Development of Reservoir Descriptions To Aid in Design of EOR Projects

Abstract

Summary A technique was developed to obtain reservoir descriptions to aid in the design of EOR projects. This technique incorporates the reservoir performance during primary and secondary processes and the geological, petrophysical, and pressure-transient data. The application of this technique is illustrated for a San Andres formation reservoir. Introduction Designing an EOR project usually requires a characterization or description of the reservoir that can be used in a mathematical model to predict EOR project performance. Use of a reservoir description in a mathematical model offers a means to assess the possible significant economic risk involved in the operation of an EOR project. The more realistic the reservoir description is, the more reliable the EOR performance predictions will be. If an EOR process is field tested before a field-scale EOR project is run, a realistic reservoir description is critical to understanding the EOR process field test and can help in validating a reservoir simulator. In addition, a realistic reservoir description allows the study of how various EOR process variables affect performance. Thus, the optimal design of an EOR process can be determined. Early efforts in the petroleum industry to develop reservoir descriptions relied on statistical layering techniques. There are two drawbacks to the statistical procedures used previously. The biggest drawback is that the geological properties of the reservoir are ignored. The statistical methods fail to address reservoir continuity, correct PV in the high layers (important for predicting cycling of a miscible gas through the reservoir), vertical ordering and position of the layers (critical when crossflow occurs between layers), and varying irreducible water saturation with depth. An additional problem with the statistical approaches is that the layering is based on variations in permeability. If the standard massbalance equations for flow through porous media are divided by porosity, a term involving arises. This term, along with phase mobility and pressure gradient, dictates the speed at which fluid moves through a reservoir layer (Appendix A shows a derivation of this concept). Several geological studies have characterized oil reservoirs. These studies have been successful in identifying important geological features: faults, changes in rock type, facies changes, continuity or discontinuity of layers, expected permeability trends for various depositional environments, and barriers to vertical flow. A limitation with these studies is that the contrast in with depth was not considered as a tool to separate the reservoir into layers. Dividing a reservoir into layers, if it has intervals with different values of, is critical in making realistic predictions of reservoir performance. Even if the contrast in is used with the geological and petrophysical (core and log) data to estimate the reservoir stratification, our experience is that the resulting reservoir description has to be further modified to be used in a simulator to predict reservoir performance satisfactorily. One of the reasons is that log data are used to estimate layer permeabilities at noncored wells. The permeability must be estimated from correlations of core permeability with porosity. These correlations may yield significant errors in the estimated permeabilities, Additional errors may occur in estimating layer permeability in thin layers because logs have poor resolution of thin layers and tend to underestimate porosity. This leads to underestimating permeability. An additional reason why the geologically/petrophysically based reservoir description alone may fail to yield a satisfactory match of predicted performance is that reservoir properties are known only at wells. Areal reservoir heterogeneities can lead to varying interwell reservoir properties. One technique used to force a geologically based reservoir description to be satisfactory for predicting waterflood performances is to modify the water/oil relative permeability data. This procedure assumes that the reservoir description, based on geology alone, is correct. This is not a good assumption because of the arguments presented in the previous paragraph. Modification of the water/oil relative permeability can cause subsequent problems in predicting EOR performance. As an example, a multiple-contact-miscible gasflood can have two and sometimes three hydrocarbon phases, depending on the temperature and pressure. In the presence of mobile water, a three-phase relative permeability correlation may be required to calculate the intermediate wetting-phase relative permeability. Modifying the water/oil relative permeability then affects the relative permeabilities used in the miscible gas process. Errors in the geologically/petrophysically based reservoir description then relate to errors in the predicted EOR performance and the ability to design an EOR project properly. An additional problem associated with modifying the water/oil relative permeability is what to do with the imbibition (decreasing water saturation in an oil-wet reservoir) water relative permeability for an oil-wet reservoir. Changing the drainage (increasing water saturation in an oil-wet reservoir) water relative permeability without changing the imbibition water relative permeability affects the trapped water saturation. The trapped water saturation affects injection and production rates of a tertiary EOR process in an oil-wet reservoir. This paper presents a technique to develop reservoir descriptions suitable for design of EOR projects that has grown out of the work of Ader and Stein. The significant improvement to the work of Ader and Stein is the use of plots to develop a reservoir layering from the geological and petrophysical data. Laboratory-measured water/oil relative permeability data are not changed. Instead, layer permeability is changed to match primary and secondary performance. The application of this technique is illustrated for a San Andres formation oil reservoir. We have applied this technique to several reservoirs and have been very successful. Procedure To Develop a Reservoir Description Step 1-History-Match Time Period. The initial conditions at the start of a reservoir simulation affect the predicted results. Therefore, it is important to know the initial reservoir conditions at the start of the history match. The safest way to obtain a realistic estimate of the initial reservoir conditions is to start the history match at a time when the saturations and pressures are uniform within layers. In absence of either water influx or crossflow between layers, the starting time can be the start of pressure maintenance. Before pressure maintenance, the water saturation is at irreducible saturation and probably fairly uniform within layers. The layer ir-reducible water saturations can be determined from native-state coror electrical log data. The gas saturation and reservoir pressure should be fairly uniform within layers except in the vicinity of wells. This is why the material-balance equation can be used to characterize reservoir performance during volumetric depletion. Initial layer gas saturations can be related to layer pressures and average reservoir pressure (to be discussed later). Water influx can be viewed as a form of pressure maintenance. Then, the history-match time period should start when the water influx begins. If there is cross-flow between layers, a secondary gas cap could form during volumetric depletion. SPERE P. 143^

