Standard Vibration Analysis Methods

Abstract

The following chapter presents the summary of signal processing methods used in vibration based condition monitoring. It starts with the examples of signals which are then used to demonstrate all the analyses. First, simple simulated signals are described. Then, more complex features are added until real vibration signals are presented and described. The Fourier Transform is used to obtain the frequency spectrum of a signal. It is a fundamental vibration analysis tool. This transform is presented in details covering continuous and discrete signals. There are factors, like leakage, windowing and digitization which have important influence on its results and they are also covered. A powerful and popular method for rolling element bearing fault detection is the envelope analysis which is presented next. Signal examples are used and the importance of a proper selection of a demodulation band is described. Since modern wind turbines work with varying rotational speed, resampling is very important to decrease this type of non-stationarity. It is also described. Next two methods are: time synchronous analysis as it is important for gear fault detection, and time-frequency method used for non-stationary states. The chapter is concluded with a presentation of broadband and narrowband features used for tracking individual drivetrain components and with description of standards applicable to wind turbines. The special importance is a new part of ISO 10816-21 devoted to onshore turbines with gearboxes.