Search

In view of the problems of low recognition accuracy, multiple extraction parameters and low real-time processing efficiency of existing coal gangue recognition and separation methods based on image processing, an automatic separation method of coal gangue based on wavelet transform was proposed. The method uses wavelet analysis to do noise reduction processing of coal and coal gangue image, and adopts constructing wavelet moment to extract and analyze features of coal and gangue, calculates the feature value, and find out the obvious differences of characteristic parameters between coal and gangue, the characteristic parameters can take as the basis of coal and gangue recognition. The experimental results show that the method improves the efficiency and of on-line identification and separation of coal and gangue with high accuracy.

In recent years, the machine learning method has been successfully applied to the intelligent health monitoring of rolling bearings. However, its effectiveness depends heavily on the vibration signal time series collected. For the purpose of abundantly excavate the fault feature information in time series, some new multi-scale entropy methods are used to solve this problem. However, the multi-scale entropy only considers the low frequency component of the signal, neglects the high frequency component of the signal, loses some fault signals with modulation phenomenon, and reduces the accuracy of fault classification. Therefore, in order to overcome the shortcomings of multi-scale entropy. In this paper, proposing a new hierarchical state space correlation entropy (HSSCEn) to extract the fault features contained in the signal. At the same time, in practice, the fault features extracted by HSSCEn are usually high and nonlinear. Directly input to the classifier for training and testing will reduce the accuracy of machine learning method recognition. Therefore, an improved standardized variable distance classifier is proposed in this paper. It effectively reduces the feature redundancy of signal and improves the accuracy of original classifier. Finally, in order to verify the validity of the fault diagnosis method based on HSSCEn fusion standardized variable distance classifier, two sets of fault classification experiments are carried out in this paper. Compared with existing fault identification classifier algorithms, the proposed method has more advantages in accuracy and stability of fault classification. [ABSTRACT FROM AUTHOR]

Variational mode decomposition provides a feasible method for non-stationary signal analysis, but the method is still not adaptive, which greatly limits the wide application of the method. Therefore, an adaptive spectrum mode extraction method is proposed in this article. The proposed method is mainly composed of spectral segmentation, mode extraction, and feedback adjustment. In the spectral segmentation part, considering the lack of robustness of classical scale space in a strong noise environment, this article proposes a method of fault feature mapping, which solves over-decomposition of variational mode decomposition guided by classical scale space. In the mode extraction part, for insufficient self-adaptability of the single penalty factor in the variational mode decomposition method, this article proposes a method of spectral aggregation factor, which solves multiple penalty factors that conform to different intrinsic modal functions. In the feedback adjustment part, this article proposes the method of transboundary criterion, which makes the result of variational mode decomposition within a preset range. Finally, using dispersion entropy and kurtosis indicators, intrinsic modal function components containing fault frequencies are extracted for envelope spectrum analysis to extract fault characteristic frequencies. In order to verify the effectiveness of the proposed method, the proposed method is applied to simulation signals and bearing fault signals. Comparing the decomposition results of different methods, the conclusion shows that the proposed method is more advantageous for the fault feature extraction of rolling bearings. [ABSTRACT FROM AUTHOR]

When the mechanical transmission mechanism fails, such as gears and bearings in the gearbox, its vibration signal often appears as a periodic impact. Considering the influence of noise, however, the fault signal is often submerged in the noise, so it is necessary to propose a feasible and effective fault extraction method. MOMEDA (multipoint optimal minimum entropy deconvolution adjusted) overcomes the tedious iterative process of MED (minimum entropy deconvolution) and overcomes the resampling trouble in MCKD (maximum correlated kurtosis deconvolution). It is suitable for dealing with periodic impact signal. Besides, aiming at the poor ability of MOMEDA to capture the deconvolution result of target function in a strong noise environment, this paper proposes an improved MOMEDA gearbox fault feature extraction method. Considering that MOMEDA has poor anti-noise performance and can easily cause misdiagnosis in a strong noisy environment, this paper constructs an autoregressive mean sliding model to improve the noise immunity of MOMEDA. Firstly, the stability of the test signal is judged by the autocorrelation coefficient (ACF) and the partial correlation coefficient (PACF). Secondly, the ARMA (autoregressive moving average) model is constructed and a set of optimal model coefficients are obtained to filter the signal, which greatly improves MOMEDA's ability to capture fault features. Thirdly, the fault feature is extracted by MOMEDA, and the fault information is extracted accurately under a strong noise environment. Finally, compared with AR-MED, ARMAMED, and other methods, the advantages of ARMAMOMEDA are verified. Moreover, the effectiveness and superiority of the proposed method are verified by simulation signals and experimental data from the Case Western Reserve University Bearing Data Center. [ABSTRACT FROM AUTHOR]

