|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
Abstract Reservoir simulation is an important tool used in the industry for reservoir management. While developing a field, a reservoir simulation model is used as a decision tool to select the best development scheme and also to forecast the oil, gas, and water production expected for the field. Uncertainties are much higher at the early phases and, when production data are gathered during the field development phase, most of the time the initial reservoir simulation model needs to be reviewed once the field observed data is not as the same as the predicted by the model. Some of these uncertainties of these input parameters are related to the reservoir rock reservoir heterogeneities. History matching techniques are used by reservoir engineers to mitigate/minimize the difference between the observed field data and the predicted data and thus assessing the uncertainties. When reservoir models become too big in terms of number of cells and features, the elapsed simulation time increases very much, making the history matching process very cumbersome and, in some cases, very difficult to achieve in an acceptable time. Parallel processing features of some commercial simulators can perform lots of simulation runs at the same time but cannot address and cannot solve the problem in a proper way. This paper presents an alternative proposition to speed up the history matching process: the application of feed-forward neural networks as nonlinear proxies of reservoir simulation. Neural networks can map the response surface in multidimensional spaces of a reservoir model (i.e. water production, bottom hole pressure etc.) or of an objective function with few number of simulations. The mapped response is then used as a substitute of reservoir simulation runs during the history matching process. The focus of this work is to shown the steps of choosing the best number of hidden layers, the neurons and the training method. An application case is presented using the workflow presented is this work and showing the validity of the proposed methodology for this complex nonlinear problem. Introduction Reservoir simulation is an important tool used in the industry for reservoir management. While developing a field, a reservoir simulation model is used as a decision tool to select the best development scheme and also to forecast the oil, gas, and water production expected for the field. Uncertainties are much higher at the early phases and, when production data are gathered during the field development phase, most of the time the initial reservoir simulation model needs to be reviewed once the field observed data is not as the same as the predicted by the model. Some of these uncertainties of these input parameters are related to the reservoir rock reservoir heterogeneities. History matching techniques are used by reservoir engineers to mitigate/minimize the difference between the observed field data and the predicted data and thus assessing the uncertainties. When reservoir models become too big in terms of number of cells and features, the elapsed simulation time increases very much, making the history matching process very cumbersome and in some cases very difficult to achieve in an acceptable time. Mohaghegh (2006) presents a simulation field case with 165 wells and one million grid blocks where the elapsed simulation time is about ten hours in a cluster of twelve 3,2 GHz processors.
This article, written by Technology Editor Dennis Denney, contains highlights of paper SPE 101233, "Assisted History Matching for Surface-Coupled Gas-Reservoir Simulation," by D. Biswas, SPE, SiteLark, prepared for the 2006 SPE Asia Pacific Oil and Gas Conference and Exhibition, Adelaide, Australia, 11–13 September. The delivery-point pressures for gas reservoirs are known, and operators are obligated to supply gas into those downstream pressures. Therefore, reservoir-deliverability prediction must account for the pressure drop in the surface network. Also, this coupled system should honor historical attributes of pressures and rates. In this study, a modified Gauss-Newton method was used in conjunction with a nonlinear parameter-estimation algorithm to history match a surface/reservoir-coupled gas-reservoir simulation. Introduction A control-volume finite-element based reservoir simulator can use unstructured grids to analyze near-wellbore and high-activity flow areas (i.e., high-permeability channels and fractures). The pressure drop in the surface network is modeled with Weymouth's equation of steady-state pipeline flow. More importantly, the total system is solved in a coupled fashion, thereby enabling control of the solution with surface-pressure constraints. In addition, both surface and reservoir decision variables are estimated with a modified Gauss Newton algorithm in the assisted history-matching step. Background Previously published methods investigate and apply various numerical experiments of coupling reservoir and surface models. Coupling explicit vs. implicit techniques, full-system-solve vs. domain decomposition, and applications to large fields to optimize production and maximize surface-facility use have been examined extensively. However, time-consuming calibration to known (observed) data points (i.e., history-matching process) is not emphasized. Much time and effort is spent to history match and