TrafoStick™

The TrafoStick was developed as a robust and compact online sensor for use on power transformers. Only materials that can be permanently exposed to transformer oils have been used. The test and calibration routines were developed on the basis of the evaluation of over 3800 oil samples from over 900 different transformers.

The core of every TrafoStick is the smart deep learning algorithm. Based on and developed through a massive evaluation of more than nineteen thousand, in laboratories analyzed, multivariant insulating oil samples of  over 3500 power-transformers from all over Europe (including Germany, Italy, Austria and Switzerland). The calculation, which takes into account the strong correlation between oil moisture (WC), acid number (TAN), temperature (T) and breakdown voltage, is carried out in the sensor by a 32-bit embedded system using a floating point process (FPU) using a look-up table (Look- up-table). The representation of the limit behavior of the oil samples has already been taken into account in the look-up table.

As a first in the industry, the transformer allows the most important parameters of transformers to be monitored continuously and throughout their lifetime. With the information to be obtained by monitoring these parameters, it is possible to make an estimation about the remaining life of transformers, which is one of the most critical components of power systems, and to create maintenance plans. Automatic and real-time monitoring of the dielectric strength of transformers makes it possible to monitor the condition of the transformer, recognize trends and, above all, take timely measures to improve the reliability of the transformer and possibly the entire system.