Search

Estimating the remaining useful life (RUL) of components is a crucial task to enhance reliability, safety, productivity, and to reduce maintenance cost. In general, predicting the RUL of a component includes constructing a health indicator (HI) to infer the current condition of the component, and modelling the degradation process in order to estimate the future behavior. Although many signal processing and data-driven methods have been proposed to construct the HI, most of the existing methods are based on manual feature extraction techniques and require the prior knowledge of experts, or rely on a large amount of failure data. Therefore, in this study, a new data-driven method based on the convolutional autoencoder (CAE) is presented to construct the HI. For this purpose, the continuous wavelet transform (CWT) technique was used to convert the raw acquired vibrational signals into a two-dimensional image; then, the CAE model was trained by the healthy operation dataset. Finally, the Mahalanobis distance (MD) between the healthy and failure stages was measured as the HI. The proposed method was tested on a benchmark bearing dataset and compared with several other traditional HI construction models. Experimental results indicate that the constructed HI exhibited a monotonically increasing degradation trend and had good performance in terms of detecting incipient faults.

Low frequency vibration monitoring has significant implications on environmental safety and engineering practices. Vibration expressed by visual information should contain sufficient spatial information. RGB-D camera could record diverse spatial information of vibration in frame images. Deep learning can adaptively transform frame images into deep abstract features through nonlinear mapping, which is an effective method to improve the intelligence of vibration monitoring. In this paper, a multi-modal low frequency visual vibration monitoring system based on Kinect v2 and 3DCNN-ConvLSTM is proposed. Microsoft Kinect v2 collects RGB and depth video information of vibrating objects in unstable ambient light. The 3DCNN-ConvLSTM architecture can effectively learn the spatial-temporal characteristics of muti-frequency vibration. The short-term spatiotemporal feature of the collected vibration information is learned through 3D convolution networks and the long-term spatiotemporal feature is learned through convolutional LSTM. Multi-modal fusion of RGB and depth mode is used to further improve the monitoring accuracy to 93% in the low frequency vibration range of 0–10 Hz. The results show that the system can monitor low frequency vibration and meet the basic measurement requirements.

Electromagnetic dampers are emerging as alternatives to conventional dampers applied to stay cables of bridges because they can reduce maintenance costs and allow vibration monitoring owing to their permanent driving characteristics and self-generation function. In this study, the main equations (including those for the induced electromotive force of the active coils and the total damping force of the damper) were derived through magnetic circuit analysis using the main parameters of the electromagnetic damper model. Characteristic tests were performed on electromagnetic damper prototypes to analyze the hysteretic dynamics and derive characteristics according to their structure and excitation conditions. On the basis of the results, we proposed a regenerative hybrid electrodynamic damper with an oxygen-free copper tube and teeth structure. Its physical and electromagnetic behaviors were examined through an electromagnetic analysis of the finite element model of the proposed damper. The results confirmed that attenuation occurred via strengthened magnetic flows, and the estimated power production is suitable for energy harvesting applications. Therefore, we confirmed the feasibility of constructing a system that can simultaneously perform cable attenuation and vibration monitoring using the proposed damper.

In this paper, the problem of optimal sensor arrangement during vibration monitoring is analysed. The wave propagation caused by horizontal excitation is investigated to predict the areas of the largest ground and structure response. The equations of motion for a transversally isotropic elastic medium with appropriate absorbing boundary conditions are solved using the finite element method (FlexPDE software). The possibility of an amplified soil medium response is examined for points located on the ground surface and at various depths. The results are presented in the form of a dimensionless vibration reduction factor, defined as the ratio of the peak particle velocity observed at the selected depth to the corresponding value observed at the ground surface. Significant amplifications (≈50%) can be observed below the ground surface, especially in the case of a weak layer below a stiff layer. The effect of vibration amplification is most significant near the boundary surface of two layers. For the points located on the ground surface, the greatest peak particle velocities are observed in the direction perpendicular to the load direction. However, the greatest vertical velocity component at the ground surface is observed in front of the applied force.

Following a small-scale wedge failure at Yukon Zinc's Wolverine Mine in Yukon, Canada, a vibration monitoring program was added to the existing rockbolt pull testing regime. The failure in the 1150 drift occurred after numerous successive blasts in an adjacent tunnel had loosened friction bolts passing through an unmapped fault. Analysis of blasting vibration revealed that support integrity is not compromised unless there is a geological structure to act as a failure plane. The peak particle velocity (PPV) rarely exceeded 250 mm/s with a frequency larger than 50 Hz. As expected, blasting more competent rock resulted in higher PPVs. In such cases, reducing the round length from 3.5 m to 2.0 m was an effective means of limiting potential rock mass and support damage.

