[en] To realize the quantitative damage detection of a rotor, firstly an impedance analytic model is built. Then the change of bending stiffness is introduced as the damage index. Given the circular boundary condition of a rotor, annular elements are used as the analyzed objects and spectral element method is used. The electro-mechanical (E/M) coupled impedance expression of an undamaged rotor is derived with the application of a low-cost impedance test circuit. A Taylor expansion method is used to obtain the approximate E/M coupled impedance expression for the damaged rotor. After obtaining the difference between the undamaged and damaged rotor impedance, a rotor damage detection algorithm is proposed. In this paper, a preset damage configuration is used for the numerical simulation and experiment validation. The detection results have shown that the quantitative damage detection algorithm based on spectral element method and piezoelectric impedance proposed in this paper can identify the location and the severity of the damaged rotor accurately. (paper)

[en] Due to the disadvantages of the current smart Geogrid for geotechnical use only being able measure strain and evaluate load location, a smart Geogrid embedded with fiber Bragg grating (FBG) sensors has been developed. Also, a deformation reconstruction technique has been investigated, which enables the newly designed smart Geogrid to evaluate the deformation fields of the key areas in geotechnical structures. After the fabricating process of the FBG embedded smart Geogrid was briefly introduced, a curvature information based deformation reconstruction method for the smart Geogrid was detailed. In order to optimize the distribution of the FBG nodes in the smart Geogrid, the finite element (FE) simulation data of the three possible causes of deformation were extracted, and the reconstruction results of the four distributions were compared. The results indicated that equidistantly distributed FBG sensors at the ribs of the smart Geogrid were the optimal distribution for the newly designed smart Geogrid. In addition, a modified deformation reconstruction technique was proposed to reduce reconstruction errors due to the stress concentration on the junctions of the smart Geogrid. The modified method, which employs FBG measured strains for calculating the deformation of the ribs and weighted strains to compute the coordinates of the two junctions, was validated by FE simulations. The simulation results illustrated that the modified method can improve the deformation reconstruction accuracy for both a Geogrid embedded with one fiber optic cable into one warp thread and a Geogrid embedded with multiple fiber optic cables in different warp threads. For the purpose of verifying the feasibility of the deformation measurements for the designed smart Geogrid using the proposed reconstruction techniques, experiments for the smart Geogrid embedded with one fiber optic cable were conducted in constant temperature environments. The curvatures of the smart Geogrid were calibrated prior to the deformation experiments in order to remove the errors induced by the strain measurement. The experimental results demonstrated that the reconstruction technique for the newly designed smart Geogrid was capable of evaluating the deformation field, and the modified reconstruction technique was able to effectively improve the reconstruction accuracy in order to fulfill the requirements of geotechnical usages. The newly developed smart Geogrid with deformation reconstruction techniques can be a promising smart Geosynthetic for the reinforcement as well as the monitoring of geotechnical engineering-related applications. (paper)

[en] In this paper, a novel probability-based classification model is proposed for real-time fault detection of power transformers. First, the transformer vibration principle is introduced, and two effective feature extraction techniques are presented. Next, the details of the classification model based on support vector machine (SVM) are shown. The model also includes a binary decision tree (BDT) which divides transformers into different classes according to health state. The trained model produces posterior probabilities of membership to each predefined class for a tested vibration sample. During the experiments, the vibrations of transformers under different conditions are acquired, and the corresponding feature vectors are used to train the SVM classifiers. The effectiveness of this model is illustrated experimentally on typical in-service transformers. The consistency between the results of the proposed model and the actual condition of the test transformers indicates that the model can be used as a reliable method for transformer fault detection. (paper)