Minimum entropy deconvolution (MED) is not effective in extracting fault features in strong noise environments, which can easily lead to misdiagnosis. Moreover, the noise reduction effect of MED is affected by the size of the filter. In the face of different vibration signals, the size of the filter is not adaptive. In order to improve the efficiency of MED fault feature extraction, this paper proposes a firefly optimization algorithm (FA) to improve the MED fault diagnosis method. Firstly, the original vibration signal is stratified by white noise-assisted singular spectral decomposition (SSD), and the stratified signal components are divided into residual signal components and noisy signal components by a detrended fluctuation analysis (DFA) algorithm. Then, the noisy components are preprocessed by an autoregressive (AR) model. Secondly, the envelope spectral entropy is proposed as the fitness function of the FA algorithm, and the filter size of MED is optimized by the FA algorithm. Finally, the preprocessed signal is denoised and the pulse enhanced with the proposed adaptive MED. The new method is validated by simulation experiments and practical engineering cases. The application results show that this method improves the shortcomings of MED and can extract fault features more effectively than the traditional MED method. [ABSTRACT FROM AUTHOR]

3D face reconstruction and dense alignment play a key role in digital humans. Learning-based approaches have been developed by employing cascaded convolutional neural networks for feature extraction and spatial-semantic relationship construction. However, these networks often struggle with accurately capturing the complex relationships among various facial components and attributes, particularly in unconstrained environments. Furthermore, the mappings modeled by convolutional weights are at low levels, which are usually implicit and local, and lack global discrimination. In this paper, we propose a global and instance-level perceived relationship matrices-based network (PRMNet) to recover high-fidelity 3D faces and perform accurate dense alignment in unconstrained environments. Specifically, a Key Information Extraction Module (KIEM) extracts crucial features from global and local feature banks. This reduces reasoning costs while maintaining high perceptual quality. Learnable global and instance-level perceived feature relationship matrices are then integrated into the Feature Relationship Reasoning Module (FRRM). This calibration combines key information from both the global macroscopic and sample-specific microscopic views, allowing accurate construction of spatial-semantic relationships to harvest both global discrimination and local relevance. Finally, we introduce a Spatial Reasoning and Guidance Module (SRGM), designed to recalibrate the joint weight responses of feature extraction and its enhancement paths using various attention mechanisms, thereby further enhancing global discrimination ability. Extensive quantitative and qualitative experiments on the benchmark datasets show that our PRMNet outperforms the state-of-the-art. Codes and all resources will be publicly available at https://github.com/Ray-tju/PRMNet. [ABSTRACT FROM AUTHOR]

• This paper proposes an adaptive maximum cyclostationary blind deconvolution (AMCBD) to diagnose bearing faults. • This paper presents the performance efficiency ratio (PER) index. This index is proposed after considering the noise reduction effect and time cost of deconvolution, which can evaluate the cost performance of deconvolution. • On the basis of PER, combined with the equal-step search strategy, an adaptive strategy of filter length is proposed. • This paper proposes a cyclic frequency estimation method based on autocorrelation function of morphological envelope. Blind deconvolution has been proved to be an effective method for fault detection since it can recover periodic impulses from mixed fault signals convoluted by noise and periodic impulses. As a new blind deconvolution technique, maximum cyclostationarity blind deconvolution (CYCBD) has great advantages over minimum entropy deconvolution (MED), maximum correlation kurtosis deconvolution (MCKD), and optimal minimum entropy deconvolution (MOMEDA) in processing bearing fault signals. However, CYCBD has the following two defects: Cyclic frequency needs to be determined in advance; The filter length of CYCBD affects its ability to recover impulses. The ability of CYCBD to recover the impulse will increase with the increase of filter length, but the signal will be distorted if the length is too large. Besides, the time cost will increase significantly after the length is increased. In this paper, an optimization strategy of CYCBD parameters is proposed, and then an adaptive maximum cyclostationarity blind deconvolution (ACYCBD) is proposed. Firstly, aiming at the determination of cyclic frequency, this paper proposes a cyclic frequency set estimation method based on autocorrelation function of morphological envelope, and the validity of the method is verified by simulation and experiment. Secondly, for the second problem, after considering the performance and time cost of CYCBD, the performance efficiency ratio index is proposed. Then, the equal-step search strategy is used to adaptively select the filter length. Finally, the effectiveness of the method is verified by simulation and experiment. [ABSTRACT FROM AUTHOR]