Wellhead Forecasting Theory

Note:   This theory only applies to forecasting in analytical models.

When forecasting with wellhead pressure, a rate is needed for these two calculations:

Since the same unknown rate is required to solve both calculations, an iterative process is needed. For wellhead forecasting in Harmony, the following procedure is used:

1. Guess an initial value for the primary fluid rate (qguess), where the primary fluid is defined by the analysis (i.e., gas or oil). If there are fluid ratios (such as WOR, GOR, or WGR) specified, they are used to calculate the secondary rates.

2. Find the sandface pressure using your primary fluid rate from step 1, corresponding secondary fluid rates, and specified wellhead flowing pressure. For details on this procedure, see Pressure Loss Calculations.

3. Use the Analytical model to calculate sandface pressure based on qguess.

4. Compare sandface pressures calculated in steps 2 and 3. Iterate on the rate until the sandface pressures calculated in steps 2 and 3 are the same.

The procedure above is followed for each timestep in the forecast. Therefore, at each timestep, Harmony looks for a rate where "sandface pressure based on the reservoir model = sandface pressure based on the wellbore model".

This iterative process can also be described graphically using the well-known nodal analysis method for wellbore optimization. In nodal analysis, we look for the intersection between the inflow performance relationship (IPR) and tubing performance curve (TPC). When relating these methods, the IPR is equivalent to the analytical model, while the TPC represents pressure calculations based on the wellbore model. An example of this is shown below:

Wellhead Forecasting Options

To see how these options are enabled, see Forecasting Using Wellhead Pressure.

Oil Forecasts

When running a wellhead forecast for an oil well, you can specify the GOR and/or WOR. Note that this GOR is assumed to be measured at the separator, so it should be greater than or equal to the solution GOR. If you specify a GOR that is less than the solution GOR, the forecast results may be inaccurate.

Gas Forecasts

When running a wellhead forecast for a gas well, you can specify WGR. If you have an OGR, we recommend that you perform an oil forecast. If you have a CGR, see Forecasting Gas Condensate Systems.

Wellhead Forecasting Considerations

We recommend that you keep the following considerations in mind:

Premature Forecast Stoppage

In some cases, a wellhead forecast may stop before the end of its specified duration, even though the production rate has not yet dropped to zero:

This may happen if there is no rate that can solve both the wellbore pressure loss calculation and the model calculation. This can be seen graphically in the diagram below. As time progresses, the average reservoir pressure gets smaller and the IPR curve “shrinks”. However, the TPC remains stationary since it only depends upon a fixed wellhead pressure and wellbore. Consequently, a time will come when the IPR shrinks so much that it cannot intersect the TPC curve, which is seen in the dashed IPR curve below. At that moment, the forecast stops, because flow is no longer possible at the specified conditions.

Convergence Issues

Sometimes TPC curves are not smooth, which happens when a multiphase correlation switches from one flow pattern to another. In this case, the iterative procedure does not always converge to a solution. If the solution fails to converge, Harmony extrapolates a rate from previous timesteps, and calculates a corresponding wellhead pressure, and a warning is given. You can check calculated wellhead pressure in Tables\Forecast to ensure that it is close to the wellhead pressure you specified.

Note:   If you are having convergence issues, one possible workaround is to change the number of steps and/or the correlation used for wellbore calculations.