Identification of Barriers and Productive Zones Using Reservoir Characterization

“…This figure illustrates zones classification based on GR reading indicating shale and sand stone zone with correlation to Winland R35 method, FZI and DRT. This figure indicates 9 unit zones in this well having zone (2,4,5,6,8) are sand stone and proven to be a productive zones with relation to low Gamma ray and relatively high range of DRT, FZI and winland R35 that has better quality of fluids flow in the pore spaces of the reservoir rock. While zone (1,3,7,9) are shale zones, in relation to relatively high gamma ray and know as barriers (seals).…”

“…This is computed using the developed exploratory interval flow unit sorted and plotted in descending flow unit speed. In the figure, unit (1,2,4,8) shows to be the speed zones that happen to have the highest flow unit speed (FUS) as shown in table 1. Unit (3,5,6,7,15,12,13,14,11,10,9) illustrates to be baffles, which happen to have high storage capacity and lower flow capacity.…”

“…This figure indicates 9 unit zones in this well having zone (1,3,5,7) are sand stone and proven to be a productive zone with relation to low Gamma ray and relatively high range of DRT, FZI and winland R35 that has better quality of fluids flow in the pore spaces of the reservoir rock. While zone (2,4,6,8,9) are shale zones, in relation to relatively high gamma ray and know as barriers (seals). (2,4,6,8,10) are sand stone and proven to be a productive zones with relation to low Gamma ray and relatively high range of DRT, FZI and winland R35 that has better quality of fluids flow in the pore spaces of the reservoir rock.…”

“…While zone (2,4,6,8,9) are shale zones, in relation to relatively high gamma ray and know as barriers (seals). (2,4,6,8,10) are sand stone and proven to be a productive zones with relation to low Gamma ray and relatively high range of DRT, FZI and winland R35 that has better quality of fluids flow in the pore spaces of the reservoir rock. While zone (1,3,5,7,9) are shale zones, in relation to relatively high gamma ray and know as barriers (seals).…”

Caracterización Petrofísica de arenas del Neógeno en un pozo gasífero de la Cuenca de Veracruz

“…La figura 8 es un grafico modificado de Lorenz, en el eje abscisas es la capacidad de almacenamiento (producto de porosidad y espesor, PHIh) y el eje de las ordenadas es la capacidad de flujo (producto de la permeabilidad y espesor, kh) (Chopra et al, 1987); la interpretación del grafico es la siguiente: líneas horizontales son barreras al flujo y las que tienden a ser verticales presentan una buena capacidad de flujo; cabe hacer mención que las barreras al flujo son consideradas como unidades de flujo.…”