Hadibeik, Hamid (University of Texas at Austin) | Chen, Dingding (Halliburton Energy Services Group) | Proett, Mark A. (Halliburton Energy Services Group) | Eyuboglu, Abbas Sami Sami (Halliburton) | Torres-Verdin, Carlos
Pressure testing in very low-mobility reservoirs is challenging with conventional formation-testing methods. The primary difficulty is the over-extended build-up times required to overcome wellbore and formation storage effects. Possible wellbore overbalance or supercharge are additional complicating factors in determining reservoir pressure. This paper addresses the above technical complications and estimates petrophysical properties of low-mobility formations using a newly developed adaptive-testing approach.
The adaptive-testing approach employs an automated pulse-testing method for very low-mobility reservoirs and uses short drawdowns and injections followed by short pressure stabilization periods. Measured pressure transients are used in an optimized feedback loop to automatically adjust subsequent drawdown and injection pulses to reach a stabilized pressure as quickly as possible.
The automated pulse data is used to determine supercharge effects, formation pressure, and mobility via analytical models by analyzing the entire pressure sequence. A genetic algorithm estimates additional reservoir parameters, such as porosity and viscosity, and confirms results obtained with analytical models (reservoir pressure and permeability). The modeled formation pressure exhibits less than 1% difference with respect to true formation pressure, while the accuracy of other parameters depends on the number of unknown properties. As a quicker method to estimate reservoir properties, a direct neural-network regression of pulse-testing data was also investigated.
Synthetic reservoir models for low-mobility formations (M < 1 mD/cp), which included the dynamics of water- and oil-based mud-filtrate invasion that produce wellbore supercharging were developed. These reservoir models simulated the pulse-testing methods, including an automated feedback-optimization algorithm that reduces the testing times in a wide range of downhole conditions. The reservoir models included both simulations of underbalanced and overbalanced drilling conditions and enabled the development of new field-testing strategies based on a priori reservoir knowledge. The synthetic modeling demonstrates the viability of the new pulse-testing method and confirms that difficult properties, such as supercharging, can be estimated more accurately when coupled with the new inversion techniques.
We introduce a new numerical algorithm to forecast gas production in organic shale that simultaneously takes into account gas diffusion in kerogen, slip flow, Knudsen diffusion, and Langmuir desorption. The algorithm incorporates the effects of slip flow and Knudsen diffusion in apparent permeability, and includes Langmuir desorption as a gas source at kerogen surfaces. We use the diffusion equation to model both lateral gas flow in kerogen as well as gas supply from kerogen to surfaces.
Slip flow and Knudsen diffusion account for higher-than-expected permeability in shale-gas formations, while Langmuir desorption maintains pore pressure. Simulations confirm the significance of gas diffusion in kerogen on both gas flow and stored gas. Relative contributions of these flow mechanisms to production are quantified for various cases to rank their importance under practical situations.
Results indicate that apparent permeability increases while reservoir pressure decreases. Gas desorption supplies additional gas to pores, thereby maintaining reservoir pressure. However, the rate of gas desorption decreases with time. Gas diffusion enhances production in two ways: it provides gas molecules to kerogen-pore surfaces, hence it maintains the gas desorption rate while kerogen becomes a flow path for gas molecules. For a shale-gas formation with porosity of 5%, apparent permeability of 59.7 µD, total organic carbon of 29%, effective kerogen porosity of 10%, and gas diffusion coefficient of 10-22 m2/s, production enhancements compared to those predicted with conventional models are: 9.6% due to slip flow and Knudsen diffusion, an extra 42.6% due to Langmuir desorption, and an additional 61.7% due to gas diffusion after 1 year of production. The method introduced in this paper for modeling gas flow indicates that the behavior of gas production with time in shale-gas formations could differ significantly from production forecasts performed with conventional models.
