查看更多>>摘要：? 2022 Elsevier LtdThe quasi-rigid body model is the relationship model between the volume error and the geometric error of the CMM under the consideration of simple deformation. The quasi-rigid body model reflects the comprehensive influence of all geometric errors and ensures the calibration accuracy of geometric errors. However, the geometric error solution based on the quasi-rigid body model has the problem of multicollinearity. To solve this problem, in this study, the geometric errors of CMM are compensated using laser tracing multistation measurement technology and the LASSO (Least absolute shrinkage and selection operator, LASSO) algorithm. First, a laser tracing multi-station measurement model is established, and then the coordinates of the laser tracer station are obtained by solving the redundant equations provided by the laser tracer multi-station measurement system using the L-M (Levenberg Marquardt) algorithm. The global positioning system localization principle combined with the L-M algorithm is used to determine the actual position of the CMM measurement point. Then, the volume error is determined by calculating the difference between the actual position of the measurement point and the planned position. A mathematical model of the volumetric and geometric errors is obtained from the CMM quasi-rigid body model, and the LASSO algorithm is used to solve the model to obtain the 17 geometric errors of the CMM. Using the relationship between the volume error and the uniaxial geometric error, the 4 rotational errors are solved of εz (x), εx- (z), εv (z), εz (z). To verify the measurement accuracy of the compensated CMM, the compensation effect is verified by using the XL80 interferometer with the measuring block. The experimental results indicate that when error compensation is performed, the measurement accuracy of the CMM is improved by >50% compared to the initial state, and no mechanical processing is involved.
查看更多>>摘要：? 2022 Elsevier LtdCurrently, most fault diagnosis methods for wind turbine gearboxes rely on certain unimodal signal, such as vibration or current, which cannot enable reliable and satisfactory performance due to its limited presentation ability. To this end, this paper proposes a new multiview enhanced fault diagnosis framework to learn the correlated and complementary features across current and vibration signals, which are regarded as two different but related views. Multiple statistic features at different wavelet packet decomposition levels are first extracted from raw vibration and current signals, respectively. Then, an unsupervised multiview learning method based on canonical correlation analysis (CCA) is developed to learn maximum correlations between vibration and current features. Finally, the learned enhanced features are used to identify different health conditions. Experimental results show that our proposed method can learn enhanced fault-related features and achieve superior fault diagnosis performance, especially on compound faults, compared with unimodal signal-based methods.
查看更多>>摘要：? 2022 Elsevier LtdNondestructive evaluation for the mechanical properties loss of equipment materials serviced in the hydrogen environment is particularly vital for monitoring the equipment operating conditions. In this study, the quantitative relationship between the introduced hydrogen content and the mechanical property degradation of 2.25Cr-1Mo-0.25V steel was investigated using electrochemical hydrogen charging technique and tensile testing, and a nondestructive testing method based on pulse-echo ultrasonic measurement technique, combined with the results of mechanical assessment for hydrogen embrittlement (HE) susceptibility, was proposed to indirectly estimate the degree of hydrogen-induced plasticity loss of the steel. Moreover, signal reconstruction technique was also used to improve the measurement accuracy of ultrasonic echo signal. The results of the study showed that hydrogen present in the metal lattice intensifies the energy attenuation of ultrasonic echo, verifying the feasibility and validity of the measurement method in the effective online prediction of HE susceptibility of the tested steel.
查看更多>>摘要：? 2022 Elsevier LtdCommercial laser scanners that have the required measurement resolutions, cover too small an area per scan for pavement texture studies. Time and skill demanding image splicing tends to introduce errors. To overcome the limitation, this study developed the laboratory Chang'an University 3D Laser Scanning Device (CUSD) to cover a single-scan area of up to 800 × 325 mm, with the capability to perform a reverse-twin scanning procedure to reduce occlusion effects. To verify the advantages of CUSD, twelve asphalt mixtures were tested by CUSD and a commercial scanner AMES HD9400. CUSD took 19 min to scan and process the image of an area measuring 750 mm long, compared to more than 3 h by AMES HD9400. The experimental verification showed that the mean profile depth (MPD), root-mean-square height (RMS) and height deviation index (Ra) obtained from CUSD images were all within 5% of the values by AMES HD9400.
查看更多>>摘要：? 2022 Elsevier LtdMagnetostrictive patch transducers (MPT) are widely used in non-destructive testing for flexibility, reputability, and durability. Based on the theory of magnetostriction, we developed a method for the design and optimization of axial magnetized MPTs for the transduction of L(0,2) ultrasonic guided waves. To get the optimum static bias magnetic field, we conducted experiments with different magnetostrictive patch aspect ratios, which were based on the theory of demagnetization. Furthermore, we analyzed the variations in solenoid coil parameters with changes in the dynamic alternating magnetic field and induced voltage via simulations and experiments. An optimized MPT was obtained, with its performance improved by 2.79 times compared with those of transducers without magnetostrictive patch.
