Gas Lift

summary gas lift traverses are computed incrementally from the wellhead node down to reservoir node there are multiple scenarios in which prodx can compute a gas lift traverse and gas lift valve diagnostics, depending on the available data no gas lift valve data present gas lift valve data present without r ratio (r) and test rack opening pressure (ptro) gas lift valve data present with r and ptro during the course of valve diagnostic calculations, a valve state is determined by the application the following states can be detected open / flowing injection pressure is > valve opening pressure and the traverse / valve performance calculations indicate injection gas is passing through the valve open / not flowing injection pressure is > valve opening pressure and the traverse / valve performance calculations indicate injection gas is not passing through the valve (i e all injection gas is passing through a shallower valve) open / can't flow injection pressure is > valve opening pressure, but the production pressure is > injection pressure in this condition gas will not pass through the valve and in packerless installations, the valve may be submerged closed injection pressure is < closing pressure indicating the valve is closed likely closed injection pressure is between opening and closing pressure this is an indeterminant state if the injection pressure was increasing from below the closing pressure, the valve would still be closed if injection pressure was decreasing from above the opening pressure, the valve would still be open for the purposes of daily gas lift modeling, these valves are assumed to be closed dummy this state is used for dummy valves which have no port for gas to pass through dummy valves do not participate in modeling, but are shown for completeness of analysis, allowing engineers to identify the potential for a valve re installation no gas lift valve data present in the absence of known gas lift valve depths, prodx will compute a pure injection traverse and production traverse (using the total gas rate = produced + injected gas) then the spread between these traverses is analyzed to find the deepest point at which the difference is greater than a minimum delta pressure (45 psi, by default) this is assumed to be the gas lift depth of injection the production and injection traverses are modified below this depth to remove the injected gas from the traverse rates gas lift valve data present without r and ptro with known valve depths and unknown r and ptro values, the design case surface opening and closing pressures (pso, psc) are used to determine which valve(s) are open to perform this calculation, the injection pressure at surface is compared to each valve's pso valves in which the injection pressure is above pso are deemed open, and valves in which the injection pressure is below psc are deemed closed valves which have injection pressures between psc and pso are indeterminant but assumed closed for modeling next, a production and injection traverse are computed simultaneously from surface to the first open valve using a total gas rate of produced + injected gas for the production traverse at the first open valve, the modified thornhill craver (tc) equation or api rp 11 v2 is used to compute the gas flow rate through the valve, depending on configuration and available data see below sections for more information any remaining injection gas that cannot pass through the first open valve will be assumed to travel to the next open valve or deepest/orifice valve the traverse is continued downward using this methodology, modifying the produced and injected gas rate accordingly by depth and passage through open valves any valves deeper than the intersection of the production and injection traverses are considered submerged gas lift valve data present with r and ptro with known valve depths, r, and ptro values, a more comprehensive calculation can occur first, a production and injection traverse are computed simultaneously from the surface node down to the first gas lift valve the production traverse assumes a total gas rate of produced gas + injected gas from the surface to the first valve at the first valve, downhole opening and closing pressures are computed by calculating the dome pressure (closing, pvc), and then using the r value and production pressure to determine opening pressure (pvo) see section below for more details on opening and closing pressure calculations then, using the injection pressure at depth, a comparison is made to pvo/pvc to determine if that valve is open, closed, or "likely closed" (indeterminant state with injection pressure between pvc and pvo) if the valve is open, the modified thornhill craver (tc) equation (default) or api rp 11 v2 is used to compute the gas flow rate through the valve, depending on configuration and available data see below sections for more information any remaining injection gas that cannot pass through the