Search

Wind turbines vibration signals carry useful information about their conditions. However, such signals are usually very complex and comprise high number of vibratory signatures. Moreover, the cost of setting up vibration measurement systems using conventional accelerometers is very high especially in the case of multi-location measuring points are required. This paper contributes towards developing an inexpensive and accurate condition monitoring method based on remotely measured air-born acoustic signals. Fault detection capabilities are assessed by determining its coherence with the well-understood vibration induced signals. A new generation of recently improved condenser microphone used as input for the proposed coherence analysis algorithm. Special arrangement to eliminate reverberation problems associated with air-borne acoustic signals was designed. Consequently, coherence analysis is carried out to compare the information supplied by the accelerometer and the microphone of a three bladed horizontal wind turbine under different operating conditions. The air-borne acoustic signal exhibits good coherence with the vibration ones at rotating speed, gear meshing frequencies and fan passing frequency bands. More important, air-borne acoustic data was found to contain more pronounced information within wider spectrum compared with the vibration measurements. The presented results open doors for deeper investigations on the diagnostic capabilities of such technique and probably lead to fully adoption by the wind turbine industry.

Helical gearboxes play a critical role in power transmission of industrial applications. They are vulnerable to various faults due to long-term and heavy-duty operating conditions. To improve the safety and reliability of helical gearboxes, it is necessary to monitor their health conditions and diagnose various types of faults. The conventional measurements for gearbox fault diagnosis mainly include lubricant analysis, vibration, airborne acoustics, thermal images, electrical signals, etc. However, a single domain measurement may lead to unreliable fault diagnosis and the contact installation of transducers is not always accessible, especially in harsh and dangerous environments. In this article, a Compressive Sensing (CS)-based Dual-Channel Convolutional Neural Network (CNN) method was proposed to accurately and intelligently diagnose common gearbox faults based on two complementary non-contact measurements (thermal images and acoustic signals) from a mobile phone. The raw acoustic signals were analysed by the Modulation Signal Bispectrum (MSB) to highlight the coupled modulation components relating to gear faults and suppress the irrelevant components and random noise, which generates a series of two-dimensional matrices as sparse MSB magnitude images. Then, CS was used to reduce the image redundancy but retain key information owing to the high sparsity of thermal images and acoustic MSB images, which significantly accelerates the CNN training speed. The experimental results convincingly demonstrate that the proposed CS-based Dual-Channel CNN method significantly improves the diagnostic accuracy (99.39% on average) of industrial helical gearbox faults compared to the single-channel ones.

Electromechanical systems as the prime power source are widely employed in industry. To ensure the high productivity and safety of the motor-gear system, motor current signature analysis (MCSA) becomes a cost-effective and effective approach to health condition monitoring of motors and gears simultaneously. In general, working conditions of the motor-gear system can be separated into stationary and nonstationary working conditions. The nonstationary working conditions usually refer to varying speeds, which have been broadly investigated in the angular domain analysis and time-frequency analysis. The stationary working conditions are assumed to have a constant rotating speed which is an ideal scenario, but practically, the rotating speed varies slightly with randomness. The random speed variation seems neglectable but it actually spreads the energy into adjacent frequency bins, which attenuates the amplitude of fault signatures and leads to inaccurate fault diagnosis. To address this issue, a squeezed modulation signal bispectrum (MSB) approach is developed to concentrate the leaked energy for accurately diagnosing gear faults with motor current signals. The squeezed MSB concentrates the energy in the frequency domain along the time axis to overcome the random speed oscillation induced energy leakage and then demodulates and aligns the modulation fault signatures from the squeezed spectra for ensemble averaging to further enhance fault signatures. The simulation study shows the performance of the proposed method under different levels of random speed variation, and the experimental studies demonstrate the effectiveness of the squeezed MSB for diagnosing gear tooth breakage faults under a wide range of working conditions.

This paper presents a novel method for diagnosing the gradual deterioration of gears using modulation signal bispectrum (MSB) and vibration measurements. A nonlinear model was derived to understand dynamic forces applied to gears that are excited by quadratic terms, e.g., shaft rotating speeds and gear meshing frequencies. Owing to its sensitivity to those quadratic terms, MSB is powerful in recovering less noisy condition related features from the measured vibration signals, e.g., gear meshing and multiples of shaft rotating speed. This allows a more pronounced representation of gear dynamic forces and makes it more effective for detecting early gear deterioration. The proposed method was verified through a run-to-failure test based on a helical gearbox system. The results show small gears at low-speed stages deteriorate faster and fail at 838 hours. This was because they prone to wear more severe due to poorer lubrication conditions compared with gears at high-speed stages. Moreover, fault detectability of the developed MSB-based method outperforms that of time synchronous averaging (TSA). Compared to TSA, clearer signs of early gear deterioration were captured using MSB, which makes it a more powerful tool for monitoring the condition of gearboxes.

The asymmetry in the manufacturing and assembling is the common issue of rotor systems. Different degrees of errors are inevitable in alternating current (AC) motors, which causes degraded performances. Furthermore, around 80 % of mechanical faults link to rotor eccentricity. The eccentricity faults (EFs) generate excessive mechanical stress and then lead to fatigue in the other parts of the motor. Motor current signal analysis (MCSA) can be used to diagnose induction machine (IM) faults. As the EF leads to an unequal air gap when the rotor rotates, the inductance of IM also responds to the EF. Moreover, the dynamic load is a typical situation due to residual dynamic unbalance and misalignment. To study how EFs and dynamic load affect the stator current. The current model of symmetrical motor, asymmetrical motors with three-level EFs and with dynamic load are investigated numerically. The correctness of models is verified through experimental study. The results show the level of EF affects the sideband peak values significantly in the stator current spectrum. These findings will provide a foundation for the accurate diagnosis of motor health conditions.