Influence of load spectrum assumptions on the expected reliability of hydroelectric turbines: A case study

• Reliability assessment of hydroelectric turbine runner blades.
• Combined High Cycle Fatigue (HCF) and Low Cycle Fatigue (LCF) loading.
• Sensitivity to load spectrum assumptions.
• Load spectrum essential parameters.

A lack of observed data can lead to significant discrepancy between estimated and actual fatigue reliability. In this study, the load spectrum derived from the strain measurements and operation history of a hydroelectric turbine are used to identify the information necessary to avoid major bias in the estimated fatigue reliability of such structure. Our results demonstrate that a limited number of parameters need to be considered. Any further simplifications may lead to major reliability estimate discrepancies. Furthermore, we observe that the parameters which influence initial reliability hide the influence of other parameters on the reliability failure rate associated with the structure’s life expectancy. We conclude that a typical sensitivity analysis made on parameter values needs to be complemented with a sensitivity study on the chosen assumptions in order to properly evaluate the risks associated with the operation of hydroelectric turbines.