The application of neutron noise analysis (NNA) to the ex-core neutron detector signal for monitoring the vibration characteristics of a reactor core support barrel (CSB) was investigated. Ex-core flux data were generated by using a nonanalog Monte Carlo neutron transport method in a simulated CSB model where the implicit capture and Russian roulette technique were utilized. First and third order beam and shell modes of CSB vibration were modeled based on parallel processing simulation. A NNA module was developed to analyze the ex-core flux data based on its time variation, normalized power spectral density, normalized cross-power spectral density, coherence, and phase differences. The data were then analyzed with a fuzzy logic module to determine the vibration characteristics. The ex-core neutron signal fluctuation was directly proportional to the CSB's vibration observed at 8 Hz and 15 Hz in the beam mode vibration, and at 8 Hz in the shell mode vibration. The coherence result between flux pairs was unity at the vibration peak frequencies. A distinct pattern of phase differences was observed for each of the vibration models. The developed fuzzy logic module demonstrated successful recognition of the vibration frequencies, modes, orders, directions, and phase differences within 0.4 ms for the beam and shell mode vibrations.

The radial and axial vibration signals of a rotating machine are crucial information to understand the machine operation and to diagnose potential faults. Instead of using three single sensors to measure the horizontal, vertical and axial displacements of rotating shafts, a novel non-projection vision-based system was proposed to realize simultaneous measurement of the radial and axial displacements with high accuracy and good reliability using an tailored artificial constant density sinusoidal fringe pattern (CDSFP), which was pasted around the shaft surface and worked as a tri-axial (i.e. the horizontal, vertical and axial) displacement sensor. The measurement principle and setup of the proposed measurement system were well established. The horizontal displacement could be correctly obtained from the fringe period density changes of the CDSFP image sequence recorded by a high-speed camera. Simultaneously, the vertical displacement could be acquired by tracking the centerline of shaft whilst the axial displacement could be obtained by locating the peaks of the cross-correlation sequence of the fringe intensities. A sub-pixel method was employed to improve the displacement resolution of the developed system. The performance of the proposed system was demonstrated by the comparison of the experiments using eddy current sensors. It showed that the proposed method was an effective and accurate technique for real-time tri-axial vibration monitoring of rotating shaft. Experimental results verified the feasibility, effectiveness and good robustness of the proposed methodology, which demonstrated that the proposed system was capable of achieving accurate tri-axial vibration displacements of rotating shaft compared to the commercial eddy current sensor which could only measure one dimensional displacement at each measurement. Therefore, the vision-based tri-axial vibration monitoring system could be recommended for real engineering applications in condition monitoring of rotating shafts. [ABSTRACT FROM AUTHOR]

Abstract A critical issue for structural health monitoring (SHM) strategies based on pattern recognition models is a lack of diagnostic labels to explain the measured data. In an engineering context, these descriptive labels are costly to obtain, and as a result, conventional supervised learning is not feasible. Active learning tools look to solve this issue by selecting a limited number of the most informative observations to query for labels. This work presents the application of cluster-adaptive active learning to measured data from aircraft experiments. These tests successfully illustrate the advantages of utilising active learning tools for SHM, and they present the first application/adaptation of active learning methods to engineering data — a MATLAB package is available via GitHub: https://github.com/labull/cluster_based_active_learning. [ABSTRACT FROM AUTHOR]

Highlights • Application of a novel digital video processing method to structural monitoring. • Viable and effective method by fast setup data acquisition and low-cost equipment. • Detection of weak points in structures through magnified motion anomalies. • Capability of providing approximate dynamic identification of low-frequency modes. Abstract A new methodology for digital image processing, namely the Motion Magnification (MM), allows to magnify small displacements of large structures. MM acts like a microscope for motion in video sequences, but affecting only some groups of pixels. The processed videos unveil motions hardly visible with the naked eye and allow a more effective frequency domain analysis. We applied the MM method to several historic structures, including a 1:10-scale mockup of Hagia Irene in Constantinople tested on shaking table, the so-called Temple of Minerva Medica in Rome and the Ponte delle Torri of Spoleto. MM algorithms parameters were calibrated by comparison with reference consolidated modal identification methods applied to conventional velocimeters data. Encouraging results were obtained in terms of vibration monitoring and modal analysis for dynamic identification of the studied structures, offering a low-cost, viable support to the standard vibration sensing equipment, such as contact velocimeters, laser vibrometers and others. [ABSTRACT FROM AUTHOR]

