Katika visima vya kuchimba, bomba la chuma la casing, ganda la saruji linalolizunguka na tabaka la mwamba huunda mfumo wa tabaka nyingi unaobeba mizigo na kutoa uzuiaji wa majimaji. Bomba la casing likitoka katikati ya kisima au mshikamano kati ya saruji na mwamba ukidhoofika, mikazo inaweza kujilimbikiza katika maeneo fulani na kusababisha nyufa kuanza mapema, kupinduka kuelekea mwelekeo tofauti na kutengeneza njia za uharibifu zinazorahisisha kupita kwa vimiminika.

Kwa ajili ya kuchunguza mchakato huu, watafiti walitengeneza modeli mseto ya fracture ya phase-field inayojumuisha pia interfaces dhaifu za nyenzo. Kwanza modeli ilithibitishwa kwa suluhisho za analitiki na matatizo ya kulinganisha yaliyochapishwa awali. Kisha eccentricity ya casing, nguvu ya interface ya cement-rock na three-dimensional crack interactions zilichunguzwa kwa majaribio tofauti ya nambari.

Kulingana na matokeo, casing ilipokuwa katikati, crack-initiation pressure ilikuwa 71 MPa, lakini kwa eccentricity ya yüzde 83,33 thamani hii ilishuka hadi 49,5 MPa. Hii ni sawa na karibu yüzde 30 loss of strength. Eccentricity ilipofikia yüzde 50, non-radial inclined shear cracks pia zilionekana ndani ya rock formation. Ikiwa cement-rock interface ilidhoofishwa hadi karibu yüzde 30 ya strength ya materials zinazozunguka, cracks zilibadilisha mwelekeo na kuendelea along interface badala ya kuingia kwenye rock, na radial cracks zaidi ziliunda ndani ya cement.

Kwa nini well integrity ni muhimu?

Visima vilivyowekwa saruji havitumiki tu katika uzalishaji wa mafuta na gesi asilia. Geothermal energy, underground hydrogen na natural-gas storage, carbon-dioxide injection, underground waste storage, groundwater production na matumizi mbalimbali ya madini pia hutegemea well architecture inayofanana. Katika mifumo hii steel casing hutoa mechanical support, huku cement sheath ikijaza nafasi kati ya pipe na rock na kupunguza fluid migration kati ya geological layers tofauti.

Hata hivyo casing isipojiweka kikamilifu katikati ya well, thickness ya cement hubadilika katika circumferential direction. Thin cement region inaweza kubeba stresses kubwa zaidi. Cement shrinkage, drilling-mud residues, voids, temperature changes na inadequate cementing pia vinaweza kudhoofisha cement-rock au steel-cement interfaces.

Aina kuu za damage zinazoweza kutokea ni radial cracks, shear cracks, steel-cement debonding, cement-rock debonding na disk-shaped fractures perpendicular to well axis. Defects hizi zikijumuika zinaweza kutengeneza fluid channels zinazojulikana kama microannulus na kuvuruga hydraulic isolation ya well.

Phase-field method inaonyeshaje crack?

Katika traditional finite-element models, crack path inaweza kuhitaji kufafanuliwa mapema. Katika phase-field approach, crack haiwakilishwi kama sharp geometric surface bali kama diffuse region ambako material hubadilika kutoka intact state kwenda damaged state.

Katika model, d phase-field variable inaonyesha damage state ya material. Kwa takribani d = 0 material ni intact, na d = 1 inawakilisha fully cracked region. Material stiffness hupunguzwa kwa function hii:

\[ g(d)=(1-d)^2 \]

Kwa njia hii crack initiation, branching, coalescence na mabadiliko ya mwelekeo kati ya materials yanaweza kutokea kama matokeo ya energy minimization bila predefined crack line.

Total potential energy inajumuisha kwa pamoja elastic-strain energy ya materials mbili, regularized crack-surface energy ndani ya volume na interface-specific fracture energy:

\[ U = \sum_{m=1}^{2}\int_{\Omega^m}g(d^m)\psi^m(\varepsilon(u^m))\,d\Omega + \sum_{m=1}^{2}\int_{\Omega^m}G_c^m \frac{1}{2} \left[ \frac{(d^m)^2}{\ell} + \ell|\nabla d^m|^2 \right]d\Omega + \int_{\Gamma_{\mathrm{int}}}G_c^{\mathrm{ib}}\tilde{d}^{\,2}\,d\Gamma \]

Hapa u ni displacement field, ε small-strain tensor, ψ elastic-energy density, G_c critical energy-release rate, ℓ phase-field length scale na Γ_int material interface.

