The failure rate of gearbox, motor and other key components in wind turbine transmission system is high, and the failure of gearbox and motor is caused by bearing failure.
With the rapid development of wind power technology, the requirements for the overhaul and maintenance of the wind turbine is becoming more and more strict, due to the change of the unit type, size, service environment as well as the installation error and so on reasons, lead to bearing failure mechanisms are also different, so the bearing failure analysis is varied, the wind generator and chooses the bearing mostly belong to large bearing, However, due to the particularity of size and working environment, large bearings are often damaged by local deformation, local vibration, and temperature rise, which is not the same as ordinary bearings. Therefore, the bearing status of wind turbines should be monitored in real-time effectively and maintained reasonably at the best time.
There are a variety of characteristic signals (vibration, acoustic emission, stress, temperature, lubricating oil, etc.) in wind turbine bearings. All kinds of characteristic signals can reflect the health state of bearings to a certain extent, so the characteristic signals can be used for fault analysis and diagnosis of bearings.
The overall strength and life of bearings can be improved by the control of load, clearance, rotational speed, and the modification of the rolling body.
Research on bearing Fault Diagnosis of Wind Turbine: Early parts change characteristic of fault symptoms will cause a series of various physical quantities, at the same time, the fluctuations of a single physical characteristic may be caused by several parts, such as, when the yaw system of gear and bearing fatigue failure, will cause the vibration of the generator rotor, which destroy the generator’s internal balance, Then the fault signal will be attached to the electrical characteristic quantity or vibration characteristic quantity of the stator and rotor of the generator. Therefore, it is necessary to combine the actual situation of bearings in different parts and the differences of different units and transform from the previous single and multi-feature longitudinal data fusion analysis method to the multi-angle analysis method, so as to seek the fault data characteristics and evolution rules of bearings in various parts.