DONATE

IAHR Document Library


« Back to Library Homepage « Book of Abstracts of the 9th IAHR Europe Congress (Lulea, 20...

Predictive Maintenance of a Reversible Francis Turbine in a Pumped-Storage Hydropower Plant Using AI and ML Methods

Author(s): Albert Kindl; Vladimir Haban; Petr Konas; Martin Hudec; Pavel Rudolf

Linked Author(s):

Keywords: No Keywords

Abstract: Artificial intelligence (AI) and machine learning (ML) are gaining increasing significance and applications in the field of hydropower. Modern predictive maintenance methods utilizing AI and ML hold strong potential to enhance the operational efficiency of hydropower plants, and consequently, the overall efficiency of the power system. This paper discusses the processing and application of predictive systems implemented at the Štěchovice II Pumped Storage Hydropower Plant, located on the Vltava River in the Czech Republic. The plant operates a reversible Francis turbine with an installed capacity of 45 MW. Since March 2023, data have been continuously collected from two sources: sensors deployed by the Brno University of Technology (BUT) and sensors belonging to the plant operator, ČEZ a.s., the leading company owning and operating power plants in the Czech Republic. The ČEZ a.s. sensors provide conventional vibration diagnostics, measuring absolute bearing vibrations and relative shaft vibrations. The measurement complies with ISO 20816-5. The high-speed sensors deployed by BUT include pressure and acceleration sensors installed in the turbine draft tube, as well as two microphones - one positioned near the draft tube and another on the turbine cover.

DOI:

Year: 2026

Copyright © 2026 International Association for Hydro-Environment Engineering and Research. All rights reserved. | Terms and Conditions