Search

• EMD manifold is proposed for true mode extraction in machinery fault diagnosis. • Sensitive modes with different noise are fused nonlinearly by manifold learning. • The true mode is adaptively learned with the assistance of random noise. • The new method has the merits of mode mixing alleviation and noise removal. • The improved performance is verified in enhanced machinery fault diagnosis. One challenge of the existing noise-assisted methods for solution of mode mixing problem of empirical mode decomposition (EMD) is that, the decomposed modes contain much residual noise derived from both added and self-contained noise. This paper proposes a new noise-assisted method, called EMD manifold (EMDM), for enhanced fault diagnosis of rotating machines. The major contribution is that the new method nonlinearly and adaptively fuses the fault-related modes containing different noise via a manifold learning algorithm, by which true fault-related transients are preserved, while fault-unrelated components including mode-mixing-induced components and the residual noise derived from both the added and self-contained noise are greatly suppressed. First, the sensitive mode is located among the modes obtained by the EMD method according to a newly proposed criterion. Then, a high-dimensional matrix is constructed of the sensitive modes obtained through a small number of EMD trials on the signals plus noise of different amplitudes. Finally, the manifold learning algorithm is performed on the high-dimensional matrix to extract intrinsic manifold of the fault-related transients. The high-dimensional matrix is repeatedly constructed with random noise added to adjust local data distribution of the matrix for adaptive EMDM feature learning. Experimental studies on gearbox and bearing faults are conducted to validate the proposed method and its enhanced performance over traditional noise-assisted EMD methods. [ABSTRACT FROM AUTHOR]

Bio-mechanoreceptors capable of micro-motion sensing have inspired mechanics-guided designs of micro-motion sensors in various fields. However, it remains a major challenge for mechanics-guided designs to simultaneously achieve high sensitivity and broadband sensing due to the nature of resonance effect. By mimicking rat vibrissae, here we report a metamaterial mechanoreceptor (MMR) comprised of piezoelectric resonators with distributed zero effective masses featuring a broad range of local resonances, leading to near-infinite sensitivity for micro-motion sensing within a broad bandwidth. We developed a mechanical frequency-division multiplexing mechanism for MMR, in which the measured micro-motion signal is mechanically modulated in non-overlapping frequency bands and reconstructed by a computational multi-channel demodulation approach. The maximum sensitivity of MMR is improved by two orders of magnitude compared to conventional mechanics-guided mechanoreceptors, and its bandwidth with high sensitivity is extendable towards both low-frequency and high-frequency ranges in 0-12 kHz through tuning the local resonance of each individual sensing cell. The MMR is a promising candidate for highly sensitive and broadband micro-motion sensing that was previously inaccessible for mechanics-guided mechanoreceptors, opening pathways towards spatio-temporal sensing, remote-vibration monitoring and smart-driving assistance., (© 2023. Springer Nature Limited.)

The method of dynamic mass isolation is utilized in a self-moving precision positioning stage actuated by a piezostack to increase its moving speed. Two prototypes, namely, the referenced stage and the modified stage, have been fabricated. The only difference between the two stages is the flexure hinge manufactured in the modified stage to achieve an efficient dynamic mass isolation method. The step response has been investigated. The modified stage with dynamic mass isolation presents the average displacement of 6.6 µm with the applied step voltage being 55 V. By contrast, the referenced stage without dynamic mass isolation presents the average displacement of 1.6 µm. As a type of quasi-static piezoactuator/motors, the modified stage moves approximately four times faster than the referenced stage under the same driving frequency. By utilizing the dynamic mass isolation method, the modified stage still features the advantages of the referenced stage, such as cost-effective controllers, heavy-load capability, and motion of nanoscale. The concept and technique presented in this study can be applied to precision positioning stages for improved speed performance.

Vibrations carry a wealth of useful physical information in various fields. Identifying the multi-source vibration information generally requires a large number of sensors and complex hardware. Compressive sensing has been shown to be able to bypass the traditional sensing requirements by encoding spatial physical fields, but how to encode vibration information remains unexplored. Here we propose a randomized resonant metamaterial with randomly coupled local resonators for single-sensor compressed identification of elastic vibrations. The disordered effective masses of local resonators lead to highly uncorrelated vibration transmissions, and the spatial vibration information can thus be physically encoded. We demonstrate that the spatial vibration information can be reconstructed via a compressive sensing framework, and this metamaterial can be reconfigured while maintaining desirable performance. This randomized resonant metamaterial presents a new perspective for single-sensor vibration sensing via vibration transmission encoding, and potentially offers an approach to simpler sensing devices for many other physical information.