first open valve will be assumed to travel downward the traverses continue downward, with gas rates modified accordingly to the next valve, where the previous step is repeated to check valve status and flow rate through the valve, if applicable any valves deeper than the intersection of the production and injection traverses are considered submerged the following diagram illustrates the methodology computing pvo, pvc, pso, and psc the downhole closing pressure of a valve is equal to its dome pressure at depth first the estimated dome pressure at 60 degf is calculated using pd' = ptro (1 r) pd' dome pressure at 60 degf ptro test rack opening pressure r r ratio then, the final downhole pvc is calculated through an iterative method using an equation of state with an estimation of nitrogen compressibility from sage and lacy the dome / pvc pressure is used to compute the downhole opening pressure (pvo) using pvo = \frac{(pvc pprod r)}{(1 r)} pvo downhole opening pressure pvc downhole closing pressure pprod production presssure r r ratio the downhole pvo and pvc values are converted to surface pressures (pso and psc) through a traverse from downhole node to surface node modified thornhill craver to compute gas passage through an open valve, one method prodx can leverage is the modified thornhill craver equation qgi = \frac{155 5 cd ap p1 \sqrt{2g (\frac{k'}{k' 1}) (r^{\frac{2}{k'}} r^{\frac{k'+1}{k'}}}) }{\sqrt{γg z t}} qgi gas flow rate, mscf/d cd discharge coefficient (see below) ap valve seat flow area, in2 p1 injection pressure, psia g acceleration due to gravity, 32 16 ft/s2 k' specific heat ratio of injection gas, cp / cv r pressure ratio of production pressure / injection pressure γg specific gravity of injection gas z z factor of injection gas t absolute temperature of injection gas, degr discharge coefficient 1 5" injection pressure operated (ipo) valves 0 76 1 5" orifice valves 0 86 1" ipo valves or any production pressure operated valves 0 65 1" orifice valves 0 75 api rp 11 v2 if configured and detailed valve experimental data is available, prodx can analyze the appropriate flow model to use following api rp 11 v2 the first step is calculating the transition injection pressure between orifice to throttling flow using ptran = pd (1 + fe) ptran transition injection pressure, psi pd dome pressure at depth, pvc, psi fe dynamic ap/ab ratio from valve test data orifice model if the injection pressure is greater than ptran, then the orifice model is used the orifice model assumes the gas lift valve operates as a simple orifice, where the flow through the valve is governed by standard orifice flow equations this is used when the valve is fully open, and the flow rate is not restricted by mechanical components like a throttling stem key assumptions flow is dominated by the port area the gas flow through the valve follows subsonic or sonic flow conditions depending on the pressure ratio there are no moving parts restricting the flow orifice flow qg = cd a \sqrt{(2 δp)/ρg} where qg gas flow rate (standard cubic feet per minute, scf/m) cd discharge coefficient (dimensionless, typically 0 6–0 8 depending on valve design) a effective flow area of the orifice (in2) δp pressure drop across the valve (psi), δp = pcasing − ptubing ρg gas density (lb/ft³) choked flow if the pressure drop (δp\delta pδp) exceeds the critical pressure ratio, the gas reaches sonic velocity (choked flow), and the flow rate is no longer influenced by downstream pressure the critical pressure ratio (ptubing/pcasing) for choked flow depends on the specific heat ratio γ (ptubing / pcasing) = (2 / γ + 1)^{γ/(γ 1)} used for unloading valves and when the valve is designed as a simple orifice with no stem to restrict the flow throttling model if the injection pressure is less than ptran, then the throttling model is used the throttling model applies when the gas lift valve includes a throttling stem or other internal mechanism that modulates the flow the stem changes the effective flow area dynamically, altering the gas flow rate through the valve key assumptions the flow area is variable and depends on the stem position flow is influenced by the throttling mechanism and the pressure drop across the valve qg = cd aeff \sqrt{(2 δp)/ρg} where aeff effective flow area (function of the stem position and orifice geometry) the stem position affects the flow area aeff, which is calculated based on the r ratio (port to bellows area ratio) and the bellows deflection due to the pressure differential across the valve throttling mechanism the stem restricts the gas flow area as the bellows compress or expand, changing the valve's effective flow rate the throttling effect becomes significant when the pressure differential across the bellows is small used for operating valves where the valve modulates the gas injection rate based on operating conditions more common in continuous gas lift systems where precise control of gas injection is needed