Software Detects Damage to Wind Power Plants

Siemens is developing an intelligent sensor system capable of detecting impending damage to offshore wind power plants. Siemens researchers are working on self-learning software that automatically analyzes the gears of a wind power plant and their vibrations and predicts potential failures on this basis. As reported in the research magazine Pictures of the Future, a prototype version of the software has successfully passed initial tests.

Because of weather conditions, offshore wind power plants are accessible for only a few days a year. Without some form of telemonitoring system, operators are therefore unable to monitor the status of equipment and identify any damage at an early stage. Many companies use the Condition Monitoring System from Siemens, which essentially consists of sensors installed on the gearing of the plant. In addition to monitoring critical parameters — on the basis of the vibrations and harmonics of the gear mechanism — and thereby the mechanical functionality of the plant, the system also transmits the results to the plant operator.

Formerly, however, it was only possible to set rough parameters for the characteristic values. This is because each gear has its own vibration characteristics, which constantly vary according to ambient conditions such as temperature, wind velocity, and oil level. In the past, Siemens engineers therefore determined warning and alarm limits only with regard to the range within which each individual gear can still be rated as fault-free.

In the future, however, the researchers will be supported by a so-called vibration diagnosis module (VDM). Developed by Siemens Corporate Technology (CT), this self-learning software provides automatic analysis and prevention of mechanical failures. In combination with a general-purpose measurement acquisition and diagnostic system, the VDM is able to determine the status of each of the gears on the basis of its vibration characteristics and ambient data. It then defines the limits at which the first changes resulting from wear or defects become perceptible.