Interface strength ilibadilishwaje?

Nguvu ya bond kati ya cement na rock ilifafanuliwa kwa parameter κ. Effective interface fracture energy iliandikwa:

\[ G_c^{\mathrm{eff}}=G_c^{\mathrm{nb}}(1+\kappa) \]

G_c^nb ni fracture energy ya neutrally bonded interface. κ = 0 inawakilisha neutral bond, κ > 0 strengthened bond, na κ < 0 weakened bond. Kwa mfano, κ = −0,7 inapofikiwa effective fracture energy ya interface hushuka hadi yüzde 30 ya neutral case.

Model inapunguza stiffness katika tension-dominated regions na kutumia tension-compression energy split inayozuia nonphysical crack growth under compression. Pia history field inayohifadhi highest tensile energy reached in past imejumuishwa ili kuzuia damage closure au healing.

Casing eccentricity

Umbali wa casing kutoka center ulifafanuliwa kwa dimensionless eccentricity parameter:

\[ e=\frac{r_e}{R_2-R_1} \]

Hapa r_e ni distance between casing center and well center, na R₁ pamoja na R₂ ni relevant inner na outer radii. e = 0 inawakilisha concentric system, wakati high e values zinawakilisha geometries ambako cement sheath ni noticeably thinner upande mmoja.

Model ilithibitishwa katika hatua tatu

1. Analytical stress solution katika eccentrically perforated pipe

Katika validation ya kwanza, cement pipe yenye eccentric hole ilimodeliwa ikiwa inner surface ina pressure ya 30 MPa na outer surface 50 MPa, kisha tangential stresses zikalinganishwa na Jeffery closed-form analytical solution. Inner radius ilikuwa 69,5 mm, outer radius 94,5 mm, center offset 10 mm na eccentricity yüzde 40.

Solution iliyotumia takribani 7,6 million quadrilateral finite elements ilireproduce kwa ujumla analytical stress distribution. Maximum relative error ilikuwa yüzde 11,37 katika 90-degree location ambako kulikuwa na high stress gradient. Researchers walihusisha difference hii na stress resolution ya first-order quadrilateral elements.

2. Two-material plate yenye weak circular interface

Katika validation ya pili, two-material plate yenye ukubwa 540 × 540 mm, initial crack ya 50 mm na weak circular interface ilichunguzwa. Interface fracture energy iliwekwa 50 N/m na weakening parameter −0,875.

Crack kwanza ilisonga ndani ya material ya kwanza kwa opening mode, ilipofikia circular weak interface ikadeflect along interface na baadaye ikaingia tena katika material ya pili. Damage path iliyopatikana ilionyesha strong qualitative agreement na published reference simulation. Small difference katika points ambako crack iliondoka interface ilihusishwa na ukweli kwamba reference model iliwakilisha interface kama finite-thickness third material, wakati study hii iliimodel kama zero-thickness sharp surface.

3. Well geometry yenye steel, cement na rock

Katika validation ya tatu, realistic well section yenye steel casing, cement sheath na Castlegate sandstone formation ilimodeliwa. Kwa sababu ya symmetry, quarter ya geometry ilitumika na kutatuliwa kwa takribani 700 thousand quadrilateral elements.

Model ilireproduce radial cracks ndani ya cement, crack transfer into rock formation, microcracks at steel-cement interface na debonding kwa agreement na published reference study. Kwa sababu steel ilikuwa na very high fracture energy, crack haikuingia steel body na badala yake ikadeflect along steel-cement boundary.

Well strength ilipungua kadiri eccentricity ilivyoongezeka

Researchers walitumia seven different eccentricity values kati ya yüzde 0 na yüzde 83,33. Internal well pressure wakati wa crack initiation ilibadilika hivi:

Eccentricity ilipoongezeka kutoka yüzde 0 hadi yüzde 83,33, crack-initiation pressure ilipungua kwa 21,5 MPa. Katika concentric geometry tangential stress ilisambaa kwa kiasi kikubwa uniformly around cement, lakini katika eccentric geometries ilijilimbikiza katika thinnest cement region. First crack pia kila mara ilianza katika thin region hiyo.

