PO083

Gearbox high speed bearing skidding risks during an electrical grid fault

Jan Helsen 1
1Vrije Universiteit Brussel/OWI-lab, Brussel, Belgium, 2ENGIE lab-Laborelec, Linkebeek, Belgium

Abstract

The higher share of distributed energy production in the overall energy mix has caused electricity grid regulators’ requirements to become stricter. Turbines are required to stay online during short voltage dips, the so-called ride through phenomena. From a grid operator point of view this requirement is logical; however, from a drivetrain structural integrity point of view this poses a challenge. The grid loss event causes a dynamic excitation that propagates from the generator to the other components of the wind turbine drivetrain. This paper focuses on the influence of this event on gearbox reliability, more particularly the high-speed stage gears and bearings.  For gears the influence of the impact loads due to electrical phenomena is important, since these loads could potentially exceed critical levels for crack initiation.  An important failure mechanism for bearings is excessive roller slip or skidding. In this case the roller does not follow its kinematic trajectory anymore. Excessive slip conditions are indicated to have potential links to complex bearing failure mechanisms such as e.g. white etching crack (WEC), amongst others by Greco and Stadler. Axial shaft movements, that are typical for helicoidal gear setup, are believed to increase the damage risk of the gear bearings. This paper discusses the results of an experimental investigation on the influence of positive to negative torque oscillations triggered by a grid loss type of event on gear impact loads and bearing slip in a gearbox high-speed stage (HSS).

Method

This paper discusses the results of an experimental investigation on the influence of positive to negative torque oscillations triggered by a grid loss type of event on gear impact loads and bearing slip in a gearbox high-speed stage (HSS).  All experiments were conducted on full-scale systems: several experiments on a nacelle on a 2.5MW dynamometer and a one-year measurement campaign in the field. In both cases the high-speed stage of the used gearbox is instrumented extensively. For the test-rig case both internally and externally, whereas in the field on external and torque instrumentation was used. Detailed measurements at the gears and bearings are used to illustrate the gear impact loads and bearing skidding.

 

Results

The paper consists of three parts: first a detailed analysis of HSS behavior during the grid loss event on the dynamometer.

The second part exposes the link between the torque signature during the event and gear impacts and HSS bearing slip. It is shown that the torque signal shape contains information about gear impacts and that it links to the amount of slip.

The final part exploits the learned information about the link between torque signature and HSS behavior to learn about HSS response from field data. Impacting gears and axial impact loads at HSS were detected by a HSS accelerometer. These impacts are linked to the torque signature. A measurement of the axial shaft position confirmed important axial movements of the HS rotor.

Conclusions

This paper showed that the excitation coming from electrical grid dynamic events has the potential to cause unfavorable mechanical loading at a wind turbine gearbox HSS. An example is the impact loads that were detected at the HSS in the field turbine during the event. The torque response linked to a grid loss event is shown to be characterizing the gearbox response during the event. Shaft torque is therefore a good parameter to characterize the appearance of unfavorable loading of (HSS) drivetrain components. This can help in further understanding these links and prevent the harmful consequences of this type of events.

Objectives

The main added value of the paper to industry is that it experimentally shows that changing electricity grid regulations has a direct impact on the drivetrain mechanical degradation. That dynamic events linked to these electricity grid events, such as grid losses are important to have on the radar. This is the case particularly for offshore farms where all turbines are connected in one local grid and the farm owner is responsible for this grid. For owners it is important to understand these consequences. Additionally for turbine manufacturers it is important to know this correlation between electrics and mechanics. In general to the scientific community it is interesting to have experiments on full scale both on dynamometer and field turbine documented.