WELLS
Inflow and outflow on the project fluid, with nodal analysis to find the operating point. Seven lift correlations are ranked against your well tests, and lift tables leave in the format your simulator reads.
WHAT IT DOES
Well performance answers the question every production engineer starts with: how much will this well make, and what limits it? Converge builds the inflow performance relationship (IPR) from reservoir pressure and productivity, the vertical lift performance (VLP) from the tubing, deviation survey and temperature, and solves for the rate at which the two agree.
Every intersection is reported and flagged as stable or unstable, along with the minimum of the lift curve. A full pressure traverse shows where the losses are, split between gravity and friction. Sensitivities on up to 19 parameters show what moves the rate, and a seeded Monte Carlo run turns uncertain inputs into a P90, P50 and P10 rate with an influence ranking.
When you have well tests, Converge ranks all seven lift correlations against them, fits gravity and friction multipliers, checks against a gradient survey and fits the IPR, so the model you forecast with is the one that reproduces the field.
KEY CAPABILITIES
Pick the inflow model that suits the well and the data you have, from a single productivity index to back-pressure tests.
Total skin is built from its parts, so you can see what each contributes.
Seven multiphase correlations cover oil and gas wells, with the deviation survey and a temperature profile.
The operating point, and what would move it.
Screen lift options and surface limits on the same well model.
Calibrate the model to the field, then hand it to the simulator.
WORKFLOW
Enter the deviation survey, tubing, temperature, wellhead conditions or choke, and any lift.
Choose the IPR model, reservoir pressure, productivity and skin, or fit them from tests.
Pick a lift correlation, or let the matching tab rank all seven against your tests.
Read the operating rate, its stability and the VLP minimum, then the pressure traverse.
Run sensitivities and the Monte Carlo rate, and screen gas lift or an ESP.
Write VFPPROD and CSV lift tables for the simulator, and gradient CSV.
VALIDATION
No textbook worked example with a stated answer exists for these correlations, so they are compared with independent open-source implementations, and the single-phase traverse with the exact answer. A dash means the benchmark reports the agreement rather than a pass threshold.
| Check | Reference | Converge | Tolerance |
|---|---|---|---|
| Beggs and Brill (1973) | Two independent implementations, 14 flow states | within 0.1% | — |
| Gray (1974) | Two independent implementations, 14 flow states | within 0.2% | — |
| Duns and Ros (1963) | Independent implementation, bubble, slug and mist flow | within 1% | — |
| Whole gas-well traverse, 10,000 ft, dry and wet | Independent implementation | 0.02–0.18% | — |
| Single-phase liquid traverse, deviated well | Exact hydrostatic plus Colebrook friction | within 10⁻⁹ | — |
| Hydrostatic gradient in the well model (V23) | 18.06 lbm/ft³ gives 0.1254 psi/ft | 0.1254 psi/ft | 10⁻⁴ psi/ft |
Hagedorn-Brown, Orkiszewski and Aziz-Govier-Fogarasi are implemented but not yet on the benchmarks page.
INPUTS AND OUTPUTS
CONNECTED MODULES
A new well seeds its fluid from the project fluid, and a gas-condensate fluid gives a gas well.
FAQ
Enter your well tests and Converge ranks all seven correlations by average error against them, then applies the best one in a click. You can fit gravity and friction multipliers on top.
Yes. Converge writes VFPPROD lift tables over rate, tubing-head pressure, water fraction, gas fraction and lift-gas rate, in field units, plus the same tables as CSV.
Horizontal wells use the Joshi inflow model. Hydraulically fractured wells are modelled through the Cinco-Ley and Samaniego fracture equivalent skin, checked against the Prats limit. Transient and multi-fracture inflow models are not built yet.
A seeded Monte Carlo run samples the uncertain inputs and reports P90, P50 and P10 rate, with a Spearman ranking of which inputs matter most.
You can screen both. Gas lift gives the rate against injection curve with its technical and economic optimum. The ESP uses your head-per-stage curve, scaled for frequency by the affinity laws. Valve spacing and pump catalogue sizing are not part of the module.