Neural network classification of hyperspectral images on the basis of the Hilbert–Huang transform

The method of image classification with its preliminary transformation to principal components and with the use of the Hilbert–Huang transform is studied by an example of neural network classification of a hyperspectral image. The efficiency of the method is demonstrated through comparisons with traditional methods of neural network classification with the use of spectral components and principal components without involving spatial information as features. Radial-basis and complex neural networks are used for classification.     

Extraction of microseismic waveforms characteristics prior to rock burst using Hilbert–Huang transform

This study provides a new research idea concerning rock burst prediction. The characteristics of microseismic (MS) waveforms prior to and during the rock burst were studied through the Hilbert–Huang transform (HHT). In order to demonstrate the advantage of the MS features extraction based on HHT, the conventional analysis method (Fourier transform) was also used to make a comparison. The results show that HHT is simple and reliable, and could extract in-depth information about the characteristics of MS waveforms. About 10 days prior to the rock burst, the main frequency of MS waveforms transforms from the high-frequency to low-frequency. What’s more, the waveforms energy also presents accumulation characteristic. Based on our study results, it can be concluded that the MS signals analysis through HHT could provide valuable information about the coal or rock deformation and fracture.     

Time–frequency interpretation of multi-frequency signal from rotating machinery using an improved Hilbert–Huang transform

The Hilbert–Huang transform (HHT) has proven to be a promising tool for the analysis of non-stationary signals commonly occurred in industrial machines. However, in practice, multi-frequency intrinsic mode functions (IMFs) and pseudo IMFs are likely generated and lead to grossly erroneous or even completely meaningless instantaneous frequencies, which raise difficulties in interpreting signal features by the HHT spectrum. To enhance the time–frequency resolution of the traditional HHT, an improved HHT is proposed in this study. By constructing a bank of partially overlapping bandpass filters, a series of filtered signals are obtained at first. Then a subset of filtered signals, each associated with certain energy-dominated components, are selected based on the maximal-spectral kurtosis–minimal-redundancy criterion and the information-related coefficient, and further decomposed by empirical mode decomposition to extract sets of IMFs. Furthermore, IMF selection scheme is applied to select the relevant IMFs on which the HHT spectrum is constructed. The novelty of this method is that the HHT spectrum is just constructed with the relevant, almost monochromatic IMFs rather than with the IMFs possibly with multiple frequency components or with pseudo components. The results on the simulated data, test rig data, and industrial gearbox data show that the proposed method is superior to the traditional HHT in feature extraction and can produce a more accurate time–frequency distribution for the inspected signal.     

Analysis of signal characteristics of swirlmeter in oscillatory flow based on Hilbert–Huang Transform (HHT)

In the paper, firstly, on an experimental facility, we investigated the measurement characteristics of a diameter 50 mm swirlmeter in uniform flow and oscillatory flow. At the same time, the interference characteristics of oscillatory flow were studied. Then, the signal characteristics of swirlmeter in oscillatory flow were analyzed by Hilbert–Huang Transform (HHT) method. Results show that the response characteristics of swirlmeter in oscillatory flow are addition of that of swirlmeter in uniform flow and the interference characteristics of oscillatory flow. They further prove the conclusions which suppose that the correlation between the velocity pressures of fluid disturbs wave and that of vortex precession in swirlmeter is linear in the literature, and a new method for the oscillatory flow swirlmeter noise removal on HHT was provided.     

Roller bearing safety region estimation and state identification based on LMD–PCA–LSSVM

The basic idea of safety region is introduced into roller bearing condition monitoring. Local mean decomposition (LMD), principal component analysis (PCA) and least square support vector machine (LSSVM) are used comprehensively for the estimation of the safety region and the identification of normal state and faulty state for the roller bearing operational status. First, the vibration acceleration data was segmented according to a certain time interval and then Product Functions (PFs) of each piece of the data were obtained by LMD. Based on this, statistics control limits T² and SPE were extracted by PCA as roller bearings’ state characteristics. Finally, LSSVM was used for the estimation of the safety region of the roller bearing operation state, and multi-class LSSVM was used for the identification of the four normal, ball fault, inner race fault and outer race fault states. The results show that both the safety region estimation and state identification are accurate, and confirm the validity of the LMD–PCA–LSSVM method.     