Non-invasive estimation of minute ventilation is important for quantifying the intensity of physical activity of individuals. In this paper, several improved regression models are presented, based on the measurement of chest and abdomen movements from sensor belts worn by subjects (n = 50) engaged in 14 types of physical activity. Five linear models involving a combination of 11 features were developed, and the effects of different model training approaches and window sizes for computing the features were investigated. The performance of the models was evaluated using experimental data collected during the physical activity protocol. The predicted minute ventilation was compared to the criterion ventilation measured using a bidirectional digital volume transducer housed in a respiratory gas exchange system. The results indicate that the inclusion of breathing frequency and the use of percentile points instead of interdecile ranges over a 60 s window size reduced error by about 43%, when applied to the classical two-degrees-of-freedom model. The mean percentage error of the minute ventilation estimated for all the activities was below 7.5%, verifying reasonably good performance of the models and the applicability of the wearable sensing system for minute ventilation estimation during physical activity.

Estimation of ventilation volume from dimensional changes of the rib cage and abdomen is of interest to researchers interested in quantifying internal exposure to environmental pollutants in the atmosphere. In this paper, we present different statistical regression models for estimating ventilation volume during free-living activities. The movements of the rib cage and abdomen were measured by piezoelectric sensor belts. Multiple linear regression as the calibration method was applied. Five regression models with different combinations out of thirteen features were developed and the performance of these models was compared through experimental study of 11 subjects. The effect of training approaches - model trained for each subject and for all subjects, and the effect of time intervals for computing features were also investigated. The results indicate that Model 2, combining respiratory features and breathing frequency, with a longer time intervals will lead to a higher accuracy.

This paper presents a new method for nonlinear trend estimation of non-stationary signals, by which the trend can be self-adaptively decomposed through calculating the midpoint-based local means. In this method, the so-called midpoints are proposed to construct the local mean of a signal instead of two envelopes in the classical empirical mode decomposition (EMD) algorithm, thus resulting in the midpoint-based empirical decomposition. Furthermore, a negentropy-based statistical method is presented to justify decomposition of the trend. Simulation results indicate that the new algorithm improves the performance of signal decomposition and trend estimation in comparison with the classical EMD algorithm. The proposed method also shows the value in self-adaptively estimating the nonlinear respiratory component from non-invasively measured ventilation signals.

Estimation of respiration commonly employs piezoelectric sensors secured to rib cage and abdominal belts. However, these respiratory signals are often contaminated by tissue artifact. This paper presents a signal decomposition technique for tissue artifact removal in respiratory signals, based on empirical mode decomposition (EMD). After introducing the theoretical foundation, this method is performed on three synthetic signals, and performance of tissue artifact removal using EMD is compared with low-pass filter and independent component analysis (ICA) techniques. A simulation study and experimental results show that EMD can effectively remove tissue artifact in respiratory signals.

Objective: To study the contribution of genetic factors to the variance of serum leptin concentration in healthy, normotensive twins., Methods: A total of 57 pairs of twins were investigated: 28 female and 19 male pairs of monozygotic(MZ) twins, and 6 female and 4 male pairs of dizygotic(DZ) twins. The zygosity of twins was determined by comparing the concordance of the genotype of nine fluorescence-labeled microsatellite markers. The genetic analysis was performed using the variance-based method. Serum leptin levels were determined in duplicate by a radioimmunoassay Kit (Linco Research, Inc., St. Charles, Missouri) as previously described., Results: The test of genetic variance revealed a significantly larger within-pair variance of serum leptin in the DZ twins, in comparison with the MZ twins. The corresponding heritability for serum leptin was 8%. Adjusted for BMI, gender, and uric acid (UA), the heritability for serum leptin was 0.18%. Log leptin correlated significantly with blood pressure (SBp r=0.355 P<0.001; DBp r=0.339 P<0.001). Stepwise multiple linear regression analysis revealed that only BMI, gender and UA were linked independently to serum leptin levels(R(2)=0.788, P<0.001)., Conclusion: The above data indicate that environmental factors other than genetic factors are important determinants of leptinemia in normal subjects.