this research, the new stochastic optimization approach to optimize the ASP flooding operational parameters has been proposed. To gain the objective of this research, a numerical simulation study was carried out and Particle Swarm Optimization (PSO) was implemented as an optimization algorithm. The net present value (NPV) served as the objective function that has to be maximized among the compared flooding processes. The used operational parameters were location of production and injection well, number of injection cycles, oil production rate, ASP injection time, ASP injection rate, alkaline-surfactant and polymer concentrations, surfactant and polymer viscosities. Sensitivity study of these parameters shows significant impact on net present value and ultimate oil recovery. Results also confirm that NPV is increased significantly after the optimization of all flooding parameters by using Particle Swarm Optimizer.
The new optimized model was developed for designing the ASP as a chemical EOR method in low permeability heterogeneous reservoir. It can be served as a handy tool for reservoir engineer to select the best ASP flood parameters to achieve maximum NPV.
After the natural depletion of the reservoir, water injection is the most commonly used improved oil recovery (IOR) method. The produced formation water is injected back to reservoir for displacing oil and maintaining pressure in the reservoir, however in low permeability reservoirs the hydrocarbon recovery after primary, and secondary recovery techniques remains very low because of gas conning and water viscous fingering. In principal water flooding can increase the recovery up to 1 to 10% (Brundred and Brudred, 1955). Therefore, there is an immense need to apply any suitable tertiary recovery method to recover the trapped oil from low permeable sands. The most critical parameters for the screening of any enhanced oil recovery EOR method are; reservoir temperature, formation water salinity, clay contents, oil viscosity, and formation permeability (Delamaide et al., 2014). Due to the limitation of thermal methods in low permeability heterogeneous reservoirs, chemical methods are often considered more favorable.
We present hybrid derivative free algorithm methods that maximize NPV while solving the problem of placement of hydraulic fracturing stages and horizontal wellbore trajectory in a tight heterogeneous gas condensate reservoir. These parameters are important in reservoir management and field development optimization as they determine the interaction of the wellbore with the reservoir. Thus, they have a large impact on production performance of the field and the cost of field development.
We couple a tight heterogeneous gas condensate reservoir model to an optimization algorithm that determines the parameters that maximize the economic value of the field. The optimization algorithm proposed in this paper exploits the advantage of stochastic global search (particle swarm optimization) and local pattern search (hill climber) techniques to find the optimized parameters. The optimization process starts with the particle swarm optimization (PSO) algorithm, which is executed with an initial guess based on engineering experience until the objective function (in this study NPV) fails to improve in the next couple of iterations. This is followed by the hill climber (HC) algorithm that improves the objective function.
Observations from our investigation show that an optimal field development plan (FDP) is essential to optimize the placement of hydraulic fracturing stages along a horizontal wellbore and to optimize the wellbore trajectory inside the reservoir. However, these parameters would have to be optimized simultaneously in a non-systematic manner. We integrate reservoir engineering experience and economics knowledge into the optimization algorithm by embedding practical constraints into the problem formulation.
The algorithm is executed in a reasonable amount of computation time, considering the complexity of the problem. The optimization process evaluates various possible field development plans involving different hydraulic fracturing stages and well placement trajectories. The investigation demonstrates how these parameters impact the economic value of the field, how to optimize the placement of hydraulic fracture stages along a horizontal wellbore, and how to optimize the wellbore trajectory inside the reservoir.
The methodology presented in this work should allow industry professionals working with unconventional reservoirs to improve the economic value of a field in a shorter timeframe while considering all possible field development plans, a task that would be time consuming and tedious if carried out manually.
We present an approach for estimating in-situ relative permeability and capillary pressure through the joint inversion of array resistivity logging and formation test data. Considering a scenario of drilling a vertical well into an oil-bearing formation with water-based mud, the mud-filtrate invasion process can be regarded as a controlled experiment under reservoir conditions. Array resistivity logging can sense the formation resistivity perturbed by the two-phase flow invasion. Formation testing with fluid sampling can also provide information on the radially varying saturation and the associated changes in mobility, as well as information on the effect of capillary pressure. A facies-based workflow is developed to invert for the relative permeability and capillary pressure from the abovementioned two data sets. The inversion strategy is adjustable based on a sensitivity analysis as well as on the data available and the operational sequence of collecting the data. A hybrid inversion framework combining deterministic and stochastic optimization approaches is developed for the inversion of the data.