A fault identification method of rotating machinerybased on t-SNEEI北大核心CSCD

A fault identification method ofrotating machinery is proposed,which combines wavelet packet of time-frequency analysis and manifold learning.Firstly,the sampled vibration signal is decomposed to multilayer information with wavelet packet decomposition(WPD) method.Andevery level data of wavelet packet decomposition is processed bydemodulatingof Hilbert transform,eliminating the high frequency noiseof FIR filterand reducing the data length of the low frequency of resampling.Further,every level data vector is deal with normalization and calculated for the auto power spectrum.Finally,the manifold learning methods of t distributed stochastic neighbor embedding(t-SNE) is applied to do dimension reduction to generate 2D manifold figure data.Different fault forms of gearbox have different manifold features,which is used to identify failure status of equipment.With the experiment test,the feasibility and effectiveness of this identification method is verified.     

Sparse decomposition method based on time–frequency spectrum segmentation for fault signals in rotating machinery

The impulse signal in large rotating machinery with damage fault is sparse, weak, coupled, and even nonperiodic in intermittent operation. To extract this complex signal is a key topic in machinery fault diagnosis. Sparse decomposition (SD) has excellent adaptability in describing arbitrary complex signals based on over-complete dictionary. However, the pursuit speed of best atom is a serious drawback. To alleviate this, a method of sparse decomposition based on time–frequency spectrum segmentation (SD-TFSS) is introduced. Generalized S transform (GST) provides the capability to show the distribution of vibration signals, but the resolution is susceptible to noise, multiresolution generalized S-transform (MGST) is developed to generate multiresolution time–frequency spectrums. Then, spectrums fusion with an appropriate threshold is adopted to acquire multiresolution binary spectrums and produce an optimal binary spectrum. From this optimal binary spectrum, all the connectivity areas are extracted and marked by spectrum segmentation. Thus, an optimal library can be constructed by selecting the optimal atoms of every connectivity area, and the signal can be expressed with this library. We conduct simulations and experiments demonstrating that the proposed method performs well with lower pursuit complexity, higher decomposition efficiency, and better approximation precision.     

Feature extraction of AC square wave SAW arc characteristics using improved Hilbert–Huang transformation and energy entropy

In order to extract the arc feature information related to welding quality in alternating current square wave submerged arc welding (AC Square Wave SAW), an improved Hilbert–Huang transform (HHT) is put forward to investigate the time–frequency distribution of arc current, and the energy entropy is employed to quantitatively judge the arc characteristics. The empirical mode decomposition (EMD) is used to decompose the collected current signal into a number of Intrinsic Mode Functions (IMFs). The method for removing the high frequency and undesirable low-frequency IMFs is proposed by using the correlation coefficient of the IMF and the original signal as criterion, and the valid IMFs are selected for the Hilbert transform and energy entropy calculation. The improved HHT combining with energy entropy can quantitatively describe the time–frequency energy distribution characteristics of the arc current signal at different duty cycle, frequency and welding speed. Experimental results are provided to confirm the effectiveness of this approach to extract the arc physical information related to welding quality.     

Non-stationary signal analysis based on general parameterized time–frequency transform and its application in the feature extraction of a rotary machine

With the development of large rotary machines for faster and more integrated performance, the condition monitoring and fault diagnosis for them are becoming more challenging. Since the time-frequency (TF) pattern of the vibration signal from the rotary machine often contains condition information and fault feature, the methods based on TF analysis have been widely-used to solve these two problems in the industrial community. This article introduces an effective non-stationary signal analysis method based on the general parameterized time–frequency transform (GPTFT). The GPTFT is achieved by inserting a rotation operator and a shift operator in the short-time Fourier transform. This method can produce a high-concentrated TF pattern with a general kernel. A multi-component instantaneous frequency (IF) extraction method is proposed based on it. The estimation for the IF of every component is accomplished by defining a spectrum concentration index (SCI). Moreover, such an IF estimation process is iteratively operated until all the components are extracted. The tests on three simulation examples and a real vibration signal demonstrate the effectiveness and superiority of our method.     