Pressure ilipofikishwa 97 MPa, concentric well iliendelea kuwa na symmetric crack pattern, wakati damage ilizidi kuwa asymmetric kadiri eccentricity ilivyoongezeka. Baada ya first crack, cracks mpya ziliunda katika thicker cement regions kwa counterclockwise sequence.

Yüzde 50 eccentricity ilileta fracture mode mpya

Eccentricity ilipofikia yüzde 50, inclined cracks zilizopotoka kutoka radial direction zilionekana ndani ya rock. Cracks hizi zilianza karibu na 10-degree azimuth region na zilihusishwa na high local shear stresses. Fracture mode hii, ambayo haikuonekana katika concentric system, ilionyesha kwamba geometric imperfection inaweza kubadilisha si fracture pressure pekee bali pia crack type na direction.

Weak interface ilibadilisha crack path

Kwa cement-rock interface, κ ilibadilishwa kati ya −0,5 na −0,9. Clear effect ya interface weakening ilionekana kwa κ ≤ −0,7. Condition hii inamaanisha interface ina effective fracture energy ya yüzde 30 au chini ya neutral-bond case.

Katika stronger interfaces, radial crack iliyoanza ndani ya cement iliingia moja kwa moja kwenye rock formation. Katika weak interfaces, crack ilideflect along cement-rock boundary badala ya kuingia rock. Kutekwa kwa crack kwenye interface kulizuia immediate stress release ndani ya cement sheath na kuunda conditions za new radial cracks.

Kwa κ = −0,8, ambapo interface ilikuwa weakened kwa yüzde 80, load-displacement curves zilionyesha multiple successive fracture jumps. Katika neutral bond, radial cracks mbili zilianza karibu simultaneously na curve ikaonyesha single dominant fracture point.

Weakest interface haikuwa kila mara case ya earliest rock cracking. Kwa κ = −0,8 crack iliingia rock formation katika 123,5 MPa, wakati kwa κ = −0,9 transition ilichelewa hadi 144 MPa. Sababu ilikuwa kwamba interface iliyodhoofishwa kwa yüzde 90 ilidebond mapema na kwa kiwango kikubwa zaidi, ikarelieve stress locally ndani ya cement na mechanically partially decouple rock system.

Result hii inaonyesha kwamba interface weakness haibadilishi well behavior kwa one-directional linear way. Weaker bond hurahisisha interface debonding na katika conditions fulani inaweza kuchelewesha crack transfer into rock; lakini pia inaweza kuunda damage zaidi ndani ya cement na potential leakage paths.

three-dimensional model ilionyesha michakato ambayo plane-strain assumption haiwezi kuona

Katika final stage, three-dimensional well model yenye depth ya 100 mm ilitatuliwa. Cement-rock interface ilidhoofishwa kwa yüzde 90. Initial mesh ilikuwa na 724 thousand trilinear eight-node elements. Kadiri damage ilivyoendelea, adaptive mesh refinement iliwashwa.

Stress shadows zilizotengenezwa around initial cracks zilizuia later cracks kuanza simultaneously across full well depth. Kwa hiyo cracks zilinucleate katika depths tofauti, zikadeviate kutoka axis na kucoalesce along interface. Behaviors hizi haziwezi kuwakilishwa na two-dimensional plane-strain models.

Calculation ilikuwa na karibu 400 thousand degrees of freedom mwanzoni na ikafikia 113 million degrees of freedom baada ya adaptive mesh refinement. Simulation ilihitaji karibu 20 thousand core-hours kwenye cluster yenye 2,45 GHz AMD 64-core processors. Kulingana na researchers, same solution kwa uniformly fine mesh ingeomba karibu 12 billion degrees of freedom.

Hii ina maana gani kwa engineering?

Study inaonyesha kwamba quality ya well centralization haipaswi kutathminiwa kwa cement consumption au installation tolerance pekee. Casing offset inaweza kupunguza pressure inayohitajika kuanzisha crack na, baada ya threshold fulani, kuunda new shear-fracture modes ndani ya rock.

Cement-rock bond strength pia haipaswi kutathminiwa tu kama “strong” au “weak”. Interface strength inaweza kuamua kama crack itaingia rock, itaendelea along interface, cracks ngapi mpya zitaibuka ndani ya cement na damage itaungana vipi through depth.