查看更多>>摘要：? 2022 Elsevier LtdIn the process of gas turbine rotor fault diagnosis based on data-driven, transfer learning is an effective method to solve the lack of gas turbines labeled data, which will result in domain shifts due to the data distribution difference between source domain data and target domain data under variable working condition. A gas turbine fault diagnosis method based on Adversarial Discriminative Domain Adaptation Transfer Learning Network (ADDATLN) is put forward to reduce domain offsets and improve the gas turbine fault diagnosis accuracy. In the proposed method, pre-trained deep Convolutional Neural Networks (CNN) models in the source domain is transferred to target domain data, then deep adversarial training between the source domain and target domain is adopted to adaptively optimize the model parameters of the target domain network, with the purpose of reducing domain offsets and improving gas turbine fault classification accuracy. Field test experiment results on gas turbine rotor fault diagnosis under different working conditions show that the average accuracy of the proposed method reaches 96.45%, and the average accuracy of fault diagnosis on different gas turbines with the same type achieved 95.13%. The field test results confirm that the method effectively reduces the domain differences caused by varying working conditions and different gas turbines, and improves the accuracy of gas turbine rotor fault diagnosis under variable working condition and for different gas turbines with small samples.
查看更多>>摘要：? 2022 Elsevier LtdIn the Internet age, medical image security is very important in the field of e-healthcare. Several encryption algorithms have been developed, but they either do not achieve high security or they are computationally overhead for medical images, and thus may not be suitable for e-healthcare applications. In this paper, we develop a medical image encryption algorithm, namely FastMIE, that can provide security at low computational time. To achieve high security, we encrypt the segmented part of the image by using a key generated by redundant-discrete wavelet transform (RDWT) and randomized-singular value decomposition (RSVD) and scrambled segmented image. To reduce the computational cost, we encrypt only the significant part of the original-image instead of considering the whole image. Compared with the existing works, our FastMIE takes much less time in encrypting and has a greater ability of securing the image.
查看更多>>摘要：? 2022 Elsevier LtdMost existing industrial process fault detection and diagnosis (FDD) techniques operate on data collected at a single scale and focus only on known faults. However, actual process data are inherently multiscale and unknown faults are always inevitable during system running. Therefore, they may perform unsatisfactorily. To tackle this problem, this paper develops a decentralized industrial process FDD framework using multiple enhanced supervised kernel entropy component analysis (enhanced SKECA) models, where each model acts as a fault indicator for one specific fault. Faults can be easily diagnosed by monitoring the outputs of all models within the framework. In particular, when new faults are identified, the framework can update itself only by adding the corresponding enhanced SKECA models without a complete rebuilding process. The monitoring results for the continuous stirred tank reactor (CSTR) process show that the proposed framework is effective in diagnosing both known and unknown faults.
查看更多>>摘要：? 2022 Elsevier LtdA calibration scheme of position and orientation errors of rotary axis average lines based on touch-trigger probing is widely available on many commercial five-axis machine tools. Such a measurement is influenced by error motions of both linear and rotary axes. This paper proposes a novel scheme to separately identify linear and rotary axis geometric errors by using a touch-trigger probe and an uncalibrated test piece. Whereas the proposed scheme is based on well-developed self-calibration schemes for the circularity measurement, an original contribution is that it is applied to separate linear and rotary axis geometric errors. Two case studies are presented. First, the proposed tests are performed on the same five-axis machine tool for half a year to observe a long-term change in linear and rotary axis geometric errors. The second case study investigates the influence of room temperature change on linear axis and rotary axis error motions.
查看更多>>摘要：? 2022 Elsevier LtdIn this study, an analytical model is proposed to measure the through-thickness axisymmetric radial residual stress. Firstly, the rigorous analytical solution for the strain of a plate with a hole subjected to the through-thickness axisymmetric radial stress is found based on the general solution for a homogeneous isotropic elastic solid. It is assumed that the through-thickness axisymmetric radial residual stress can be expanded into the Maclaurin series. Then, the reverse procedure is proposed to determine the coefficients of the series by minimizing the predicted and measured surface radial strains with hole drilling. The proposed model is validated by finite element method (FEM) for a few typical distribution types. The influence of individual terms of the series on the surface radial strain is investigated, and the results show that the high order term results in a small strain value. The model is able to predict the residual stresses by peening as well as those with linear, bi-linear, quad-linear, and sinusoidal distributions. Moreover, a method to measure the resultant force and moment per unit hole circumference is proposed.
NSTL
Elsevier