DFITAnalysis-Python

Diagnostic Fracture Injection Test Analysis Program

This program analyzes the fracture injection/fall off data using standard G-Function and Square Root Time analysis techniques.

Requirements

1. Python 3
2. Numpy
3. Matplotlib (with appropraite backends)

How to use

Open, modify and run Well_1Analysis.py file. Well_1Analysis.py script loads data from DFITData.csv file. It requires time, pressure and rate data as columns, which can be chosen as t_col, p_col and r_col respectively. Header rows may also be skipped with skip_rows. Most importantly, the injection time (seconds) is also required and inputted as tp.

Following line loads and creates the well object, and makes it ready for further analysis.

Well_1 = DFITAnalysis(filename, skip_rows, t_col - 1, p_col - 1, r_col - 1, tp)

Run all

The script will automatically run both: G-function and square root time analyses. Well_1.runAll(...) may be commented out if the user requires only one of the analysis or wants to run it interactively. In which case, following instructions must be followed.

G-Function analysis

G-function analysis can then be called on the above (Well_1) object as follows (run once):

  Well_1.GFunction.analysis(Well_1, dwindow)

Multiple G-function plot(s) can be prepared as:

  Well_1.GFunction.plotData(Well_1, 0, 0, 0, 0, 0, 0, 0, 0)

Finally the user can choose a straight line through the origin, followed by identifying closure point (for each plot):

  Well_1.GFunction.identifyClosure(Well_1)

After running, the program will report the closure pressure on console as well as an annotation on the figure.

Square root time analysis

Similar to above, square root time analysis may also be performed by replacing GFunction with SquareRoot in the above code.

Quirks

1. Use the dwindow parameter to adjust the smoothness of the derivative, depending on your data.
2. If the lines seem to plot outside the chart, limits may be adjusted while calling plotData function.

Sample data output figures

Further reading

1. Barree slides