Condition-based maintenance of wind turbines not only involves maintenance but also encompasses servicing, inspection, measurement and evaluation of the condition of the unit. The current condition can be evaluated on the basis of machine-specific overall vibration values. Until now, overall vibration values had not been defined for wind turbines. In fact, ISO 10816-3 explicitly excludes wind power plants. The new VDI 3834 closes this gap, as shown in Sheet 1, which covers vibration values for wind turbines up to 3MW.By Dr Edwin Becker, Head of Service & Diagnostic Centre, Prüftechnik Condition Monitoring, Germany
{access view=!registered}Only logged in users can view the full text of the article.{/access}{access view=registered}Wind power is an industry undergoing turbulent growth, both with regard to its volume and its energy generating capacity. System vendors have had little opportunity to test the wind turbines as their power output has grown, and this has had a negative impact on system reliability. Particularly because of the frequent occurrence of gearbox damage in 2002/03, system insurers introduced a so-called revision clause and cancelled almost all old contracts. The revision clause stipulates that ‘all roller bearings in the drive train must be replaced after 5 years or 40,000 operating hours at the latest, unless a suitable Condition Monitoring System (CMS) has been installed’. These types of CMS can be used for function diagnosis and/or for fault diagnosis.
Function diagnosis refers to the measurement of functional and operating parameters and overall vibration values. That knowledge is required for the proper functioning and long-term operation of rotating machinery. Fault diagnosis is the determination of damage conditions on machinery and machine components.The use of CMS has made it clear that wind turbines are highly complex machines for which the overall vibration values must be systematically determined and made available. The following points had to be taken into consideration in creating the new guideline:
- The functional and structural design of wind turbines and their components.
- The interaction between the individual machine train components (modules) being tested.
- Information and experience regarding the possible faults and damage occurring in the individual modules during operation and their economic impact.
- Knowledge of operation-related and machine-related vibration influences, the diagnosis procedures that can be applied, and the diagnosis conditions that need to be adhered to and their limits.
Measurement Technology and Characteristic Values
Piezoelectric accelerometers can be used to measure both low frequency vibrations beginning with 0.1Hz and high frequency vibrations up to 6kHz, as defined by VDI 3834. In chapters 2 and 3 of the VDI 3834, clear specifications are given for the installation location of these types of accelerometers. Acceptance measurements should be made in the following typical measurement directions: radial vertical, radial horizontal and axial, Figure 1 shows some typical locations where measuring devices may be fitted. A minimum load of 20% is required. Because of the natural fluctuations in wind load, the VDI 3834 specifies longer measurement times ranging from 1 to 10 minutes and even requires a root mean square to obtain stable and meaningful vibration values for slowly rotating components.
Characteristic Values for Drive Train Components
The VDI guideline divides the drive train sections into their main components and assigns overall vibration values to the most important of these. In this way, component-specific vibrations can be classified and wind turbines and their components showing unfavourable behaviour with respect to their vibrations can be identified. The VDI 3834 is based on a statistical analysis of vibration measurement results of more than 450 wind turbines and defines threshold values in terms of vibration velocity in mm/s and vibration acceleration in m/s² for the drive train components (generator, gear and main bearing) and for the nacelle/tower. The threshold values were defined as component-specific frequency bands. The VDI 3834 also gives recommendations for warning and alarm thresholds. Figure 2 shows the acceptable levels for overall vibration velocity and Figure 3 shows them for overall acceleration.
In Level 1 monitoring, we differentiate between the remote monitoring of these overall vibration values and the remote monitoring of characteristic diagnosis values. But these methods are not new. In machines and industrial plants, the vibration values of the ISO-10816-3 are used to monitor the general vibration condition and, increasingly, detailed monitoring uses frequency-based or order-based characteristic trending values.
Assessing the Vibration Level and Reducing Vibrations
Based on the overall vibration values, it is now possible to assess the vibration levels of wind turbines and to compare these. After all, the experience found during service that ‘less vibration means a longer machine life’ must also apply to wind turbines. The early detection and reduction of elevated vibration levels therefore extends wind turbine service life.
Identifying Corrective Measures
The required measures can be identified by means of a condition diagnosis. Diagnosis specialists use amplitude spectra, envelope spectra, time signals and/or cepstra to detect unusual vibration signals, to identify dominant excitations and to evaluate frequency specific trends using the waterfall display function (www.telediagnose.com, issue 12). A few examples of how the availability of systems can be increased using vibration results are presented below:
- Detecting additional vibrations resulting from a generator fault: Figure 4 shows the trend of a wind turbine generator in which an increase in vibration amplitudes indicated an impending machine fault several weeks in advance. After the generator was replaced, the overall vibration values returned to normal. It should be pointed out that such vibration changes only arise if the affected drive train component is dominant in the frequency band. Overall vibration values do not rise when the damage is not dominant in the amplitude spectra.
- Identifying deviations in the machine alignment: During telemonitoring of a wind turbine, elevated overall vibration values were detected. The frequency analyses showed additional vibrations due to poor machine alignment. The machine was then aligned according to suitable alignment targets (Becker and Holstein, 2007). The overall vibration values became significantly lower, as seen in Figure 5.
- Reducing rotor blade imbalances: Rotor blade imbalances lead to rotational excitations and, among other things, increased loads on bearings and components. While they are of very low frequency in wind turbines, they can result in vibration amplitudes of 100mm/s. Measurements must be taken with linear vibration sensors and longer measuring times, as prescribed by VDI 3834. Figure 6 shows the effect of field balancing on vibrations, applying the recommended G16 balancing grade for rotor blades (www.telediagnose.com, issue 12). In this particular case, the additional vibrations, caused by imbalance, were actually reduced to the point where the difference was noticeable in the nacelle.
These examples alone illustrate how the targeted use of measuring and testing techniques make it possible to reduce vibrations in already installed wind turbines. The VDI 3834 enables manufacturers and operators to assess the vibration condition of wind turbines and reduce them by implementing specific corrective measures in order to reach state-of-the-art threshold levels.{/access}




