PSI - Issue 18

A. Kostina et al. / Procedia Structural Integrity 18 (2019) 301–308 Author name / Structural Integrity Procedia 00 (2019) 000–000

306

6

3. Results of numerical simulation To demonstrate the efficiency of the proposed model (1)-(17) we will simulate structural changes arising in the rock mass during oil recovery by the steam-assisted gravity drainage method. The simulated domain represents the part of the reservoir with the length of 44 meters and the height of 24 meters. The thickness of the domain is equal to 1. The physical and mechanical properties of the formation correspond to Yarega oil deposit (Russian crude oil deposit in Komi Republic) (Kostina et al. (2018)). According to this technique, two horizontal wells are drilled one above the other. The distance between the injection and the production well was equal to 5 meters. The temperature and the pressure of the injected steam were equal to 496 K and 3.6 MPa respectively. The formation in the virgin state is under compressive loading. Typical viscous oil reservoir is located at the depth of 200 meters where confining stresses are not so high. Propagation of the thermal front under the high pressure leads to the increase in stresses ahead of the front. Initially, the structural changes of the reservoir are characterized by the compaction of pores and microcracks. Large thermal strains induce rise in volumetric strains, which converts compressive state to the volume expansion. This process leads to the surface heave. However, as it has been mentioned by Shafiei and Dusseault (2013), the single mechanism of thermal expansion could not provide observed values of the vertical heave. The soils (as well as other granular materials) have a tendency to increase volume when they are subjected to the shear strain. Structural damage defined by equations (14)-(15) simulates additional contribution to strain induced by the growth of the volumetric-type defects (microcracks) due to the increase in the shear-type defects (microshears). Parameters 0 c σ and dc σ can be considered as the initiation criterion for this mechanism. Figure 1 shows values of vertical heave in case of the absence (Fig. 1(a)) and the presence of structural strains (Fig. 1(b)). As it can be seen, the estimated surface heave without defect-induced strain is one order of magnitude lower in comparison to the case when it is taking into account.

(a)

(b)

Fig. 1. Vertical heave (a) thermo-elastic model; (b) thermo-inelastic model.

Increase in volumetric strains affects the porosity of the reservoir. Figure 2 presents distribution of the porosity evolution after 300 days of the heating obtained with (Fig. 2(b)) and without (Fig. 2(a)) structural strains. It can be seen that the difference between the results is not only quantitative but also qualitative. The distribution of the volumetric damage is strongly affects the shape of the porosity distribution. Moreover, consideration of the structural changes allows one to obtain a rise of 4-6% while thermo-elastic strains gives significantly less values. Similar results can be observed for the permeability (Fig. 3). A rise in the porosity considerably improves permeability and provides higher values of the oil production rate. Figure 4 demonstrates specific oil production rate after the establishment of the interwell communication. On the initial stage when the influence of the structural strain is small the oil production rates are slightly different from each