Fault diagnosis systems for rotating machines operating with fluid-film bearings

The paper deals with the application of deep learning methods to rotating machines fault diagnosis. The main challenge is to design a fault diagnosis system connected with multisensory measurement system that will be sensitive and accurate enough in detecting weak changes in rotating machines. The experimental part of the research presents the test rig and results of high-speed multisensory measurements. Six states of a rotating machine, including a normal one and five states with loosened mounting bolts and small unbalancing of the shaft, are under study. The application of deep network architectures including multilayer perceptron, convolutional neural networks, residual networks, autoencoders and their combination was estimated. The deep learning methods allowed to identify the most informative sensors, then solve the anomaly detection and the multiclass classification problems. An autoencoder based on ResNet architecture demonstrated the best result in anomaly detection. The accuracy of the proposed network is up to 100% while the accuracy of an expert is up to 65%. A one-dimensional convolutional neural network combined with a multilayer perceptron that contains a pretrained encoder demonstrated the best result in multiclass classification. The detailed fault detection accuracy with the determination of the specific fault is 83.3%. The combinations of known deep network architectures and application of the proposed approach of pretraining of the encoders together with using a block of inputs for one prediction demonstrated high efficiency.

Use of the Correlation Coefficient for Rotating Machine Monitoring

The use of electronics and computer technology in production systems has greatly improved the quality of our industrial products. The productivity of these installations is a function of the maintenance quality applied to the equipment. Several methods are used to monitor the functioning of industrial installations. One of these methods is vibration analysis. The vibration signals from the rotating machines support several types of information related to the working state of the production tool. The processing of this information makes it possible to have decision tools for maintenance. In this work, we propose a method of anticipating the maintenance of rotating machines. The algorithm we propose starts with the removal of 512 point windows during the running time of the ball bearing. Each signal is decomposed by DWT: we obtain the approximation coefficients. These coefficients make it possible to determine the correlation coefficient between the so-called reference window and the other windows following the functioning of the ball bearing. The correlation coefficient is then the fundamental element of the decision. This algorithm has been applied to real vibration data and the results are encouraging.

The Band Gap Formation in Rotors with Longitudinal Periodicity and Quasi-Periodicity

Modal spacing (band gaps) in the frequency spectrum of rotating machines can be imposed by geometric periodicity. By designing the rotor with a geometry that repeats periodically, we can impose to the vibration response of the rotor a modal "gap" considerably large, where no resonances appear. In this work, we consider that the rotating elements of the machine (e.g. the stages or the impellers) are the periodic elements of the rotor. In this disk-like configuration of the rotor, the system can present band gaps due to two different reasons: due to matching between the number of disks and the eigenmode wavenumber (usually in slender rotors); due to the presence of local-mode shapes (usually in large rotors). Analytical modeling of the system is presented, whose approximated solution can be used to predict the start and stop frequencies of the band gaps. It is also shown the limitations in band gap formation when the rotor is not perfectly periodic (quasi-periodic geometry). In this case, disk positioning plays an important role in the band gap formation.

Torsional vibrations in a shaft train of an air compressor: modification, calculation and measurement

Purpose Due to the installation of the new, more powerful gearbox and the dismantling of the exciter machine, the vibration characteristics of the shaft train will be changed. Therefore, it is necessary to reassess the shaft train. It is to be investigated if the shaft train of the compressor meets the general requirements for bending and torsional vibrations and can be safely operated within the electrical network. The purpose of this paper is to show the necessary alignment of modification, calculation and measurement in such a project.. Design/methodology/approach After some modification work on the shaft train of an air compressor, it was necessary to do some engineering calculations regarding the bending natural and torsional natural frequencies and their mode shapes. The correctness of the calculated values was proven by vibration measurements performed at the shaft train in operation. Findings It can be concluded that the change and replacement of rotating equipment in a shaft train never should be done without any engineering calculations in advance and measurements after the component modification. Most important is that the calculation results have to be compared with the measurement results for verifying the calculation assumptions. In the case described above, one can see that theory and practice match well. In addition to this, the very low damping of torsional vibrations is proved again, which can be a significant problem in some situations. Originality/value Also, today one can find torsional vibration measurements of rotating machines, including frequency, magnitude and damping factor, very seldom. Especially for smaller machines, there are no real comparisons between calculation and measurement are usual. This paper shows that an alignment between theoretical and practical approaches is necessary to avoid operational problems for rotating machines.

Rotating Machinery Diagnosing in Non-Stationary Conditions with Empirical Mode Decomposition-Based Wavelet Leaders Multifractal Spectra

Diagnosing the condition of rotating machines by non-invasive methods is based on the analysis of dynamic signals from sensors mounted on the machine—such as vibration, velocity, or acceleration sensors; torque meters; force sensors; pressure sensors; etc. The article presents a new method combining the empirical mode decomposition algorithm with wavelet leader multifractal formalism applied to diagnosing damages of rotating machines in non-stationary conditions. The development of damage causes an increase in the level of multifractality of the signal. The multifractal spectrum obtained as a result of the algorithm changes its shape. Diagnosis is based on the classification of the features of this spectrum. The method is effective in relation to faults causing impulse responses in the dynamic signal registered by the sensors. The method has been illustrated with examples of vibration signals of rotating machines recorded on a laboratory stand, as well as on real objects.

Comparative Study of Electric Machines for Stirling Generator Application

The choice of a machine for an application and a given specification remains a complex problem. This will involve, for example, bringing together criteria such as: performance, space saving, economical, reliable, little acoustic noise and others. The best machine selection to fulfill all constraints is an important step for the project to be realized. This work focus on Stirling Engine based Generator and study all types of rotating machines that can be employed for maximum electric power production. Analytical electromagnetic models where developed for all types of rotating machines that satisfied minimum requirement for the project by solving Maxwell equations. The purpose is to develop the design model and combine electromagnetic and thermal study of the machines. Finite Element Method is used to compare the performances of the generators for the best choice. Results show that for applications not requiring bigger output power, the major criteria for the selection is the optimal magnetic induction created by the inducer in the stationary part of the machine. For application such as Stirling generators, permanent magnet (PM) machine satisfy many comparison criteria such as maximum power at low speed, torque density, high efficiency. Beyond exposing a selection method for a project, this work lay down a step-by-step method for engineers and scientists for the crucial stage of design and conception work