The signals emanating from the bearings are complex and contribute to various distributions. The effect of the distribution and mathematical operations are responsible for the change in the statistical moments. This paper investigates the effect of noise on statistical moments of the bearing vibration signals. Initially, the distribution function for Healthy, inner race defect (IRD), outer race defect (ORD), and ball defect (BD) are tested using Kolmogorov Smirnov test (K-S test). The resulting distributions obtained from the K-S test are normal and Laplacian distributed patterns and convey the faulty state of the bearings. The change in noise levels and their influence on the statistical moments are verified. It is observed, the kurtosis for IRD and ORD decreases with increase in noise, whereas, the trend increases for healthy and BD faults. [ABSTRACT FROM AUTHOR]

Copyright of Transactions of Beijing Institute of Technology is the property of Beijing University of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The objective of this article is to improve bearing monitoring via vibration indicators, by developing combinations of indicators, which is more effective than using isolated classical indicators. We use a new perceptive approach to seek correlations between vibration analysis and the perception of noise generated by the defects of rotating machines. Therefore, an experimental bench was developed to simulate machine defects (bearings, alignments and gears) and to perform vibration and acoustic measurements. This study was split into three stages: the first was to process the signals of simulated defects using different vibration analysis techniques (scalar indicators and spectral analysis). The second phase was devoted to a perception test of the noises generated by the simulated defects; this test was carried out on different auditors. The results obtained from the perception test were then processed using multidimensional analysis (MDS), to calculate the dimensions of the sound perception space. The final stage of this study consisted in searching the dimensions correctly representing the deterioration of the simulated defects, and in correlating these subjective dimensions with objective classical indicators (RMS, Kurtosis, Crest factor, CGS, etc.), to define more sensitive indicators calculated from combinations of classical vibration indicators. [ABSTRACT FROM AUTHOR]

Highlights • Vibration monitoring is applied for speed estimation of rotating machinery. • Vibration spectrum with sample of short duration produces large error in speed estimation. • Speed estimation using harmonic component improve its accuracy. However, higher harmonics component are found to be absent in many cases. • An algorithm is proposed to generate artificially higher order harmonic components of the fundamental rotational frequency, for speed estimation. • Accurate speed estimation using the proposed algorithm reduces error in estimating speed by 16 times. Abstract The use of vibration sensors for rotational speed estimation has been found to be an effective and reliable technique. In general, the fundamental rotational frequency (f) component observed in the vibration spectrum represents the rotational speed. However, for a better accuracy of estimation, large sample duration is required which results in a higher data acquisition time, computational time and data storage. Hence, the estimation of speed accuracy with the vibration signal of a short duration is a challenging task. This paper proposes an algorithm, which requires extraction of the fundamental rotational frequency component and later, the artificial generation of its corresponding harmonic components, for the accurate estimation of speed. The proposed algorithm can be used to estimate the speed of any rotating machinery. [ABSTRACT FROM AUTHOR]

Submersible well pumps, widely used for groundwater, are installed at great depths which make typical maintenance actions unworkable and costly. Condition-based maintenance (CBM) through vibration monitoring could be an effective approach for these critical machines. According to ISO 10816-7 (2009) suggestions, a procedure to measure vibration was established and applied on submersible well pumps both in a test facility and in situ. Instruments and measurement location were selected and a methodology for data processing was applied in order to set maintenance alarms. The procedure was tested on several submersible well pumps directly in the testing room of one of the main producers of pumps. Measurements in a testing room made it possible to avoid problems and costs related to instrument installation in the field. In order to demonstrate the technical feasibility of the procedure also in field, the method was also applied in real groundwater plant with a special design of the measurement system, to overcome strict operational work conditions. This paper shows the experimental campaign results, which are the basis to conduct CBM on pumps, in terms of: (1) definition of the typical vibration values of new pumps, not available even by producers, plant managers and standards, and (2) identification of the parameters which influence pump vibration, fundamental to data processing and the setting of proper alarms. The procedure will also be applied for other categories of pumps, used in several industrial applications, to measure reliable vibration data and set specific alarms for maintenance interventions. [ABSTRACT FROM AUTHOR]