Nonlinear dynamic systems identification using recurrent interval type-2 TSK fuzzy neural network – A novel structure

In this study, a novel structure of a recurrent interval type-2 Takagi-Sugeno-Kang (TSK) fuzzy neural network (FNN) is introduced for nonlinear dynamic and time-varying systems identification. It combines the type-2 fuzzy sets (T2FSs) and a recurrent FNN to avoid the data uncertainties. The fuzzy firing strengths in the proposed structure are returned to the network input as internal variables. The interval type-2 fuzzy sets (IT2FSs) is used to describe the antecedent part for each rule while the consequent part is a TSK-type, which is a linear function of the internal variables and the external inputs with interval weights. All the type-2 fuzzy rules for the proposed RIT2TSKFNN are learned on-line based on structure and parameter learning, which are performed using the type-2 fuzzy clustering. The antecedent and consequent parameters of the proposed RIT2TSKFNN are updated based on the Lyapunov function to achieve network stability. The obtained results indicate that our proposed network has a small root mean square error (RMSE) and a small integral of square error (ISE) with a small number of rules and a small computation time compared with other type-2 FNNs.     

Influence of Aerated Lubricants on Gear Churning Losses–An Engineering Model

It is currently assumed that churning losses can be described by using only two physical parameters representative of the lubricant properties; that is, density and viscosity. To verify this hypothesis, a number of transient measurements were carried out on a specific gear test rig over a range of oil temperatures. It appears that, for high temperatures and/or high rotational speeds, the drag torque can suddenly increase with a larger Reynolds number. Based on extensive online lubricant aeration measurements, it is demonstrated that this particular behavior can be related to churning losses when the fraction of air in the lubricant reaches a certain threshold. In order to quantify the influence of oil sump aeration on churning losses, a simplified original model, based on surface tension and lubricant aeration, is proposed. This study shows that density and viscosity are not sufficient to estimate churning losses under some specific operating conditions and the need to account for other physical properties of the lubricant is emphasized.     

Robust integral of neural network and precision motion control of electrical–optical gyro-stabilized platform with unknown input dead-zones

Parametric uncertainty associated with unmodeled disturbance always exist in physical electrical–optical gyro-stabilized platform systems, and poses great challenges to the controller design. Moreover, the existence of actuator deadzone nonlinearity makes the situation more complicated. By constructing a smooth dead-zone inverse, the control law consisting of the robust integral of a neural network (NN) output plus sign of the tracking error feedback is proposed, in which adaptive law is synthesized to handle parametric uncertainty and RISE robust term to attenuate unmodeled disturbance. In order to reduce the measure noise, a desired compensation method is utilized in controller design, in which the model compensation term depends on the reference signal only. By mainly activating an auxiliary robust control component for pulling back the transient escaped from the neural active region, a multi-switching robust neuro adaptive controller in the neural approximation domain, which can achieve globally uniformly ultimately bounded (GUUB) tracking stability of servo systems recently. An asymptotic tracking performance in the presence of unknown dead-zone, parametric uncertainties and various disturbances, which is vital for high accuracy tracking, is achieved by the proposed robust adaptive backstepping controller. Extensively comparative experimental results are obtained to verify the effectiveness of the proposed control strategy.     

Brittle–ductile mode cutting of glass based on controlling cracks initiation and propagation

Owing to brittleness and hardness, functional glass is one of the most difficult to cut materials. This paper proposes a new machining method—brittle–ductile mode machining combining both properties of brittle breakage and plastic flow of glass. Edge-indention experiments are first conducted in order to deduce the laws of crack initiation and propagation in the process of glass cutting, then a single-straight tool with big inclination angle is designed for glass cutting based on the laws of crack initiation and propagation and properties of plastic flow. With this new tool, the lateral and subsurface cracks initiation can be suppressed, and media cracks propagate away from machined surface. At the same time, the requirements for machining glass in ductile manner can be fulfilled. Validation experiments show that highly efficient and precise glass cutting can be achieved at the cutting depth of sub-millimeter level, and an integral and crack-free surface with good finish can be obtained. This method overcomes the process restriction on critical cutting depth and tool feed for ductile regime turning technology and can be transferred to mass production.     

