How to use xDGA

xDGA works by adding functionality to any existing Excel workbook with Dissolved Gas Analysis data.

To get started with xDGA, begin by downloading the *.XLL file which contains the add-in. There are two versions of the add-in.

• Download this version if you have a 32-bit Windows computer
• Download this version if you have a 64-bit Windows computer

When the ZIP file is decompressed you’ll find the three files shown in the image below.

The three files are:

• “LICENSE.txt” which contains the text of the MIT license that convers the xDGA project.
• “xDGA.ADDIN[64 or 32].xll” which is the add-in file with all the custom formulas
• “xDGA Example Usage.xlsx” is a standard Excel file with some examples of the way the formulas are used.

The easiest way to explore xDGA’s functions is to use the examples file as a starting point.

Open the “xDGA Example Usage.xlsx” file. You’ll see it contains two tabs.

The “Individual Formulas” tab contains examples of how the formulas are used on a couple of consecutive DGA samples. The “Population Analysis” tab contains an example of how the formulas would be used on a table of results for a fleet of transformers.

The next step is to ensure that the add-in is active. Activate the add-in by double clicking on the “xDGA.ADDIN32.xll” or “xDGA.ADDIN64.xll” depending on your computer architecture. If you get a Security Notice or Security Warning and allow it to run.

Once the add-in is active, go to the “Individual Formulas” tab. The input cells identify the minimum amount of data that you need to enter in order to run all the algorithms.

In order to normalise and standardise the way that data is fed into the DGA algorithms, the first thing to do is to serialise the inputs into a standard format.

This is accomplished by using the SERIALISEDGA() function as shown below.

The SERIALISEDGA() function takes the date and the nine gases that are produced as a part of a DGA analysis. This function or formula returns a standard representation of the DGA sample as a JSON formatted string. If you don’t know what JSON is, have a read here for a primer. The type of output this formula produces is shown in the image below.

You are now ready to use any of the algorithms included in the library. For example, if you wanted to know what the Duval Triangles evaluation is for this sample, you’d use the DUVALTRIALGLES() function which only takes a serialised DGA string as an input as shown below.

This formula produces the following output on the cell.

For algorithms that use two samples as an input such as the IEC_60599() function which takes the current DGA, the previous DGA, a flag that indicates whether the transformer has an OLTC and the volume of litres in the transformer. These inputs are shown below.

The IEC_60599() formula will produce the following output.

If you only have one sample, you can still use this formula by simply leaving the “Previous DGA” parameter empty. The algorithm will try to evaluate as much as the rules contained in the IEC 60599 guidelines as possible with the data available.

And, that’s it. You can use the same principle to use the other formulas contained in this Add-in, such as DUVALPENTAGONS(), ROGERSRATIOS(), etc.

The latest release of the IEEE C57.104 guidelines are currently in DRAFT and are being added to the library. Once the final version of the guidelines is published, the add-in will be updated to reflect the changes.