Locating Faulty Planet Gears Using External Vibration Measurements

In this study, a mesh phasing-based approach is developed to locate the positions of faulty planet gears using external vibration measurements. Previous studies have illustrated how this can be achieved using internal vibration measurements recorded from a sensor placed on the planet carrier. It was shown in these studies that the timing of identifiable fault symptoms in the vibration signal relative to the phase of the gear-mesh component depends on which of the planet gears carries a fault. A signal processing technique is then developed to locate the position of a spalled gear using internal vibration measurements. However, internally mounted sensors are not commonly used in planetary gearboxes and it is much more convenient to mount sensors externally, for example on the gearbox casing. Therefore, this study extends the concept of using mesh phasing relationships to locate faulty planet gears, this time using external vibration measurements. The updated procedure is validated using experimental data collected from a test-rig running under a range of operating conditions. The results show that the updated procedure is able to identify the locations of faulty planet gears so long as an absolute phase reference (for example from a tachometer) of the planet carrier is available.

The Effect of Wear Evolution on Vibration-based Fault Detection in Tapered Roller Bearings

Rolling element bearings (REBs) are typical tribological components used widely in rotating machines. Their failure could cause catastrophic damage. Therefore, condition monitoring of bearings has always had great appeal for researchers. Usually, the detection and diagnostics of incipient bearing faults are achieved by characterising the weak periodic impacts induced by the collision of defective bearing components. However, race wear evolution, which is inevitable in bearing applications, can affect the contact between bearing elements and races, thereby decreasing the impact magnitudes and impeding detection performance. In this paper, the effect of wear evolution on the condition monitoring of rolling bearings is firstly analysed based on internal clearance changes resulting from the wear effect. Then, an experimental study is ingeniously designed to simulate wear evolution and evaluate its influence on wellknown envelope signatures according to measured vibrations from widely used tapered roller bearings. The fault type is diagnosed in terms of two indices: the magnitude variation of characteristic frequencies and the deviation of such frequencies. The experimental results indicate a signature decrease with regard to wear evolution, suggesting that accurate severity diagnosis needs to take into account both the wear conditions of the bearing and the signature magnitudes.

Comparison of Electrical and Vibration Analysis Methods for Mechanical Fault Monitoring in Wind Turbine Drivetrains

Mechanical faults occurring in drivetrains are traditionally monitored through vibration analysis and, more rarely, by analysing electrical quantities measured on the electromechanical system involved. However, a monitoring method that is able to take into account all of the information contained in three-phase electrical quantities was recently proposed. The goal of this paper is to compare this threephase electrical approach with the usual vibration-based method in terms of its capability to detect mechanical faults in drivetrains. In this context, a 2 MW geared wind turbine operating in an industrial wind farm was equipped with accelerometers near the main bearing and electrical sensors on the stator of the electrical generator for several months. During this period, an important mechanical fault occurred in the main bearing of the system. The evolution of the fault indicators computed by the two previous approaches were compared throughout this period of time. All of the indicators behaved similarly and showed the development of an inner bearing fault in the main bearing. This demonstrated that a mechanical fault occurring in a drivetrain can be monitored and detected by analysing electrical quantities, even if the fault is located some distance from the electrical generator.

Kalman Filtering using Trigonometric Polynomials for Gear Fault Diagnosis Under a Variable Speed Condition

It is well known that the presence of cracks on a gear tooth is manifested by both amplitude modulation (AM) and phase modulation (PM) and their estimation is crucial for the diagnosis and prognosis of the state of the gearbox. In this paper, a new demodulation technique for gear fault diagnosis is proposed based on Kalman filtering combined with a polynomial basis function of the rotating speed. First, the problem of demodulation is formulated in terms of the state-space modelling of the vibration signal. This is achieved by approximating the modulations using the orthogonal trigonometric polynomial function of the rotating speed. Kalman filtering is then used to estimate the coefficients of the polynomials in order to reconstruct the modulation signals. This approach is different from classic approaches such as the narrowband demodulation technique (NBDT) or time-synchronous averaging (TSA). The advantages of the proposed approach over the latter are discussed in this paper and the efficiency of the new approach is also evaluated using both synthetic and real gearbox vibration signals.