Standard filtering techniques for structural parameter estimation assume that the input force is either known or can be replicated using a known white Gaussian model. Unfortunately for structures subjected to seismic excitation, the input time history is unknown and also no previously known representative model is available. This invalidates the aforementioned idealization. To identify seismic induced damage in such structures using filtering techniques, force must therefore also be estimated. In this paper, the input force is considered to be an additional state that is estimated in parallel to the structural parameters. Two concurrent filters are employed for parameters and force respectively. For the parameters, an interacting Particle-Kalman filter is used to target systems with correlated noise. Alongside this, a second filter is used to estimate the seismic force acting on the structure. In the proposed algorithm, the parameters and the inputs are estimated as being conditional on each other, thus ensuring stability in the estimation. The proposed algorithm is numerically validated on a sixteen degrees-of-freedom mass-spring-damper system and a five-story building structure. The stability of the proposed filter is also tested by subjecting it to a sufficiently long measurement time history. The estimation results confirm the applicability of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Road quality is an important issue in everyday life for all car owners. This issue seems to be critically important in underground mines, where LHD machines are used for material transport. One of the biggest problems for LHD operation is relatively quick tires degradation. One of possible reasons might be road surface quality, indeed. However, driver's skills as well as ways of machine operation (loading, acceleration, breaking...) might also play a crucial role. Nowadays, many of machines are equipped with onboard monitoring system that allows to monitor basic parameters (speed, torque, temperatures, pressures etc.) at some predefined components. To complete the picture, we propose to use proposed already (but not for mining applications) vibration measurement for road roughness evaluation. To measure vibration acceleration is relatively easy task (we used simple smartphone here), unfortunately method of parametrization and concluding about road quality is still a challenge in mining case. In this paper we have presented a short communication related to first experimental work and some ideas how to deal with this problem using statistical tools for signal modeling. [ABSTRACT FROM AUTHOR]

In this study, a practical vibration analysis approach to obtain the dynamic characteristics of an in-situ caisson breakwater under the ambient wave load is presented. Firstly, a scheme of vibration monitoring and modal analysis is designed for the caisson breakwater. The limits of vibration monitoring on the caisson system is examined based on sensor placement and excitation source. An output-only modal analysis approach which combines the time-domain stochastic subspace identification method and the frequency domain decomposition method is designed to estimate modal parameters from wave-induced ambient vibration signals. Secondly, in-situ tests on a real caisson breakwater are described. Acceleration signals recorded under ambient wave-induced excitation are experimentally analyzed for a few caisson units under sea-level variation. Finally, experimental modal analyses are performed on the measured acceleration signals to evaluate the practicality of the proposed scheme. The numerical modal analysis of the caisson-foundation system was also conducted to support the experimental mode identification result. Relative modal responses of three adjacent caisson units are analyzed to estimate relative differences in dynamic characteristics due to their structural conditions. Also, the effect of water-level variation on modal parameters such as natural frequencies, modal damping and mode shapes is analyzed from the in-situ vibration records of a single caisson unit. [ABSTRACT FROM AUTHOR]

Summary: Large‐scale, long‐term structural health monitoring systems have become more feasible in recent years as the required data acquisition and analysis systems are more affordable to deploy. These long‐term systems must process and store vast amounts of data without wasting computational power and storage capacity with redundant or poor quality data. While not a primary system for damage detection, large‐scale, long‐term vibration monitoring systems aim to leverage changes in the dynamic signature of a structure to assess global structural changes. Although the ability to continually collect vibration data at high rates exists, it is not always feasible to store all these data long term. As more long‐term monitoring systems are deployed, efficient methods need to be developed to quickly and efficiently analyze large quantities of vibration data so that only the most pertinent information is archived. Previous researchers have used scheduled approaches, eg, taking data every hour, or triggered sensing systems. A monitoring system on the I‐35W St. Anthony Falls Bridge, which crosses the Mississippi River in Minneapolis, Minnesota, has been collecting vibration and temperature data since the structure's opening in 2008. This provides a uniquely large data set to establish the characteristics of a good signal for output‐only system identification to consistently and efficiently capture natural frequencies and mode shapes. To this end, a system identification routine using a novel signal selection approach and modal sorting routine that leverages NExT‐ERA/DC is proposed to analyze this large data set. The resulting information allows long‐term and temperature‐based trends to be identified. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Machine Construction & Maintenance is the property of Institute for Sustainable Technologies - National Research Institute and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The blasting for bedrock excavation on land for open channel project has a great influence on lock gate in Taishan Nuclear Power Station, therefore, based on blasting vibration monitoring data, the attenuation law of blasting vibration signal has been studied through regression analysis of practical test data by Sadaovsk empirical formula and corresponding time-frequency characteristics was analyzed by Empirical Mode Decomposition based on Hilbert-Huang transform. As for those monitoring data, the results of blast vibration velocity for vertical direction are generally larger than horizontal radial and horizontal tangential direction in the near field of blasting source and the peak particle velocity of vertical direction is usually lower than horizontal radial and horizontal tangential direction in the far field of blasting source; at the same time, their main vibration frequency mostly vary from 10 Hz to 80 Hz which is much higher than natural frequency of lock gate and is beneficial to structural safety and stability of surrounding rock mass for reducing the probability of resonance. To ensure the safety of lock gate, it is of great significance to control maximum explosive weight per delay in advance for different distance from monitoring point to the explosion source according to Safety Regulations for Blasting (GB6722-2014), which shows the excellent effect on blasting damage control of the lock gate and surrounding rock mass. The results from the analysis can be for reference to similar blasting design and blasting construction. [ABSTRACT FROM AUTHOR]

