TCM^® & PythiaTM – a symbiosis for vibration diagnostics

Based on data from the TCM^® solution, Siemens Gamesa has developed a system, PythiaTM, which can simulate the damage development in each wind turbine component.

In Brande, Denmark, a fully automated system based on the TCM^® setup and Siemens Gamesa’s PythiaTM framework monitors more than 11,000 Siemens Gamesa wind turbines worldwide, analyzing the steady flow of data to detect minute irregularities and impending failures. The remote monitoring and diagnostics of wind turbines are part of Siemens Digital Services. Approximately 2300 parameters on each turbine are monitored around the clock.

PythiaTM is named after the Oracle of Delphi, as the purpose of PythiaTM is to look into the future through advanced vibration analysis. PythiaTM is a framework that has been operational since the fall of 2014. Based on the TCM^®, it enables advanced algorithms to look for changes in the vibrational behaviour, according to specific overall patterns defined by machine learning algorithms. The models are centered on generic mathematical algorithms containing many comparable and reliable condition indicators extracted and processed by the TCM^® system.

PythiaTM performs millions of extractions every day, enabling the human experts in Brande to dig even deeper into the condition indicators detected by the TCM^® system.

Creating algorithms with TCM^® data
The output from PythiaTM is the condition classification for the components in the drive train. These are used to detect impending failures and classify their severity.

Big data analysis of large historical data sets can be done using the TCM^® Advanced Diagnostics tool. This tool helps find the root cause and classifies vibrational patterns, enabling Siemens Gamesa to predict potential damage up to five years in advance. When possible damages occur, a notification is created in the case handling system “MORS” and is immediately sent to the service technician and the Siemens Gamesa. Twenty-four hours later, this information will appear in the Vibration Diagnostic Health Trend reports on the “Wind Dialogue” customer portal. The service technician uses this information for the maintenance of the wind turbine.

“Big data analysis of our large historical data pool from the TCM^® system results in the intelligent algorithms that power our diagnostic models running in PythiaTM, an agile platform allowing us to improve our diagnostics continuously.”

- Bo Roemer-Odgaard, Head of Vibration Diagnostics, Remote Diagnostic Center, Siemens Gamesa Renewable Energy, Denmark

Turning big data into valuable knowledge and insights
Vibration signals from the drive train (main bearing, gearbox and generator) are gathered by the TCM^® M-System in the wind turbine's nacelle and processed to extract features, i.e. condition indicators, per component in the drivetrain.

Together with select measurements, these condition indicators are transferred to Siemens Gamesa’s TCM^® server in Brande, where the PythiaTM platform integrates and extracts data using its RESTful interface. A RESTful interface is a secure and flexible data model exposing an API with which other platforms, such as PythiaTM, can interface.