Forecast a well on an Arps hyperbolic decline

Rate and cumulative by month from an initial rate, a nominal decline and a b-factor, to the month the well reaches its economic limit.

Open in the workbench

Engineering use case UC-DCA-001 (Forecast Single Well Production). To fit the decline to a history first, see dca-model-select-eur; to set the limit from the economics, dca-economic-limit.

ArpsHyperbolicDecline(Qi, Di, b) is unit-agnostic: Di is per unit of the time you pass to Rate and Cumulative, and the cumulative is in rate × that time. Here the rate is bbl/month against months, so the cumulative is a true volume in barrels. Mixing bbl/d with months puts the cumulative out by about 30.4.

The decline is nominal, not effective. A 15 %/yr effective decline is a nominal one of −ln(1 − 0.15) ≈ 0.1625 /yr; DeclineRateConversions converts between them.

b = 0 is exponential, b = 1 harmonic. A b above 1 fits early transient data from tight wells but runs to an unbounded cumulative; cap the forecast at an economic limit, as TimeToReachRate does here, or switch to a modified hyperbolic.

Source

using System;
using System.Linq;
using Team76.Engineering.DCA.DeclineModels;

public class S : ICalcScript
{
    public ScriptResult Run(ICalcContext ctx)
    {
        // 30,000 bbl/month initial rate, 6 %/month nominal decline, b = 0.6. Rate and time
        // share the month, so the cumulative is in barrels.
        var well = new ArpsHyperbolicDecline(Qi: 30_000, Di: 0.06, b: 0.6);

        // Run to an economic limit of 1,500 bbl/month.
        var months = (int)Math.Ceiling(well.TimeToReachRate(1_500));
        var t = Enumerable.Range(0, months + 1).Select(m => (double)m).ToArray();
        var rate = t.Select(well.Rate).ToArray();
        var eur = well.Cumulative(months);

        return ctx.Result()
            .Set("Forecast", new ScriptChart(t, rate, title: "Arps hyperbolic forecast", xLabel: "month", yLabel: "bbl/month"))
            .Set("Life (months)", months)
            .Set("EUR (Mbbl)", Math.Round(eur / 1000, 1));
    }
}