In rotating machines, rolling element bearings are widely used in many applications, such as marine gear box and diesel engine. Vibration monitoring is highly capable to detect any abnormalities on bearings before failure occurs. In this paper, vibration measurement is utilised to detect the locations on rotor bearing systems that record the highest levels of vibration. Vibration spectrums captured on two units of healthy rolling element bearings in three different directions at various speeds were analysed, with the measurements found to have high level of accuracy. Based on the measurement procedure developed in this experimental work, condition monitoring (CM) for rotor-bearing systems can be developed in the future. [ABSTRACT FROM AUTHOR]

The dynamic behavior at part load has been a major problem for low head and medium head Francis turbines. Vortex rope (draft tube swirl), caused due to the flow instabilities in the draft tube under off-design part load operating conditions, create low frequency pressure pulsations at a frequency of 0.2 to 0.4 times the rotational frequency of the runner. High amplitude vibrations due to resonance can occur in case these pulsations coincide with the natural frequency of a hydraulic, mechanical, or structural component. This study presents the results of the vibration analysis done using the online vibration monitoring & diagnosis system installed on Francis turbines in a hydroelectric power plant which involve high overall vibrations exceeding the alarm limits at partial loads. Orbit and FFT spectrum plots are obtained from relative shaft vibration measurements performed using eddy-current displacement sensors wired to a realtime multichannel vibration analyzer. Measurement results at the turbine guide bearing revealed the drft tube swirl that rotates at 27% of the rotational speed at certain partial loads. [ABSTRACT FROM AUTHOR]

Abstract: With the diverse development in electronic devices, high cost, low reliability, and the need for battery power have been severe obstacles to restrict the wide application of sensors. In this work, we present a highly reliable, real‐time, self‐powered vibration sensor based on an enhanced piezoelectric nanogenerator (NG), which is heat resistant, waterproof, relatively low cost, and with a long lifetime. The enhanced piezoelectric NG is constructed into a double‐arched structure: the upper arched layer is the poly(vinylidene fluoride) (PVDF) film with designed optimum size, which can utilize the excellent piezoelectric potential of the d31 mode; the lower arched layer is the polydimethylsiloxane (PDMS) film covered by uniform trapezoidal microstructures and thereby increases the charge density of the piezoelectric NG. The piezoelectric NG output voltage with the double‐arched structure increased by 35.1 % upon applying a mechanical force of 5 N. Relating to improved steady voltage output performance, the enhanced piezoelectric NG was used as the sensitive element of a sensor with sensitivity of 2.21 V g−1 and linearity error of 5.91 % under a vibration amplitude of 6 mm. Moreover, the sensor not only exhibited stable repeatability and stability over long times, but it also steadily produced output performances after soaking in water temperatures up to 70 °C. Therefore, the highly reliable self‐powered vibration sensor presents prominent prospect for future application. [ABSTRACT FROM AUTHOR]

Centrifugal pumps are widely used in the industry, especially in the oil and gas sector for fluids transport. Classically, these are designed to transfer single phase fluids (e.g., water) at high flow rates and relatively low pressures when compared with other pump types. As part of their constructive feature, centrifugal pumps rely on seals to prevent air entrapment into the rotor during its normal operation. Although this is a constructive feature, water should pass through the pump inlet even when the inlet manifold is damaged. Modern pumps are integrated in pumping units which consist of a drive (normally electric motor), a transmission (when needed), an electronic package (for monitoring and control), and the pump itself. The unit also has intake and outlet manifolds equipped with valves. Modern systems also include electronic components to measure and monitor pump working parameters such as pressure, temperature, etc. Equipment monitoring devices (vibration sensors, microphones) are installed on modern pumping units to help users evaluate the state of the machinery and detect deviations from the normal working condition. This paper addresses the influence of air-water two-phase mixture on the characteristic curve of a centrifugal pump; pump vibration in operation at various flow rates under these conditions; the possibilities of using the results of experimental investigations in the numerical simulations for design and training purposes, and the possibility of using vibration and sound analysis to detect changes in the equipment working condition. Conclusions show that vibration analysis provides accurate information about the pump's functional state and the pumping process. Moreover, the acoustic emission also enables the evaluation of the pump status, but needs further improvements to better capture and isolate the usable sounds from the environment. [ABSTRACT FROM AUTHOR]