condition reservoir models to dynamic data. However, past performance of history-matched predictions has been erratic for several reasons. In brute-force history-matching procedures, the reservoir model is adjusted in an ad hoc fashion until the available production-history data are matched. In the process, the model may cease to be consistent with the prior geological model. History-matching algorithms presume that the deviation of the predicted response from the observations is attributable to only a small set of reservoir variables, such as permeability or porosity. Actually, the production response is affected by many variables (e.g., relative permeability, capillary pressure, and fractures). Uncertainty in reservoir-parameter distribution and the influence on the production response should be considered in more detail. The trend has been to start with a single reservoir model and perform history matching to obtain a single perturbed reservoir model. The history-matched model is not unique, and a suite of models is possible, each satisfying the available production data. To represent uncertainty in predicted response realistically, the predictions obtained from several history-matched realizations must be aggregated. Therefore, a new method was proposed in which the underlying heterogeneity was conditioned to the production response. The proposed approach focuses predominantly on gas reservoirs for which the number of decision variables (parameters) is very low and benefits of cost-effective history matching are high. History matching with this reduced variable set could reduce inaccuracies in the predictions for future performance.
This paper presents the preparation of a model of the Hoadley- Westerose gas reservoir and surface pipeline network using a single phase 3-Dimensional reservoir / single phase network simulator. The 3-D reservoir description was preferable to the alternative tank type approach based on the dual permeability nature of the reservoir and the desire to calculate lease line migration. In spite of the complex reservoir description, the 3-D history match was not difficult to obtain, and computing times were very acceptable. The resulting model was found to be a reasonable representation of the network, even with the approximations which were required to treat the network as a single phase system. A previous multiphase model of the network did not provide a better match of the system.
The Hoadley-Westerose Glauconitic 'A' Pool is a lean retrograde condensate reservoir which has been developed with 1 00 wells and covers approximately 180,000 acres. The involvement of multiple operators has resulted in a competitive drainage situation and a complex surface gathering system. The reservoir is characterized by several high permeability aeolian sand bars with low permeability shut-in sand reservoir between the bars. The network is complicated by flow splitting, in-line compressors, externally sourced gas, and multiple operators /flow lines.
Previous studies concentrated on multiphase network flow calculations while treating the reservoir as a series of single phase tanks. In view of the dual permeability nature of the reservoir posed by the sand bars, a more reliable representation of the reservoir was warranted. In this study, the pressure performance of the pool was successfully modelled with a 3-Dimensional gridded reservoir description. In addition to honouring the dual permeabilities, the 3-D model enabled a calculation of lease-line migration. A single phase reservoir approach was validated by phase behaviour calculations.
Although liquids are known to occur in the network, particularly at start-up conditions, the associated pressure drops can not be completely matched by multi-phase calculations. The single phase network model used in this study provided a good match of the observed network pressures. The larger than expected pressure drops were modelled with increased frictional losses and increased lengths.
The fully integrated reservoir / network model as calibrated for the Hoadley-Westerose system will provide a tool for investigating network modifications, such as line looping and adding compression, and the further development of the reservoir, including infill drilling in specific locations.
Masoudi, Rahim (Petronas MPM) | Mohaghegh, Shahab D. (West Virginia University) | Yingling, Daniel (Intelligent Solutions, Inc.) | Ansari, Amir (Intelligent Solutions, Inc.) | Amat, Hadi (Petronas MPM) | Mohamad, Nis (Petronas COE) | Sabzabadi, Ali (Petronas MPM) | Mandel, Dipak (Petronas COE)
Using commercial numerical reservoir simulators to build a full field reservoir model and simultaneously history match multiple dynamic variables for a highly complex, offshore mature field in Malaysia, had proven to be challenging, manpower intensive, highly expensive, and not very successful. This field includes almost two hundred wells that have been completed in more than 60 different, non-continuous reservoir layers. The field has been producing oil, gas and water for decades. The objective of this article is to demonstrate how Artificial Intelligence (AI) and Machine Learning is used to build a purely data-driven reservoir simulation model that successfully history match all the dynamic variables for all the wells in this field and subsequently used for production forecast. The model has been validated in space and time.