Wavelength switchable and stable single-longitudinal-mode erbium-doped fiber laser based on Mach–Zehnder interferometer and tunable filter

An erbium-doped fiber laser with all-fiber Mach–Zehnder interferometer (MZI) and tunable filter was proposed and experimentally demonstrated. In the designed fiber laser, 6?m C-band erbium-doped fiber was selected as a gain medium; the MZI comprised two waist-enlarged fiber bitapers. In the experiment, the laser threshold was 93?mW, whereas a switchable single-longitudinal-mode laser was realized within 1519.7–1564.6?nm by adjusting the tunable filter and the line interval was less than 2.5?nm; for single-wavelength laser, the peak power difference of each line was less than 4?dB, and the power fluctuation was less than 0.77?dB within 10-min scan time. A stable and switchable dual-wavelength laser was realized, the wavelength spacing of each dual-wavelength laser was less than 0.7?nm, the side-mode suppression ratio was more than 30.2?dB, and the power shift was less than 0.39?dB. The laser’s 3-dB linewidth was less than 0.1?nm.     

Bearing remaining useful life estimation based on time–frequency representation and supervised dimensionality reduction

The extraction of ideal age feature is a challenging task in vibration-based bearing remaining useful life (RUL) estimation. Aiming at this problem, a new approach is proposed on the basis of time–frequency representation (TFR) and supervised dimensionality reduction. Firstly, S transform and Gaussian pyramid are employed to obtain TFRs at multiple scales. Textural features of TFRs are used as the high-dimensional features. Then, a two-step supervised dimensionality reduction technique, i.e. principal component analysis (PCA) plus linear discriminant analysis, is employed to reduce the dimensionality, in which the target dimension and number of classes are taken as variable parameters. Finally, the simple multiple linear regression model is utilized to estimate the RUL. Experimental results indicate that the proposed approach outperforms the methods using traditional statistical features and/or PCA. Additionally, variable conditions of load and speed should be considered in the future to further improve the proposed approach.     

Reliability analysis based on the principle of maximum entropy and Dempster–Shafer evidence theory

The Probability density functions (PDFs) of some uncertain parameters are difficult to determine precisely due to insufficient information. Only the varying intervals of such parameters can be obtained. A method of reliability analysis based on the principle of maximum entropy and evidence theory was proposed to address the reliability problems of random and interval parameters. First, the PDFs and cumulative distribution functions of interval parameters were obtained on the basis of the principle of maximum entropy and Dempster–Shafer evidence theory. Second, the normalized means and standard deviations of interval parameters were obtained using the equivalent normalization method. Third, two explicit iteration algorithms of reliability analysis were proposed on the basis of the advanced firstorder and second-moment method to avoid solving the limit state function and obtain the reliability index. Finally, the accuracy and efficiency of the proposed methods were verified through a numerical example and an engineering case.     

A flexible artificial neural network–fuzzy simulation algorithm for scheduling a flow shop with multiple processors

One of the most popular approaches for scheduling manufacturing systems is dispatching rules. Different types of dispatching rules exist, but none of them is known to be globally the best. A flexible artificial neural network–fuzzy simulation (FANN–FS) algorithm is presented in this study for solving the multiattribute combinatorial dispatching (MACD) decision problem. Artificial neural networks (ANNs) are one of the commonly used metaheuristics and are a proven tool for solving complex optimization problems. Hence, multilayered neural network metamodels and a fuzzy simulation using the α-cuts method were trained to provide a complex MACD problem. Fuzzy simulation is used to solve complex optimization problems to deal with imprecision and uncertainty. The proposed flexible algorithm is capable of modeling nonlinear, stochastic, and uncertain problems. It uses ANN simulation for crisp input data and fuzzy simulation for imprecise and uncertain input data. The solution quality is illustrated by two case studies from a multilayer ceramic capacitor manufacturing plant. The manufacturing lead times produced by the FANN–FS model turned out to be superior to conventional simulation models. This is the first study that introduces an intelligent and flexible approach for handling imprecision and nonlinearity of scheduling problems in flow shops with multiple processors.     

Artificial neural network model in surface modification by EDM using tungsten–copper powder metallurgy sintered electrodes

Electro-discharge machining (EDM) is a widely accepted nontraditional machining process used mostly for machining materials difficult to machine by conventional shearing process. Surface modification by powder metallurgy sintered tools is an uncommon aspect of EDM. Of late, it is being explored by many researchers. In the present paper, attempts have been made to model the surface modification phenomenon by EDM with artificial neural networks. Two output measures, material transfer rate and average layer thickness, have been correlated with different process parameters and presented in the form of plots. The predicted results are matching well with the experimental results.     

Erratum to “Research on feature enhancement method of weak fault signal of rotating machinery based on adaptive stochastic resonance”

Journal of Mechanical Science and Technology -