Identification of hydro-turbine runner with cracked blade based on power spectral density transmissibility in operation

For peaking hydropower plants, the damage to the runner blades is a common problem. Real-time online detection of runner blade damage can effectively prevent major accidents in the hydropower unit. However, the online monitoring of runner status is still impossible to achieve due to the lack of fundamental research on the vibration response characteristics of runner damage. Hence, in this research, the vibration response of Francis turbine runner with cracked blade compared with healthy runner under hydraulic excitation are numerically studied and the Power Spectral Density Transmissibility (PSDT) method is employed to conduct a detailed analysis of the response frequencies. The results indicate that there are significant differences in the vibration response frequencies between healthy runner and runner with cracked blade under the influence of hydraulic excitation. The PSDT method can effectively identify the characteristic frequencies of runner with cracked blade during operation and the identification of characteristic frequencies does not depend on changes of the monitoring location. The high-order frequencies, such as 532, 660 and 720 Hz, caused by cracked blade compared with healthy runner can also be captured by PSDT. This study can provide a theoretical research foundation for the online condition monitoring of runner condition.