Non-contact displacement measurement based on the phase-based microwave interferometry technology has attracted growing interest in recent years. In this paper, aiming to exploit the frequency-modulated continuous wave (FMCW) radar-based vibration monitoring technique, the principle and detailed procedure for extracting the displacement time series of vibration movement are illustrated. The approximate maximum likelihood approach is proposed to extract the phase history of beat signals across multiple sweep periods. Moreover, the measurement accuracy of vibration displacement extraction and several practical issues of this technique are evaluated and concerned. Both simulated and experimental validations are provided to show the effectiveness and robustness of the proposed method and approach. [ABSTRACT FROM PUBLISHER]

A WSNs model of vibration monitoring of mine equipment was given combined with actual situation of vibration monitoring of mine equipment. In order to solve problems of large energy consumption of TPSN algorithm and low precision of DMTS algorithm, an improved WSNs time synchronization algorithm named TPDM algorithm was proposed. In TPDM algorithm, cluster head nodes are selected by use of dynamic selection algorithm of cluster head. TPSN algorithm is used for time synchronization among the cluster head nodes to improve synchronization precision and DMTS algorithm is used among nodes in a cluster to reduce energy consumption. Meanwhile, synchronous clock is compensated by use of clock skew compensation algorithm based on the minimum square linear regression. The simulation results show that TPDM algorithm gets a compromise between precision and energy consumption compared with TPSN algorithm and DMTS algorithm, which can meet with requirements of vibration monitoring of mine equipment preferably.

The prediction of electrical machines’ Remaining Useful Life (RUL) can facilitate making electrical machine maintenance policies, which is important for improving their security and extending their life span. This paper proposes an RUL prediction model with similarity fusion of multi-parameter and multi-sample. Firstly, based on the time domain and frequency domain extraction of vibration signals, the performance damage indicator system of a gearbox is established to select the optimal damage indicators for RUL prediction. Low-pass filtering based on approximate entropy variance (Aev) is introduced in this process because of its stability. Secondly, this paper constructs Dynamic Time Warping Distance (DTWD) as a similarity measurement function, which belongs to the nonlinear dynamic programming algorithm. It performed better than the traditional Euclidean distance. Thirdly, based on DTWD, similarity fusion of multi-parameter and multi-sample methods is proposed here to achieve RUL prediction. Next, the performance evaluation indicator Q is adopted to evaluate the RUL prediction accuracy of different methods. Finally, the proposed method is verified by experiments, and the Multivariable Support Vector Machine (MSVM) and Principal Component Analysis (PCA) are introduced for comparative studies. The results show that the Mean Absolute Percentage Error (MAPE) of the similarity fusion of multi-parameter and multi-sample methods proposed here is below 14%, which is lower than MSVM’s and PCA’s. Additionally, the RUL prediction based on the DTWD function in multi-sample similarity fusion exhibits the best accuracy.

Nowadays, the subject of machine diagnostics is gathering growing interest in the research field as switching from a programmed to a preventive maintenance regime based on the real health conditions (i.e., condition-based maintenance) can lead to great advantages both in terms of safety and costs. Nondestructive tests monitoring the state of health are fundamental for this purpose. An effective form of condition monitoring is that based on vibration (vibration monitoring), which exploits inexpensive accelerometers to perform machine diagnostics. In this work, statistics and hypothesis testing will be used to build a solid foundation for damage detection by recognition of patterns in a multivariate dataset which collects simple time features extracted from accelerometric measurements. In this regard, data from high-speed aeronautical bearings were analyzed. These were acquired on a test rig built by the Dynamic and Identification Research Group (DIRG) of the Department of Mechanical and Aerospace Engineering at Politecnico di Torino. The proposed strategy was to reduce the multivariate dataset to a single index which the health conditions can be determined. This dimensionality reduction was initially performed using Principal Component Analysis, which proved to be a lossy compression. Improvement was obtained via Fisher’s Linear Discriminant Analysis, which finds the direction with maximum distance between the damaged and healthy indices. This method is still ineffective in highlighting phenomena that develop in directions orthogonal to the discriminant. Finally, a lossless compression was achieved using the Mahalanobis distance-based Novelty Indices, which was also able to compensate for possible latent confounding factors. Further, considerations about the confidence, the sensitivity, the curse of dimensionality, and the minimum number of samples were also tackled for ensuring statistical significance. The results obtained here were very good not only in terms of reduced amounts of missed and false alarms, but also considering the speed of the algorithms, their simplicity, and the full independence from human interaction, which make them suitable for real time implementation and integration in condition-based maintenance (CBM) regimes.