[en] As one of the Geometrical Product Specifications that are widely applied in industrial manufacturing and measurement, sphericity error can synthetically scale a 3D structure and reflects the machining quality of a spherical workpiece. Following increasing demands in the high motion performance of spherical parts, sphericity error is becoming an indispensable component in the evaluation of form error. However, the evaluation of sphericity error is still considered to be a complex mathematical issue, and the related research studies on the development of available models are lacking. In this paper, an intersecting chord method is first proposed to solve the minimum circumscribed sphere and maximum inscribed sphere evaluations of sphericity error. This new modelling method leverages chord relationships to replace the characteristic points, thereby significantly reducing the computational complexity and improving the computational efficiency. Using the intersecting chords to generate a virtual centre, the reference sphere in two concentric spheres is simplified as a space intersecting structure. The position of the virtual centre on the space intersecting structure is determined by characteristic chords, which may reduce the deviation between the virtual centre and the centre of the reference sphere. In addition,two experiments are used to verify the effectiveness of the proposed method with real datasets from the Cartesian coordinates. The results indicate that the estimated errors are in perfect agreement with those of the published methods. Meanwhile, the computational efficiency is improved. For the evaluation of the sphericity error, the use of high performance computing is a remarkable change. (paper)

[en] The vibration signals measured from a rolling bearing are usually affected by the variable operating conditions and background noise which lead to the diversity and complexity of the vibration signal characteristics, and it is a challenge to effectively identify the rolling bearing faults from such vibration signals with no further fault information. In this paper, a novel optimization deep belief network (DBN) is proposed for rolling bearing fault diagnosis. Stochastic gradient descent is used to efficiently fine-tune all the connection weights after the pre-training of restricted Boltzmann machines (RBMs) based on the energy functions, and the classification accuracy of the DBN is improved. Particle swarm is further used to decide the optimal structure of the trained DBN, and the optimization DBN is designed. The proposed method is applied to analyze the simulation signal and experimental signal of a rolling bearing. The results confirm that the proposed method is more accurate and robust than other intelligent methods. (paper)

[en] One microwave propagating mode extraction algorithm is proposed for microwave waveguide using variational mode decomposition (VMD). The reflected signal acquired by the waveguide can be seen as the mixture of the propagating mode and evanescent modes. The propagating mode contains information regarding defects and evanescent modes can be treated as noise. By using VMD, the propagating mode can be extracted. Currently, decomposition models are mostly limited by lacking mathematical theory, backward error correction not being allowed in most methods due to the recursive sifting, or the inability to properly cope with noise. In VMD, the bands have been determined adaptively and the corresponding modes are estimated concurrently. An ensemble of modes are derived, and these modes collectively reproduce the input signal while each is being smoothed after demodulation into the baseband. This proposed model is particularly robust to sampling and noise. The bridge between the physical and mathematical models is demonstrated. A coated steel defect detection experiment is conducted using an X-band open-ended rectangular waveguide to evaluate the efficacy of the VMD method. Two samples are demonstrated. The steel with hole sample has a regular and clear defect, whereas the defect of steel with peening is fuzzy. For both samples, the VMD results can accurately identify the defects. (paper)

[en] To improve the image quality of electromagnetic tomography (EMT), a new image reconstruction method of EMT based on a particle filtering algorithm is presented. Firstly, the principle of image reconstruction of EMT is analyzed. Then the search process for the optimal solution for image reconstruction of EMT is described as a system state estimation process, and the state space model is established. Secondly, to obtain the minimum variance estimation of image reconstruction, the optimal weights of random samples obtained from the state space are calculated from the measured information. Finally, simulation experiments with five different flow regimes are performed. The experimental results have shown that the average image error of reconstruction results obtained by the method mentioned in this paper is 42.61%, and the average correlation coefficient with the original image is 0.8706, which are much better than corresponding indicators obtained by LBP, Landweber and Kalman Filter algorithms. So, this EMT image reconstruction method has high efficiency and accuracy, and provides a new method and means for EMT research. (paper)