Model inatoa computational framework kwa cementing design, casing-centralizer placement, interpretation of cement-bond logs na evaluation ya well pressure limits. Hata hivyo, kabla ya kutumia results moja kwa moja kama field safety factor, zinapaswa kurecalibrate kwa actual well geometries, material properties, temperature history, pore pressure na fluid effects.

Mbinu na Matokeo ya Utafiti

Muundo wa utafiti

Study inatumia validated computational experiments badala ya physical experiment. Hybrid phase-field fracture model inatumika kwa multi-material finite-element systems zenye weak interfaces. Solution inafanywa kwa staggered iterative algorithm ambapo displacement na damage fields zinatatuliwa kwa zamu.

• Plane-strain assumption ilitumika katika two-dimensional analyses.
• Bilinear quadrilateral elements zilitumika katika two-dimensional meshes na trilinear hexahedral elements katika three-dimensional mesh.
• Convergence tolerance kwa L2 norm difference ya nodal damage vectors kati ya two successive iterations iliwekwa 10⁻².
• Katika visualizations, elements zenye damage variable kubwa kuliko 0,9 zilichukuliwa kuwa cracked region.
• Model ilitekelezwa katika deal.II open-source finite-element library.
• Parallel computing na adaptive mesh refinement zilitumika katika three-dimensional solution.

Materials zilizotumika katika main numerical experiments

Main findings

• Casing eccentricity ilipoongezeka kutoka yüzde 0 hadi yüzde 83,33, crack-initiation pressure ilishuka kutoka 71 MPa hadi 49,5 MPa.
• First crack ilianza consistently katika thinnest region ya cement sheath katika eccentric wells.
• Kwa eccentricity yüzde 50 na zaidi, inclined cracks ziliunda katika rock formation kwenye maeneo yenye high shear stress.
• Effective cement-rock interface strength iliposhuka hadi yüzde 30 ya neutral case au chini, deflection na debonding along interface vilitawala kuliko crack transfer.
• Weak interface ilikuza successive new radial cracks ndani ya cement.
• Most weakened interface haikuwa case ya earliest rock cracking; early extensive debonding ilirelieve stress locally na kuchelewesha transfer into rock.
• three-dimensional model ilionyesha crack initiation katika depths tofauti, stress shadowing, off-axis crack growth na coalescence along interface.

Mapungufu

• Study ni preprint na haijapitia peer review.
• Results si physical field experiments bali validated numerical-model results.
• Model inategemea linear isotropic elasticity na brittle phase-field assumptions.
• Katika interface only opening-dominated damage imemodeliwa; frictional slip haijajumuishwa.
• Fluid flow, poroelasticity, thermal effects na hydro-mechanical coupling hazijajumuishwa.
• Most two-dimensional results zinategemea plane-strain assumption.
• three-dimensional model ni demonstration problem yenye depth ya 100 mm na haiwakilishi direct field scale.
• Maximum relative error katika analytical stress validation ni yüzde 11,37.
• Calculated pressure thresholds zinategemea selected geometry, material properties na phase-field length scale; hazipaswi kutafsiriwa kama universal limits kwa wells zote.
• Potential leakage paths na sustained casing-pressure risk zinainferiwa kutoka mechanical crack geometry; direct fluid transport haikusimuliwa.

Dokezo la Chanzo na Mbinu

Utafiti wa asili: “Phase-field analysis of fracture in heterogeneous wellbore systems: effects of casing eccentricity and cement-formation interface strength.”

Waandishi: Tharunsarathy Sachithanantham, Wasim Niyaz Munshi, Chandrasekhar Annavarapu na Birendra Jha.

Corresponding author: Chandrasekhar Annavarapu. Hakuna equal-contribution statement.

Taasisi: Indian Institute of Technology Madras, Department of Civil Engineering; University of Southern California, Mork Family Department of Chemical Engineering and Materials Science.

Mwaka na aina ya chanzo: Original numerical engineering research na preprint ya 2026.

Hali ya peer review: Study bado haijapitia peer review. Results hazipaswi kutathminiwa kama final published journal article.

Platform: SSRN.

DOI:10.2139/ssrn.6947717

Kiungo rasmi cha study:SSRN study record

Scientific method, numerical values, equations, material properties, visual interpretations, findings na limitations katika content hii zinategemea only study iliyochunguzwa. External sources zilitumika tu kwa bibliographic verification ya author identity, institution, platform, publication date na DOI; hakuna scientific result ambayo haikuwepo katika study iliyoongezwa.