This paper introduces a rock typing method for application in hydrocarbon-bearing shale (specifically source rock) reservoirs using conventional well logs and core data. Source rock reservoirs are known to be highly heterogeneous and often require new or specialized petrophysical techniques for accurate reservoir evaluation. In the past, petrophysical description of source rock reservoirs with well logs has been focused to quantifying rock composition and organic-matter concentration. These solutions often require many assumptions and ad-hoc correlations where the interpretation becomes a core matching exercise. Scale effects on measurements are typically neglected in core matching. Rock typing in hydrocarbon-bearing shale provides an alternative description by segmenting the reservoir into petrophysically-similar groups with k-means cluster analysis, which can then be used for ranking and detailed analysis of depth zones favorable for production.
A synthetic example illustrates the rock typing method for an idealized sequence of beds penetrated by a vertical well. Results and analysis from the synthetic example show that rock types from inverted log properties correctly identify the most organic-rich sections better than rock types detected from well logs in thin beds. Also, estimated kerogen concentration is shown to be the most reliable property in an under-determined inversion solution.
Field cases in the Barnett and Haynesville shale gas plays show the importance of core data for supplementing well logs and identifying correlations for desirable reservoir properties (kerogen/TOC concentration, fluid saturations, and porosity). Qualitative rock classes are formed and verified using inverted estimates of kerogen concentration as a rock-quality metric. Inverted log properties identify 40% more of a high-kerogen rock type over well-log based rock types in the Barnett formation. A case in the Haynesville formation suggests the possibility of identifying depositional environments as a result of rock attributes that produce distinct groupings from k-means cluster analysis with well logs. Core data and inversion results indicate homogeneity in the Haynesville formation case. However, the distributions of rock types show a 50% occurrence between two rock types over 90 ft vertical-extent of reservoir. Rock types suggest vertical distributions that exhibit similar rock attributes with characteristic properties (porosity, organic concentration and maturity, and gas saturation).
The interpretation method considered in this paper does not directly quantify reservoir parameters and would not serve the purpose of quantifying gas-in-place. Rock typing in hydrocarbon-bearing shale with conventional well logs forms qualitative rock classes which can be used to calculate net-to-gross, validate conventional interpretation methods, perform well-to-well correlations, and establish facies distributions for integrated reservoir modeling in hydrocarbon-bearing shale.
Petrophysical interpretation of well logs acquired in organic shales and carbonates is challenging because of the presence of thin beds and spatially complex lithology; conventional interpretation techniques often fail in such cases. Recently introduced methods for thin-bed interpretation enable corrections for shoulder-bed effects on well logs but remain sensitive to incorrectly picked bed boundaries.
We introduce a new inversion-based method to detect bed boundaries and to estimate petrophysical and compositional properties of multi-layer formations from conventional well logs in the presence of thin beds, complex lithology/fluids, and kerogen. Bed boundaries and bed properties are updated in two serial inversion loops. Numerical simulation of well logs within both inversion loops explicitly takes into account differences in the volume of investigation of all well logs involved in the estimation, thereby enabling corrections for shoulder-bed effects.
The successful application of the new interpretation method is documented with synthetic cases and field data acquired in thinly bedded carbonates and in the Haynesville shale-gas formation. Estimates of petrophysical/compositional properties obtained with the new interpretation method are compared to those obtained with (a) nonlinear inversion of well logs with inaccurate bed boundaries, (b) depth-by-depth inversion of well logs, and (c) core/X-Ray Diffraction (XRD) measurements. Results indicate that the new method improves the estimation of porosity of thin beds by more than 200% in the carbonate field example and by more than 40% in the shale-gas example, compared to depth-by-depth interpretation results obtained with commercial software. This improvement in the assessment of petrophysical/compositional properties reduces uncertainty in hydrocarbon reserves and aids in the selection of hydraulic fracture locations in organic shale.
