Specic experiments have been designed and the experimental measurements obtained show that, not only the absolute permeability but also the gas relative permeability are sensitive to connement and that the residual gas saturation (through permeability "jail") increases with loading. This observation represents an additional source of complexity in the evaluation of low-permeability sandstone gas reservoirs. INTRODUCTION Low-permeability sandstone gas reservoirs, also called tight reservoirs, are generally considered stress-sensitive reservoirs. Numerous laboratory tests under single-phase ow have shown that the absolute permeability of these reservoir rocks decreases strongly with connement. This dependence on connement is attributed to the existence of joints and interfaces in tight rocks, which close when loading increases, as pointed out by Walsh and Brace (1984) and Warpinski and Teufel (1992).
ABSTRACT: This experimental study on rock salt is dedicated to the measurements of water and gas permeability under confinement and to a characterization of the poromechanical behavior. Poromechanical experiments with gas highlight the effect of an internal fluid pressure on mechanical behavior. These experiments show indeed that the salt rock skeleton is coupled to internal fluid pressure: mechanical measurements under isotropic loading allow to estimate an effective stress and define a Biot coefficient for rock salt. Its value is far below usual values measured for sandstones, shales or other sedimentary rocks.
From an engineering perspective, rock salt can be considered as a tough impermeable rock. In the case of salt caverns for underground gas storage or for brine production, rock salt mass is indeed a strong permeability barrier.
Tightness well tests (Berest et al., 2001) and abandonment experiments of salt caverns (Karami-Jafari, 2007) show that, in these particular situations, salt can be regarded as a very low permeable medium, which permeability cannot be neglected. For example, in the case of the abandonment of a cavern filled with brine, the cavern fluid pressure starts to increase, after the cavern has been sealed due to salt creep and cavern shrinkage. However, equilibrium is reached, when salt mass creep is balanced by brine permeation through the cavern wall.
The objective of this experimental work is thus on one hand to characterize permeation and specify the effect of the confinement on single phase salt permeability -considering successively gas permeability and brine permeability- and on the other hand to examine the effect of an internal fluid pressure on the mechanical behavior of the salt. A specific experimental set-up in order to perform both hydraulic and poromechanical measurements has been developed and is presented in this paper.
Rock salt permeability has been examined in many studies, mainly related to development of salt caverns for underground storage; measurements are based on laboratory tests and in-situ tests (see for example Stormont 2001, Pusch et al. 2002, Berest and de Greef 2001) have been studied. Gas permeability under confinement is first investigated. As for other geomaterials such as concrete, argillite or shale, permeability is strongly sensitive to loading. The experimental set-up has also been adapted to measure permeability to brine. The brine permeability may be at the limit of what it can be measured in lab with confidence and is several orders of magnitude higher for gas than that of water.
The field development phase prior to investment sanction is characterized by relatively large uncertainties at the time important decisions have to be made. It is, for instance, crucial to select an appropriate recovery strategy (depletion or injection) to obtain optimal hydrocarbon cumulative production whilst ensuring good profitability of the project. Evaluation of reservoir as well as economic uncertainties and quantification of their impact are needed before the field development concept selection.
This paper describes how to stochastically assess reservoir and economic uncertainties and the screening process used to select the best recovery strategy. The chosen methodology is the combination of uncertainty studies, including both continuous, discrete and controllable parameters. The different screened scenarios are combined in a stochastic decision tree, built-up through decision and chance nodes, to establish a distribution of recoverable volumes and rank the recovery strategies given a chosen criterion. A second uncertainty study is performed by adding economic uncertainties to the initial set of reservoir uncertain parameters. Eventually a new decision tree is established and scenarios ranked using economic criteria.
The application of this methodology to an oil field from the Norwegian continental shelf and how recovery strategies are ranked are presented in this paper. The described methodology has exhibited the risks and uncertainties carried by the project, as it was possible to rank the different solutions based on the dispersion of the recoverable volumes distribution and/or on the net present value (NPV). In the context of a marginal or large capex project, a robust P90 case is required and this may therefore influence the choice of the recovery strategy. For instance, a scenario yielding the largest hydrocarbon volume may not be selected because it requires too many wells and/or too large investment if one of these criteria is defined as the most important. In addition, the combination of uncertainty studies enabled a full economic evaluation covering the entire recoverable volumes distribution whereas in many projects economic evaluation is focused on the P90, Mean and P10 scenarios.
The two-step integrated approach allows a decision to be made whilst taking into account both reservoir and economic aspects. Having a combined stochastic approach to the reservoir and economic uncertainties avoids a biased decision. All cases are stochastically covered and screened using a systematic and unified methodology that gives the same weight to each scenario.
Underground gas storage (UGS) into aquifers causes a limited dissolution of gas into the water at the gas-water interface. This phenomenon was characterized and quantified in a study carried out at a UGS site in an aquifer in the Parisian Basin (France).
The study methodology consisted of simulation of water and gas phase equilibrium and comparison of the results with in-situ measurements.