[en] The uncertainty quantification of particle image velocimetry (PIV) measurements has recently become a topic of great interest as shown by the recent appearance of several different methods within the past few years. These approaches have different working principles, merits and limitations, which have been speculated upon in subsequent studies. This paper reports a unique experiment that has been performed specifically to test the efficacy of PIV uncertainty methods. The case of a rectangular jet, as previously studied by Timmins et al (2012) and Wilson and Smith (2013b), is used. The novel aspect of the experiment is simultaneous velocity measurements using two different time-resolved PIV systems and a hot-wire anemometry (HWA) system. The first PIV system, called the PIV measurement system (‘PIV-MS’), is intended for nominal measurements of which the uncertainty is to be evaluated. It is based on a single camera and features a dynamic velocity range (DVR) representative of typical PIV experiments. The second PIV system, called the ‘PIV-HDR’ (high dynamic range) system, features a significantly higher DVR obtained with a higher digital imaging resolution. The hot-wire is placed in close proximity to the PIV measurement domain. The three measurement systems were carefully set to simultaneously measure the flow velocity at the same time and location. The comparison between the PIV-HDR system and the HWA provides an estimate of the measurement precision of the reference velocity for evaluation of the instantaneous error in the measurement system. The discrepancy between the PIV-MS and the reference data provides the measurement error, which is later used to assess the different uncertainty quantification methods proposed in the literature. A detailed comparison of the uncertainty estimation methods based on the present datasets is presented in a second paper from Sciacchitano et al (2015). Furthermore, this database offers the potential to be used for comparison of the measurement accuracy of existing or newly developed PIV interrogation algorithms. The database is publicly available on the website www.piv.de/uncertainty. (paper)

[en] Cardiovascular diseases (CVDs) remain the biggest cause of deaths worldwide. ECG monitoring is a key tool for early diagnosis of CVDs. Conventional monitors use monopolar electrodes resulting in poor spatial resolution surface recordings and requiring extensive wiring. High-spatial resolution surface electrocardiographic recordings provide valuable information for the diagnosis of a wide range of cardiac abnormalities, including infarction and arrhythmia. The aim of this work was to develop and test a wireless recording system for acquiring high spatial resolution ECG signals, based on a flexible tripolar concentric electrode (TCE) without cable wiring or external reference electrode which would make more comnfortable its use in clinical practice. For this, a portable, wireless sensor node for analogue conditioning, digitalization and transmission of a bipolar concentric ECG signal (BC-ECG) using a TCE and a Mason-likar Lead-I ECG (ML-Lead-I ECG) signal was developed. Experimental results from a total of 32 healthy volunteers showed that the ECG fiducial points in the BC-ECG signals, recorded with external and internal reference electrode, are consistent with those of simultaneous ML-Lead-I ECG. No statistically significant difference was found in either signal amplitude or morphology, regardless of the reference electrode used, being the signal-to-noise similar to that of ML-Lead-I ECG. Furthermore, it has been observed that BC-ECG signals contain information that could not available in conventional records, specially related to atria activity. The proposed wireless sensor node provides non-invasive high-local resolution ECG signals using only a TCE without additional wiring, which would have great potential in medical diagnosis of diseases such as atrial or ventricular fibrillations or arrhythmias that currently require invasive diagnostic procedures (catheterization). (paper)

[en] Knowledge of the temperature distribution of an in-duct gaseous medium is essential in the monitoring of combustion status. To obtain the temperature distribution, an inverse relationship based on the Radon transform is formulated by using the measured time retardation data from a set of acoustic sensors and actuators. The entire spatial distribution can be obtained by interpolating the estimated discrete temperature data using either a path-based or spaced-based method. An interpolation method then determines the precision of the final imaging result. The characteristics and performance of two interpolation methods are investigated in a simulation study by reconstructing the temperature distribution of a rectangular cross-section. To calculate the temperature field, the path-based interpolation method adopts a direct expression of temperature variation along the propagation path, whereas the space-based interpolation method uses data obtained at predetermined points deployed inside the field. The average reconstruction accuracy of the space-based interpolation for temperature fields with 1 and 4 local maxima is 22% and 183% better than that of path-based interpolation, respectively. Also, the space-based interpolation method is more robust with regard to measurement noise than the path-based interpolation method. (paper)