Vibration based condition monitoring and fault diagnosis of wind turbine planetary gearbox: A review

A review of failure modes, condition monitoring and fault diagnosis methods for large-scale wind turbine bearings

The relationship between kurtosis- and envelope-based indexes for the diagnostic of rolling element bearings

Vibration-based bearing fault detection for operations and maintenance cost reduction in wind energy

A comparative study of the effectiveness of vibration and acoustic emission in diagnosing a defective bearing in a planetry gearbox

Separation and enhancement of gear and bearing signals for the diagnosis of wind turbine transmission systems

A relatively recent development for the condition monitoring of oil-wetted machinery has been the in-line full-flow inductive wear debris sensor (hereafter referred to as inductive wear debris sens...

Determining inductive sensor wear debris limits for rolling contact fatigue of bearings

Spectrometric oil analysis has been a popular technique used by oil analysis laboratories to determine the concentration of elements in an oil sample. The primary application of this technique has been the identification of fine wear in lubricated systems as an indicator of abnormal wear in order to prevent failure. It has also found some use for monitoring contaminants and fluid additives. While the literature has varying accounts of the effectiveness of this technique for machinery condition monitoring purposes, the introduction of fine filtration to many modern machines has profound implications for the effectiveness of spectrometric oil analysis. This paper will assess the effectiveness of spectrometric oil analysis in a fine filtered system and compare it with the results from an inductive wear debris sensor. The inductive wear debris sensor is a relatively new sensor for detecting abnormal wear in recirculating lubrication systems. A combination of experimental data and aircraft data is presented.

On the impact of fine filtration on spectrometric oil analysis and inductive wear debris sensors

Measuring a machinery health parameter in a condition monitoring program is only half of the equation; at some point a limit must be set that initiates either further intensive monitoring or maintenance action. Limits are a fundamental and critical component of an effective condition monitoring program (often an integral part of a HUMS) and they can directly impact aircraft availability (e.g. rotor track and balance checks), or result in substantial maintenance burden (e.g. removal of machinery for overhaul based on wear debris). Limits therefore need to have a solid basis and often this is unknown, unclear or is questionable resulting in sub-optimal performance of the condition monitoring program. The traditional statistical-based philosophy behind setting limits dates back to early Statistical Process Control (SPC) work credited to Shewart [1] and Deming [2] in the 1930’s. Whilst the familiar Gaussian-based statistical approach (based on the average plus or minus some number of standard deviations) and the Cumulative Distribution Function (CDF) approach have been used extensively, the authors argue that such a fundamental component of a condition monitoring program now deserves more attention and rigour. Interestingly there is scant information in the literature about methodologies for setting in-service limits where the requisite minimum data sample is unavailable, unreliable or simply where the traditional approaches are not optimal or lack physical meaning. Conceptually, methodologies can be researched and developed that have a reasoned and validated foundation in the physics or chemistry of the machine element. This paper will provide some examples of areas where alternative approaches may have provided better outcomes. The authors argue that a purely statistical approach may not necessarily serve condition monitoring programs well in all instances and can in fact result in sub-optimal limits being set. Similarly, limits based on small data sets derived during accelerated production testing of aviation machinery may not be representative of the true in-service behaviour of the equipment. Questioning the pedigree of limits and reviewing limits in-service is essential for ensuring optimal performance of condition monitoring programs, however experience shows that this is often neglected.

https://www.humsconference.com.au/Papers2013/45_Becker.pdf

Questioning how limits are set

The prevention of catastrophic failure of helicopter gearboxes is the primary objective of gearbox vibration monitoring. In the past, gearbox vibration monitoring involved the temporary installation of sensors and cabling followed by dedicated flights to record data. This amounted to an added workload for squadron personnel simply to acquire the vibration data. The Royal Australian Navy (RAN) recently commenced permanently installing (hard-wiring) the Sea King and Seahawk helicopters with vibration sensors, cabling and junction box. This enables rotor track and balance, airframe vibration survey, engine vibration survey and gearbox vibration analysis to be performed with significant time savings. The gearbox vibration analysis component of this system has been developed by the Defence Science and Technology Organisation (DSTO). The other aspects of vibration analysis are conducted using commercially available equipment. This paper discusses the DSTO developed gearbox vibration analysis element of the RAN hard-wired system. The system is described in detail and the intended operating procedures outlined. The current status of the RAN hard-wire program is discussed together with proposed systems for other Australian Defence Force (ADF) helicopters.

/pdf/a-review-of-permanently-installed-helicopter-gearbox-4uw96a29d7.pdf

David Forrester

Papers

Separation and enhancement of gear and bearing signals for the diagnosis of wind turbine transmission systems

Determining inductive sensor wear debris limits for rolling contact fatigue of bearings

On the impact of fine filtration on spectrometric oil analysis and inductive wear debris sensors

Questioning how limits are set

A review of permanently installed helicopter gearbox vibration monitoring systems in the australian defence force.