Reliable downhole gas/water holdup data were obtained by modifying a wireline tool based on optical refraction, discriminating gas from liquids and deriving a direct gas holdup. The modified optical tool was combined with an inline spinner and pressure and temperature sensors to allow an optimal evaluation of the zones of interest. Data were acquired in a monitoring well in flowing conditions (25 m3/day water). The survey was recorded from surface down to the producing water interval, revealing gas bubbles freeing from solution only at shallow depth, above 47 m.
Water was also sampled at reservoir conditions with a dedicated wireline tool. The samples show stable concentrations of dissolved gas over time, with methane as the prevalent dissolved gas.
Phase equilibrium was calculated at multiple depths using different thermodynamic equations of state and models. Results from PSRK and MHV2 models, including nucleation overpressure effects, fit well with the acquired data.
Thanks to the innovative logging tool and procedure, the consistency between the acquired and simulated data and the results of the equilibrium models, there is enhanced confidence in the thermodynamic modelling for UGS. The results from this analysis can be now integrated in a reservoir simulator to model the gas/water phase exchanges with a better accuracy.
This workflow can also be applied to other UGS fields for water and gas equilibrium modelling.
Underground gas storage (UGS) into aquifers causes a limited dissolution of gas into the water at the gas-water interface. A study was carried out for the characterization and the quantification of the phenomenon at a UGS site in an aquifer. In-situ measurements of the gas/water holdups were made in a monitoring well, with water samples collected at reservoir depth and along the well using a wireline down-hole measurement technique. The in-situ measurements allowed the validation of a theoretical thermodynamic model of dissolution.
Reliable forecasts of pressure, gas saturation and water production are of primary importance for the performance optimization of underground gas storages (UGS). This paper presents the development of a new methodology to achieve the history matching of radial reservoir models taking into account subsurface uncertainties.
Radial models are a simple but consistent way to modelize an aquifer underground gas reservoir, assumed single layered, with about few tens of cells in general to model explicitly the storage and the aquifer. Each reservoir cell holds porosity, permeability, thickness and depth properties. STORENGY has developed a specific radial reservoir simulator called PREPRE dedicated to pressure, gas column thickness and water predictions. Water coning options and Turner parameters (well water loading) have been accounted for semi-analtically, in order to make more reliable water predictions. A major advantage of such models is to run in a few seconds which fits with optimization process requirements when performing the history matching. In the standard workflow, we need to optimize hundreds of reservoir parameters in addition to the dynamics parameters like relative permeability / capillary pressure curves, residual saturation, etc.. This large number of parameters usually results in a very challenging and time-consuming optimization process.
We developed a methodology based on multi-parameters organized classification, based on the Kohonen method (Self Organizing Map algorithm). Models are built using maps of geological property classified in a matrix of about a hundred of classes. Using this approach, it is only needed to optimize the position of the model into the matrix with two parameters in addition to the dynamic parameters. This method then reduces the number of optimized parameters from a hundred to ten.
The objective function to be optimized contains two or three sub-objective functions like well pressure, gas column thickness and water production history-matching errors. We can combine all objectives into one in order to perform mono-objective optimization, but the best approach is to perform multi-objective optimization for uncertainty management purposes. The result is a population of models lying on the pareto front, thus providing information on the models uncertainty.
The topology of the objective functions is very complex, irregular and contains a lot of local minima. An efficient optimizer capable of reaching the global minimum with a good probability is needed. For that purpose, genetic-based CMA-ES optimizer was chosen. Although it converges slowly and needs a lot of function evaluations, usually it always finds a better minimum, compared with other standard optimizers. A new multi-objective CMA-ES version has also been developed, combining elitist and non-elitist methods to get the best compromise between speed and performance, in order to better describe the global pareto front.
These workflows are tested through several optimization processes conducted on various natural gas storage assets. The results illustrate the added value of such approach, particularly the quality of the history matching and the possibility to assess the uncertainties using several matched models for more reliable exploitation forecasts.
Rapid gas depressurization leads to gas cooling followed by slow gas warming when the cavern is kept idle. Gas temperature drop depends upon withdrawal rate and cavern size. Thermal tensile stresses, resulting from gas cooling, may generate fractures at the wall and roof of a salt cavern. These fractures are perpendicular to the cavern wall; in most cases their depth of penetration is small. The distance between two parallel fractures becomes larger when fractures penetrate deeper in the rock mass, as some fractures stop growing. These conclusions can be supported by numerical computations based on fracture mechanics. Salt slabs are created. These slabs remain strongly bounded to the rock mass and it is believed that in many cases their weight is not large enough to allow them to break off the cavern wall. However, depth of penetration of the fractures must be computed to prove that they cannot be a concern from the point of view of cavern tightness.
Gas storage caverns used to be developed mainly for seasonal storage, with one cycle per year and a moderate pressure rate between the maximum and minimum operation pressure (1MPa/day often was a maximum depressurization rate). However, the needs of energy traders are prompting change toward more aggressive operating modes. Typically, high-deliverability caverns (HFCGSC) can be emptied in 10 days and refilled in 30 days or less. At the same time, Compressed Air Energy Storage (CAES) is experiencing a rise in interest. They are designed to deliver full-power capacity in a very short time period.
Both types of facilities imply high gas-production rates and multiple yearly pressure cycles. This cycled mode of operation raises questions regarding frequently repeated, extreme, mechanical and thermal loading.