ABSTRACT: Hydraulic rock typing is based on pore geometry, which relates to saturation-height modeling at a later stage in reservoir characterization. Additionally, pore geometry affects mud-filtrate invasion under over-balanced drilling conditions. Reliable hydraulic rock typing should simultaneously honor saturation behavior in the vertical direction and mud-filtrate invasion in the radial direction. Such a condition becomes critical when hydraulic rock typing is performed with well logs acquired in multiple wells penetrating the same or different capillary transition zones. This paper considers three conventional core-based rock typing methods, namely Leverett’s k /f , Winland R35, and Amaefule’s flow zone index, to appraise whether rock classifications can be extrapolated from core-data to well-log domains. A new quantitative log attribute is derived from well logs to assist hydraulic rock typing, which integrates in-situ reservoir capillary pressure (Pc) and initial water saturation (Swi). The assumption is that the reservoir under study underwent hydrocarbon migration wherein vertical fluid distribution is still well represented by the primary-drainage capillary pressure curve. Petrophysical properties that are closely linked to pore geometry are quantified by invoking both Leverett’s J-function and Thomeer’s G-factor. The new log attribute is based on standard well-log analysis and only requires conventional well logs for its application. Thus, it can be generally applied to both clastic and carbonate reservoirs in multi-well contexts. It overcomes the limitation of the bulk volume water method which is only applicable to reservoir zones that are at nearly irreducible water saturation. Most importantly, it provides good initial estimates to constrain in-situ dynamic rock-fluid properties such as capillary pressure and relative permeability. The method proceeds with initial estimates of dynamic properties to construct multi-layer petrophysical models with a common stratigraphic framework (CSF) for each rock type, and to simulate the process of mud-filtrate invasion.
Hadibeik, Nishaboori Abdolhamid (U. of Texas at Austin) | Proett, Mark A. (Halliburton Energy Services Group) | Chen, Dingding (Halliburton) | Eyuboglu, Abbas Sami Sami (Halliburton) | Torres-Verdin, Carlos (U. of Texas at Austin) | Pour, Rohollah Abdellah
Tight formation testing when mobilities are lower than 0.01 mD/cP poses significant challenges because the conventional pressure transient buildup testing becomes impractical as a result of the large buildup stabilization time. This paper introduces a new automated pulse test method for testing in tight formations that significantly reduces testing time and makes the determination of formation pressure and permeability possible. A pulse test is defined as a drawdown followed by an injection test, and the source is shut in to record the pressure transient. Based on pressure data during the shut-in period, the next drawdown or injection test is designed, such that the flow rate is a fraction of the initial pulse rate, followed by another shut-in test. This procedure continues until the difference in pressure at the beginning and at the end of the shut-in period is reduced to within a specified limit of pressure change; then, an extended transient is recorded to a stabilized shut-in pressure. The overall advantage is to reduce the pressure stabilization time by implementing an adaptive pressure feedback loop in the system. The method can be applied to a straddle packer test using conventional drillstem testing tools or formation testers, using either straddle packers or probes.
The effects of wellbore storage and fluid compressibility are found to reduce the pressure drop and positive pressure pulse in the drawdown and injection tests, respectively; they also affect the decay rate to the asymptote of the shut-in pressure response. Consequently, the combined pulse test method with the pressure feedback system and wellbore storage effect reduces the reservoir pressure testing time in tight formations.
The automated pulse-test method has been successfully validated with consideration of the effects of wellbore storage and overbalance pressure in tight gas and heavy oil formations. In addition, the effects of invasion with water- and oil-based mud filtrate were considered in the modeling. The method uses successive pressure feedbacks and automated pulses to yield a pressure to within 0.5% range of the initial reservoir pressure while decreasing the wait time by a factor of 10 for a packer type formation tester.
To account for various tool options and storage effects, the packer-type, oval probe, and standard probe-type formation testers have been simulated in various tight formation conditions. The method enables a rapid appraisal of pressure measurements in comparison to conventional testing. Simulations also indicate that the analytical spherical model can be used to analyze a pulse test, even when encountering multi-phase compositional fluid effects.