On the use of Vibrational Hill Charts for improved condition monitoring and diagnosis of hydraulic turbines

To cope with the intermittent power supply of new renewable energies and demand fluctuations, Francis turbines are required to operate more in an extended operating range, far away from design point. With this behavior, it is very complex interpret trend vibration parameters typically used Condition Monitoring define reasonable alarm trip levels valid for all range unit working steady conditions. As efficiency curves represented as a function net head load (Hill Chart), paper we propose represent most relevant surfaces, called Vibrational Hill Charts, which allow accurate evaluation indicators their trends better classification abnormal values. show potential complete database obtained after 2 years monitoring large Unit (444 MW rated power) has been used. The mapping performed by means Artificial Neural Networks. It shown that setting action based on resulting maps, rather than constant value, diagnosis capacity achieved Receiver Operating Characteristic will be improved. Furthermore, phenomena such erosive cavitation, hard detected, could also assessed use multidimensional analysis Chart. conclusion, Chart, condition hydraulic