查看更多>>摘要：? 2022 Elsevier LtdRecently, the valve leakage detection of subsea Christmas tree (SCT) attracts considerable attention in the field of underwater resource exploitation. However, most existing leakage detection methods rely on contact-sensors, which are expensive and troublesome to install and maintain. Additionally, it is still a great challenge to extract sensitive fault features for the non-linear and unsteady signal. To address these issues, a novel remote acoustic detection method based on acoustic sensor and deep learning is proposed in this paper. Firstly, the feasibility of SCT valve leakage detection based on acoustic sensors is theoretically demonstrated by acoustic analysis. Secondly, a multi-scale attention module (MSAM) is proposed to obtain rich feature information according to the characteristics of valve leakage acoustic signals. Subsequently, A deep residual shrinkage network based on multi-scale attention module (MSAM-DRSN) is designed to perform valve leakage detection. The proposed method is evaluated by the valve leakage experiment. The results show that the proposed leakage detection method can obtain sensitive fault features from strong background noise signals and the detection performance is better than existing detection methods.
查看更多>>摘要：? 2022Precision deep-hole parts are widely used in various fields of industrial production. Their machining quality has a significant impact on fatigue limit, geometric accuracy, and stability of products. Since roundness and axis straightness errors are essential technical indexes to evaluate the machining quality of deep-hole parts, their accurate measurement is essential to ensure the performance of related products. In this study, a multi-sensor integrated system that can measure two kinds of shape errors simultaneously was developed based on laser displacement sensor, circular grating, two-dimensional position-sensitive detector (PSD) and laser interferometer. It is suitable for deep-hole parts with a diameter of 150–165 mm and a length of 600–20000 mm. Furthermore, the uncertainty analysis models of roundness and axis straightness errors for this system were constructed. Eventually, the effectiveness of the measurement system was successfully verified through experiments.
查看更多>>摘要：? 2022Data augmentation technology has achieved great success to expand the training set for several years. As a representative technology, generative adversarial network and its variants are widely applied in many data augmentation tasks. But the quality of training samples is rarely considered. In this paper, a novel assessable data augmentation named ADA is proposed for mechanical fault diagnosis under noisy labels. First, a sample quality assessment procedure including assessment model construction, approximate calculation based on influence function and screening decision is presented. Thereby, the optimized training set can be obtained. Then, the WGAN-gp model can be established based on the optimized training set and the data augmentation can be accomplished. Finally, a classifier can be trained with the expanded training set and achieve the task of fault diagnosis. The results of two experiments show that the proposed ADA method can effectively improve the fault diagnosis accuracy for various classifiers.
查看更多>>摘要：? 2022In the fast kurtogram (FK), kurtosis is used as an indicator to locate the fault frequency band, and is widely aplied to fault diagnosis. However, kurtosis has been proven to favor a single large impulse rather than the required small fault characteristics, especially in the strong interference environment. To eliminate the impact of large-amplitude impact and further improve the accuracy of fault extraction, a method based on generalized nonlinear spectral sparsity (GNSS) is proposed for fault diagnosis of bearings. First, Z-score normalization and generalized nonlinear sigmoid activation function are used for signal preprocessing, and the scale distribution of the signal will be changed to eliminate the effects of large amplitude shocks under noisy environment. Then, to improve the sparsity measure capability, an improved L3/2 norm is used to replace kurtosis as the basis for selecting the best resonance frequency band. Finally, the effectiveness of the GNSS is verified by simulation data and experimental data. Compared with FK method, the performance of fault extraction of the proposed method is significantly improved, especially for the interference of abnormal impact.
查看更多>>摘要：? 2022Intelligent fault diagnosis in mechanical condition monitoring has emerged in recent years. Consequently, the training efficiency and diagnostic accuracy of fault diagnosis are urgent research topics for real applications. On this basis, this paper proposes fast general normalized convolutional sparse filtering (FGNC-SF) via the L1-L2 mixed norm for intelligent fault diagnosis. The contributions of this paper are as follows. The L1-L2 mix norm is used to penalize the objective function of the algorithm, generalized normalization is used to achieve the row and column normalization of sparse filtering, convolutional activation is used to improve the diagnostic efficiency, and pseudo-normalization is used to improve the test feature distribution. The L1-L2 mixed norm can have two different functions in the algorithm by adjusting different normalization parameters. Generalized normalization no longer limits the normalization parameters to specific values, the characteristic of the proposed approach is illustrated for different normalization parameters. Our method is characterized under different normalization parameters. The selection of optimal parameters is studied in terms of the diagnostic accuracy, the computing time and the standard deviation. The proposed FGNC-SF is validated through two collected rolling bearing datasets. Results show that FGNC-SF exhibits a strong learning ability and is superior to the existing methods for rotating machinery fault diagnosis, obviously improved diagnosis accuracy, efficiency and robustness, reduces the need of priori knowledge and makes intelligent fault diagnosis handle big data more easily.
