The dual-rotor structure is susceptible to vibration, causing the malfunction of the entire operating system. In order to avoid the resonance during operation, it is significant and necessary to conduct modal analysis of such rotors. The dynamic analysis of the full dual-rotor system under operating conditions is also necessary to study dynamic characteristics of the rotating system. In this paper, one-dimension Timoshenko beam-type model, including the effects of gyroscopic moments, rotary inertias, bending and shear deformations, and three-dimension model for dual-rotor system with inter-shaft bearing are developed. Critical speed tests of dual-rotor are carried out to verify the analytical results. Based on the finite element models, the first critical speed excited by inner rotor and the first two critical speeds excited by outer rotor are calculated. The comparisons between both finite element models indicate that 1D model costs less time, which can be used to predict the critical speeds. Good agreement between the theoretical and experimental results shows the accuracy of the FE models. The Campbell diagram, critical speeds, operational deflection shapes and unbalance response of the dual-rotor are obtained to fully study the dynamic characteristics of the dual-rotor system.

Crack failures continually occur in shafts of turbine generator, where grid disturbance is an important cause. To estimate influences of grid disturbance, coupled torsional vibration and fatigue damage of turbine generator shafts are analyzed in this work, with a case study in a 600MW steam unit in China. The analysis is the following: (i) coupled system is established with generator model and finite element method (FEM)-based shafts model, where the grid disturbance is signified by fluctuation of generator outputs and the shafts model is formed with lumped mass model (LMM) and continuous mass model (CMM), respectively; (ii) fatigue damage is evaluated in the weak location of the shafts through local torque response computation, stress calculation, and fatigue accumulation; and (iii) failure-prevention approach is formed by solving the inverse problem in fatigue evaluation. The results indicate that the proposed scheme with continuous mass model can acquire more detailed and accurate local responses throughout the shafts compared with the scheme without coupled effects or the scheme using lumped mass model. Using the coupled torsional vibration scheme, fatigue damage caused by grid disturbance is evaluated and failure prevention rule is formed.

To diagnose rotating machinery fault for imbalanced data, a kind of method based on fast clustering algorithm and decision tree is proposed. Combined with wavelet packet decomposition and isometric mapping (Isomap), sensitive features of different faults can be obtained so the imbalanced fault sample set is constituted. Then the fast clustering algorithm is applied to search core samples from the majority data of the imbalanced fault sample set. Consequently, the balanced fault sample set consisted of the clustered data and the minority data is built. After that, decision tree is trained with the balanced fault sample set to get the fault diagnosis model. Finally, gearbox fault data set and rolling bearing fault data set are used to test the fault diagnosis model. The experiment results show that proposed fault diagnosis model could accurately diagnose the rotating machinery fault for imbalanced data.

Due to the complex conditions in rotating machinery, the blades usually encounter fluctuating loads, and failures may occur in the operation. Blade vibration is rarely monitored in the practical rotating machinery in power plants as the traditional blade vibration method using strain gauge method is hard to be applied in long-time and online monitoring. The blade tip timing method is a promising approach to monitor the blade vibration and several special sensors are developed to realize the online monitoring. Based on the traditional blade tip timing method, an improved approach is established in this work. The proposed method can modify the expected arriving time when the torsional vibration exists. The accuracy is therefore improved in the torsional vibration case. Then the blade vibration monitoring with the traditional blade tip timing method and the proposed method are carried out in the test rig in laboratory. The results validate that, the torsional vibration influences the expected arriving time, and the proposed method can modify the expected arriving time and improve the accuracy of the blade vibration monitoring.

This paper proposes an improved rigid multibody model for the dynamic analysis of the planetary gearbox in a wind turbine. The improvements mainly include choosing the inertia frame as the reference frame of the carrier, the ring, and the sun and adding a new degree of freedom for each planet. An element assembly method is introduced to build the model, and a time-varying mesh stiffness model is presented. A planetary gear study case is employed to verify the validity of the improved model. Comparisons between the improvement model and the traditional model show that the natural characteristics are very close; the improved model can obtain the right equivalent moment of inertia of the planetary gear in the transient simulation, and all the rotation speeds satisfy the transmission relationships well; harmonic resonance and resonance modulation phenomena can be found in their vibration signals. The improved model is applied in a multistage gearbox dynamics analysis to reveal the prospects of the model. Modal analysis and transient analysis with and without time-varying mesh stiffness considered are conducted. The rotation speeds from the transient analysis are consistent with the theory, and resonance modulation can be found in the vibration signals.

Root causes of several recent crack failures in turbine units are attributed to oscillation and interaction between generator of turbine unit and devices on the grid, where torsional vibration of the rotor bearing system is observed and identified as an important cause. Exploring vibrational (lateral, torsional, and axial) features in the cracked rotor system with torsional excitation (TE) present can provide a novel view in crack detection and isolation. This work presents dynamic analysis of a cracked rotor system in a steam turbine unit (a typical rotor system with multiple rotors, multiple supports, and oscillating loads), and the vibrational features of the cracked rotor system with comparisons to typical features in monitored vibration data. The results show that coupled vibration in both lateral and torsional components is an effective indicator for cracks in the presence of torsional excitation. Also, vibration characteristics evaluated in different locations of the rotor system are beneficial for fault detection and isolation.