The AI and Machine Learning technology that was used to build the dynamic reservoir simulation and modeling is called spatio-temporal learning. Spatio-temporal learning is a machine-learning algorithm specifically developed for data-driven modeling of the physics of fluid flow through porous media. Spatio-temporal learning is used in the context of Deconvolutional Neural Networks. In this article Spatio-temporal Learning and Deconvolutional Neural Networks will be explained. This new technology is the result of more than 20 years of research and development in the application of AI and Machine Learning in reservoir modeling. This technology develops a coupled reservoir and wellbore model that for this particular oil & gas field in Malaysia uses choke setting, well-head pressure and well-head temperature as input and simultaneously history matches Oil production, GOR, WC, reservoir pressure, and water saturation for more than a hundred wells through a completely automated process.
Once the data-driven reservoir model is developed and history matched, it is blind validated in space and time in order to establish a reliable and robust reservoir model to be used for decision making purposes and opportunity generation to maximise the field value. The concepts and the methodology of history match of multiple wells, individual offshore platforms, and the entire field will be presented in this article along with the results of blind validation and production forecasting. Results of using this model to perform uncertainty quantification will also be presented.
A case study of a highly complex mature field with large number of wells and years of production has been used to be studied and simulated by this data-driven approach. Time, efforts, and resources required for the development of the dynamic reservoir simulation models using AI and Machine Learning is considerably less than time and resources required using the commercial numerical simulators. It is validated that the TDM developed model can make very reasonable prediction of field behavior and production from the existing wells based on modification of operational constraints and can be a prudent complementary tool to conventional numerical simulators for such complex assets.
Abstract Germik, a mature heavy oil field in Southeast Turkey, has been producing for more than 60 years with a significant decline in pressure and oil production. To predict future performance of this reservoir and explore possible enhanced oil recovery (EOR) scenarios for a better pressure maintenance and improved recovery, generation of a representative dynamic model is required. To address this need, an integrated approach is presented herein for characterization, modeling and history matching of the highly heterogeneous, naturally fractured carbonate reservoir spanning a long production history. Hydraulic flow unit (HFU) determination is adopted instead of the lithofacies model, not only to introduce more complexity for representing the variances among flow units, but also to establish a higher correlation between porosity and permeability. By means of artificial intelligence (AI), existing wireline logs are used to delineate HFUs in uncored intervals and wells, which is then distributed to the model through stochastic geostatistical methods. A permeability model is subsequently built based on the spatial distribution of HFUs, and different sets of capillary pressures and relative permeability curves are incorporated for each rock type. The dynamic model is calibrated against the historical production and pressure data through assisted history matching. Uncertain parameters that have the largest impact on the quality of the history match are oil-water contact, aquifer size and strength, horizontal permeability, ratio of vertical to horizontal permeability, capillary pressure and relative permeability curves, which are efficiently and systematically optimized through evolution strategy. Identification and distribution of the hydraulic units complemented with artificial neural networks (ANN) provide a better description of flow zones and a higher confidence permeability model. This reduces uncertainties associated with reservoir characterization and facilitates calibration of the dynamic model. Results obtained from the study show that the history matched simulation model may be used with confidence for testing and optimizing future EOR schemes. This paper brings a novel approach to permeability and HFU determination based on artificial intelligence, which is especially helpful for addressing uncertainties inherent in highly complex, heterogeneous carbonate reservoirs with limited data. The adopted technique facilitates the calibration of the dynamic model and improves the quality of the history match by providing a better reservoir description through flow unit distinction.