查看更多>>摘要：? 2022A generalised Warblet tensor rank-1 decomposition (GWTTR1) method is developed in this study to realise location diagnosis for bearing outer raceway defects in low signal-to-noise ratio (SNR) scenarios. Firstly, a novel third-order tensor model (channel–time–frequency) with comprehensive information is established to reveal key information hidden in multidimensional signals and realise feature fusion among two-channel signals. Secondly, the defect characteristic frequency and correlation coefficient between rank-1 tensors are introduced as optimal selection indices for the factor matrix to decompose the tensor model effectively and overcome the inherent shortcomings of decomposition. Thirdly, a novel location diagnosis dimensionless index, namely, the horizontal-vertical synchronisation factor (HVSF), is proposed for the optimised tensor model. Finally, the performance of the proposed GWTTR1 method and novel HVSF index in location diagnosis for bearing outer raceway defects and noise interference elimination is evaluated comprehensively with low SNR dynamic simulation and experimental signals. The comparison results reveal that the GWTTR1 method outperforms existing noise reduction techniques.
查看更多>>摘要：? 2022Amputees suffer from metabolic diseases. Thus, for a healthy life, management through body composition (BC) tests is useful. We aimed to validate between the bioimpedance analysis (BIAInBody_S10) and dual-energy X-ray absorptiometry (DXALunar Prodigy) method for evaluating BC in amputees. 78 male (n = 66) and female (n = 12) unilateral amputees, with either trans femoral amputation (TFA) or trans tibial amputation (TTA), were recruited. Correlation, agreement, and differences between fat free mass (FFM) and percentage fat mass (PFM), computed with the two methods, were tested using methods such as Pearson's and Spearman's correlation, Lin's concordance correlation coefficient (CCC), Bland–Altman plot, bivariate linear regression, and %Diff = 100*(BIA-DXA)/DXA. In all groups, the FFM_BIA value was significantly overestimated compared to FFM_DXA; by contrast, the PFM_BIA value was significantly underestimated with respect to PFM_DXA. Additionally, differences between the results from the two methods were significantly higher for TFA than for TTA. In addition, a lower agreement between the two methods was observed in the TFA compared to the TTA group based on the correlation estimated through Lin's CCC. Moreover, body composition assessment with BIA needs to be carefully interpreted in amputees with some length of residual limb, especially regarding the TFA group.
查看更多>>摘要：? 2022 Elsevier LtdThe prefabricated bridges group refer to several medium- and small-span prefabricated beam bridges locating in a continuous elevated corridor. It is usually challenging to accurately localize the damage in all bridges within one group due to the lack of monitoring data of bridges in the healthy state. To address this issue, a damage localization method for prefabricated bridges group is proposed based on the area-ratio of the strain time-history curve. First, using the definition of the bridge strain influence line, an area equation of the strain time-history curve is derived theoretically under a moving load for a simple beam bridge and a continuous beam bridge. Second, a damage localization index is established based on the area-ratio of the strain time-history curve by analyzing the constitutive characteristics of the area equation of the strain time-history curve at each measurement point. On this basis, a normalization method is presented for the damage localization index at all measurement points, and then the normalized index is implemented to accurately localize damage in all bridges. Finally, the effectiveness and anti-noise performance of the proposed method are demonstrated through both numerical examples and experiment. The proposed method does not need historical health data or a finite element model (FEM) as the reference and can transform strain time-history curves with different amplitudes from different positions along bridges into a unified normalized index that relates to structural stiffness changes only, which is especially suitable for structural damage localization of long prefabricated bridges group.
查看更多>>摘要：? 2022 Elsevier LtdFor the problem that the flow of fluid around the tuning fork affects the detection accuracy of tuning fork sensor, the current methods to evaluate and reduce this influence still rely on manual experience. To improve the detection accuracy of tuning fork densitometer (TFD), we reported a comprehensive evaluation index (CEI) of fluid velocity to analyze the influence of fluid state on the TFD detection accuracy. In this work, several TFD detectors with specific structures were designed to explore the influence of fluid velocity on TFD. Firstly, the Fluent was used to complete numerical simulation, and the CEI was used to analyze the results. Secondly, an experimental platform was built to collect and analyze the TFD output frequency. Finally, it showed that the evaluation results of CEI were consistent with the experimental results. This research can improve the working stability of TFD in the detection process to a certain extent.
查看更多>>摘要：? 2022 Elsevier LtdIdentifying the calorific value of food requires a correct estimate of its volume and size dimensions. The food volumetric estimation can be done rationally and efficiently by measuring the food dimensions in terms of surface parameters. Food volume estimation can be effectively implemented with a computer vision-based application. The food image size can be estimated for its volumetric and calorific calibration with food area measures. However, studies in this area are limited to finding dimensions of a food item with geometrically regular, irregular, amorphous, and solid food shapes. There is a particular challenge with amorphous food items which do not have any shape and are usually calibrated with subjective container sizes by the dietitians and hence cause relative measures. Instance segmentation techniques are implemented at the pixel level and classify a pixel into a food type leading to higher accuracy in classification and segmentation of food over the background. In this work, mask-based RCNN is employed that helps accurate segmentation of food images with regular and irregular shapes in multi-food dish scenarios. The RCNN based food segmentation is applied as a volume estimator model. It is developed by fine-tuning the pre-trained ResNet model and trained over a dataset of 8 different classes of Indian breakfast food images in all shapes. The estimator model yields a precision of 90.9% for convex-shaped food images, 90.46% for amorphous food images in regular serving containers, and 98.5% to 98.9% for regular shaped (square and circle) food items. The accuracy of the presented volume estimator thus opens opportunities for further research with diverse food types and shapes.
NSTL
Elsevier