Gearbox is an important equipment in an automobile to transfer power from the engine to the wheels with various speed ratios. The maintenance of the gearbox is a top criterion as it is prone to a number of failures like tooth breakage and bearing cracks. Techniques like vibration monitoring have been implemented for the fault diagnosis of the gearbox over the years. But, the experiments are usually conducted in lab environment where the actual conditions are simulated using setup consisting of an electric motor, dynamometer, etc. This work reports the feasibility of performing vibrational monitoring in real world conditions, i.e. by running the vehicle on road and performing the analysis. The data was acquired for the various conditions of the gearbox and features were extracted from the time-domain data and a decision tree was trained for the time-domain analysis. Fast Fourier Transform was performed to obtain the frequency domain which was divided into segments of equal size and the area covered by the data in each segment was calculated for every segment to train decision trees. The classification efficiencies of the decision trees were obtained and in an attempt to improve the classification efficiencies, the time-domain and frequency-domain analysis was also performed on the normalised time-domain data. From, the results obtained, it was found that performing time-domain analysis on normalised data had a higher efficiency when compared with the other methods. Instantaneous processing of the acquired data from the accelerometer enables faster diagnosis. Hence, online condition monitoring has gained importance with the advent of powerful microprocessors. A windows application that has been developed to automate the process was found to be essential and accurate. [ABSTRACT FROM AUTHOR]

Designed to break the paradigm for efficiency, the new generation of engines promises double-digit reductions in fuel burn, as well as an unparalleled single-leap improvement in emissions and lower noise to fulfil societal environmental objectives for a more sustainable future. The end-use consumer and environmental policy requirements for aircrafts of the next generation translate into components with higher temperature and speed. Furthermore, new instrumentation techniques are needed to closely monitor rolling contact during testing of the next generation of aero engine bearing to check its behavior under the new application condition. Vibration analysis for condition assessment and fault diagnostics is widely used nevertheless interpretation and correlation of collected data is often cumbersome. That is why combination of both techniques giving different types information in two different frequency band can help to understand the behavior of new gear box. This study proposes a correlation between low and high frequency signals with different strategy of signal acquisition and processing. Real time transient analysis with feature extraction can be done in parallel with streaming acquisition. Then pattern recognition of individual AE signal is possible and can be correlated with more traditional analysis based on "multiple chocks" vibration analysis. Continuous monitoring of an aging gear box is giving genuine information on no stationary regime and also time of stabilization. Long term experiments are conducted on damaged and defect free gear boxes at several rotating speed and loading level. [ABSTRACT FROM AUTHOR]

The performance of a few localization schemes using measured mode shapes was tested in an experimental case study with respect to the Ultimate Limit State (ULS). The first question to be posed was: Is it possible to indicate and locate damage under laboratory conditions before the ULS is reached? Relatively simple localization criteria were chosen, which do not require extensive FE analysis. A new combination of them, designated here as Combined Localization Criterion (CLI), was proposed that performed well in the presented case. A simple supported beam with the damage progressing in three consecutive stages was used for the experiments. Mode shapes in the range of up to 100 Hz were extracted from the response of the structure to the air stream. [ABSTRACT FROM AUTHOR]

Due to the increasing share of new renewable energies like wind and solar in the generation of electricity the need for power regulation and energy storage is becoming of paramount importance. One of the systems to store huge amounts of energy is pumped storage using reversible hydropower units. The machines used in these power plants are pump-turbines, which can operate as a pump and as a turbine. The surplus of electrical energy during low consumption hours can be converted into potential hydraulic energy by pumping water to a higher level. The stored energy can be converted into electricity again by operating the runner as a turbine. Due to new regulation requirements machines have to extend the operating range in order to match energy generation with consumption for the grid stability. In this paper the consequences of extending the operating range in existing pump-turbines have been studied. For that purpose, the data obtained after two years of condition monitoring were analyzed. Vibrations and pressure fluctuations of two pump-turbines of 85MWeach have been studied during pump and turbine operation. For turbine operation the effects of extending the operating range from the standard range of 45-85 MW to and increased range of 20-85 MW were analyzed. The change in vibration levels and signatures at very low load are presented with the identification of the phenomena that occur under these conditions. The influence of head in the vibration behavior is also presented. The appearance of fluid instabilities generated at part load that may produce power swing is also presented. Finally, the effect of head on the vibration levels for pump operation is shown and analyzed. [ABSTRACT FROM AUTHOR]

