Extrapolation of dynamic load behaviour on hydroelectric turbine blades with cyclostationary modelling

• Published in: Mechanical systems and signal processing Vol. 82; pp. 193 - 205
• Main Authors: Poirier, Marc, Gagnon, Martin, Tahan, Antoine, Coutu, André, Chamberland-lauzon, Joël
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.01.2017
• Subjects: Behavior; Cyclostationarity; Dynamic tests; Extrapolation; Extremes values; Fatigue; Fluctuation; Hydroelectric turbine; Modelling; Sampling; Signal extrapolation; Stochastic simulation; Strain; Turbines; Stochastic simulation; Fatigue; Signal extrapolation; Extremes values; Hydroelectric turbine; Cyclostationarity
• ISSN: 0888-3270; 1096-1216
• DOI: 10.1016/j.ymssp.2016.05.018

In this paper, we present the application of cyclostationary modelling for the extrapolation of short stationary load strain samples measured in situ on hydraulic turbine blades. Long periods of measurements allow for a wide range of fluctuations representative of long-term reality to be considered. However, sampling over short periods limits the dynamic strain fluctuations available for analysis. The purpose of the technique presented here is therefore to generate a representative signal containing proper long term characteristics and expected spectrum starting with a much shorter signal period. The final objective is to obtain a strain history that can be used to estimate long-term fatigue behaviour of hydroelectric turbine runners. •Stochastic simulation using cyclostationary modelling.•Dynamic behaviour analysis from signals measured on Hydrolectric turbines.•Data collected on short periods of time.•Extrapolation of fatigue loads from stationary signal.•Proper extrapolation of extreme values.