A piezoelectric generator is a device used to convert mechanical energy into electrical energy. The basic element of the generator is made from piezoelectric material in which electrical energy is created as a result of deformations caused by reactions of mechanical structure of the generator. The generators, in which the main element is a cantilever beam constructed from basic material and from piezoelectric material, are one of structures, which are used to the energy harvesting from the mechanical vibrations. The piezoelectric generator are used as a part of self-powered wireless sensor in many scientific research. Such self-powered sensor consists of a sensor measuring selected parameter, a radio transmitter, and aforementioned piezoelectric beam generator. However, the piezoelectric beam generator has a limited application in supply a sensor, because the amount of electric energy generated by such generator does not exceed a few mili Watts and is achieved only if the vibration frequency of mechanical energy source equals the resonant frequency of a cantilever beam of piezoelectric generator. This problem is mainly tried to solve by modifications of the generator beam structure or by introducing additional elements with geometry adapted to the source of mechanical vibration. In contrast to these approaches, the article presents the experimental investigation on the action of self-powered wireless sensor in which a decrease of consumed energy amount was achieved through an using of piezoelectric beam generator, which power the radio transmitter, as a vibration sensor. The piezoelectric beam generator should product an amount of electric energy which is enough to supply only the radio transmitter. Such self-powered wireless sensor is dedicated to a vibration monitoring, in which the basic purpose is a detection of an appearance of the dangerous for building structure vibration. The presented self-powered vibration sensor based on a piezoelectric beam generator equipped with a standard energy harvesting circuit, a radio transmitter and a system of energy transfer control enabling power of a radio transmitter for established level of stored energy in capacitor allows a detection of dangerous vibrations for building structure, monitored by this sensor. In laboratory stand vibration amplitudes of the free end of piezoelectric generator were obtained by changes of vibration frequency and vibration amplitudes of its fixed end. In laboratory experiments the time among radio signal was measured for selected values of capacitor capacity and for changed vibration amplitude of free end of piezoelectric generator beam. On the basis of this experiments it was found that full robustness of tested self-powered sensor was achieved for the capacitor capacity equal to 440 μF. The sending of a full package of data by such sensor can be realized only above a minimum value of the capacity of applied capacitor equals 220 μF. The capacity of capacitor in piezoelectric self-powered vibration sensor should be selected between minimum value for which full packages of data are sending, in tested sensor it equals 220 μF, and maximum values for which full robustness to disturbances is achieved. It was found that an increase of the capacitor capacity caused an improvement of data sending robustness to disturbances in radio transmission. The full robustness manifested the sending of every package of data. On the other hand, an increase of capacitor capacity led to the decrease of detection sensitivity of appearance of short-lived vibrations with the big amplitudes due to longer time of capacitor charging. It was noted that the increase of capacitor capacity did not improve the acting of such sensor in case of appearance of long-lived vibrations. [ABSTRACT FROM AUTHOR]

This study proposes a damage detection method for bridges using Bayesian hypothesis testing, aiming at efficient inspection based on vibration monitoring. In the proposed damage detection method, firstly a posterior distribution of the parameters composing multivariate auto-regressive model is acquired from a bridge under healthy condition by means of Bayesian inference. Secondly, based on the distribution representing vibration of the healthy bridge, a Bayesian hypothesis test is conducted to detect change in the modal properties caused by damage. To investigate feasibility of the proposed method for damage detection, this study utilized data from a field experiment on an actual steel truss bridge whose truss member was artificially severed. The proposed method detected two different damage levels successfully. [ABSTRACT FROM AUTHOR]

Structural changes are usually associated to damage occurrence, which can be caused by design flaws, constructive problems, unexpected loading, natural events or even natural aging. The structural degrading process affects the dynamic behavior, leading to modifications in modal characteristics. In general, natural frequencies are sensitive indicators of structural integrity and tend to become slightly smaller in the presence of damage. Despite this, it is very difficult to state the relationship between decreasing values of natural frequencies and structural damage, since the dynamic properties are also influenced by uncertainty on experimental data and temperature variation. In order to contribute to improving the quality of natural frequency-based methods used for damage identification, this paper presents a simple and efficient strategy to detect structural changes in a set of experimental tests from a real structure using a computational intelligence method. For a full time monitored structure, the evolution of natural frequencies and temperature are used as input data for a Support Vector Machine (SVM) algorithm. The technique consists on detecting structural changes and when they occur based on the structural dynamic behavior. The results obtained on a historic tower show the capacity of the proposed methodology for damage identification and structural health monitoring. [ABSTRACT FROM AUTHOR]