Search

Face recognition is a biometric technique that uses data on facial features to identify individuals. It is also a key area of study for computer vision researchers. CNN is a subclass of feedforward neural networks with convolutional processing and depth structure and one of the illustrative deep learning techniques. Since the deep learning theory was put forth and computational power increased, CNN has rapidly advanced and is now utilized in computer vision, natural language processing, and other fields. Our research is focused on face recognition, and because the mini-Xception model has a condensed volume and few parameters, it is used in this study. The dataset we used is fer2013, which is a classical dataset among CNN algorithms and is used in many studies. We also used data augmentation methods, and Keras’ ImageDataGenerator image generator was the optimal data augmentation method we came up with after reading the paper. Finally, we came up with a final model with 61% accuracy, which we are satisfied with and within the error results of the papers we reviewed.

Cl−-induced corrosion is one ineluctable problem we face when metals and alloys are exposed to marine environment. Herein, we give a mini review covering the aspects of combating Cl−-induced corrosion from the perspective of surface chemistry. Firstly, the interaction of Cl− with oxide surface is briefly reviewed. Then, we highlight some recent strategies for repelling, blocking and capturing Cl− by using surface chemistry control and surface modification via self-assembling at molecular levels.

As a promising semi-active device, magneto-rheological damper has been widely used in low-frequency vibration isolation fields (within 20 Hz) such as bridge damping and building seismic resistance. Recently, the application of magneto-rheological damper has extended to medium and high frequency fields such as satellite and power engine vibration control, accompanied with an urgent need of detailed understanding of its output characteristics. In this paper, a comprehensive physical model is established to analyze dynamic performance of the magneto-rheological damper. The model, derived from both Poiseuille and Couette flow, aims to describe the relationship between the flow rate and pressure difference. The compressibility of the magneto-rheological fluid, the inertia of both the fluid and piston assembly, and the friction are involved to capture the medium and high frequency dynamics of the damping force. Theoretical calculation and simulation verification of magnetic circuit are conducted. Then the experiment based on a self-made prototype is carried out. The results show that the damping force calculated by proposed physical model matches well with the experimental results across the predefined range of frequency and coil current levels.

Material degradation is accompanied by the changes in surface structure, morphology, and composition. These changes can be recorded by a variety of image acquisition devices that export digital images in grayscale or true color to a detector. Information regarding corrosion type and extent can be extracted with image processing methods. This paper provides a comprehensive review of material degradation assessed by digital image processing. Digital image processing systems used to assess material degradation are briefly reviewed, and the algorithms developed to process metallic materials degradation images are described. Physical and electrochemical methods that can be used to support digital image processing results are summarized, and future work that will augment the present methods of evaluating material degradation are discussed.

In this work, the carbothermal reduction and hydrochloric acid leaching were utilized for separation and recovery of iron and rare earth from the Bayan Obo tailings. The effects of temperature, add...

Stainless steel (SS) is one of the most widely used engineering materials in marine engineering. However, its corrosion in the marine atmospheric environment due to the high concentration of Cl− is a problem. The SS corrosion is a threat to the development and security of marine industry; therefore, evaluating the corrosion resistance of SSs is necessary. In this work, atmospheric corrosion detection probes based on a symmetrical electrode system were used to study the corrosion behaviors of 304 SS and 2205 duplex stainless steel (DSS) in a simulated marine atmosphere. A theoretical model for electrochemical noise (EN) data analysis based on the Thevenin electrochemical equivalent circuit (EEC) model was established. The relationship between the EN characteristic parameters and the corrosion rate was obtained. The Thevenin EEC model analysis showed that the relationship between the noise resistance (Rn), the noise impedance [Rsn(f)], and the impedance modulus (|Z(f)|) was $$R_{{\text{n}}} \approx R_{{{\text{sn}}}} = {}^{\sqrt 3 }\left| {Z(f)} \right|$$Rn≈Rsn=3Z(f). Thus, Rn and Rsn can be used as indicators for quantitative corrosion evaluation. The results of EN detection for the 304 SS and 2205 DSS showed that in a simulated marine atmospheric environment, the passive films on the two SSs were relatively intact at the initial exposure stage, and their dissolution rates were slow. The corrosion resistance of the 2205 DSS was higher than that of the 304 SS. With the deposition of Cl− on the SS surface, pitting was initiated and the dissolution rate increased. The pitting initiation process on the SS surface was random, and part of the active pores could be repassivated.

The purpose of this work was to enhance the corrosion resistance of the passive film on 304 stainless steel (SS) by chemical modification in alkaline phosphate–molybdate solutions. The 304 SS was passivated in both phosphate and phosphate–molybdate mixed solutions to investigate the effect of molybdate on its corrosion resistance. The experimental results indicated that the passive film showed better corrosion resistance in Cl−-containing solutions after modification in phosphate–molybdate solutions than in phosphate-only solutions. Energy-dispersive spectroscopy analyses revealed that the passive film formed in phosphate–molybdate solutions contained Mo and P after modification, which is the reason for the enhanced corrosion resistance.

Purpose This paper aims to quantify atmospheric corrosion by image analyses. The corrosion extent, form and distribution of corrosion product on Q235B and T91 steels exposed to a Zhoushan marine atmosphere over one year are characterized by image analysis. Design/methodology/approach Image analysis of corrosion images were achieved using the gray value, wavelet analysis and fuzzy Kolmogorov–Sinai (K–S) entropy. Findings As corrosion becomes extensive, the gray value of corrosion images decreases, and the energy value of nine subimages after wavelength decomposition decreases. Fuzzy K–S entropy increases as localized corrosion propagates but decreases as uniform corrosion spreads. Originality/value The methods proposed in this work open a new way for fast corrosion evaluation of metallic materials exposed to atmospheric conditions.

A new comprehensive utilization of vanadium slag (VS) method focusing on inhibiting leaching of iron (Fe) during the leaching of vanadium (V) using sulfuric acid (SA) was proposed. In this process, Cr2O3 was added to VS to conjugate with Fe in the VS to form (Fex,Cr1-x)2O3 which is insoluble in SA, resulting in the decrease of leaching ratio (LR) of Fe to avoid the subsequent separation difficulty of V in leachate. The phase evolutions of VS during the roasting and SA leaching process, and the influences of roasting temperature, roasting time, and addition of Cr2O3 on the LR of V, Fe and chromium (Cr) from VS were studied. When the addition of Cr2O3 is 12 wt.%, the mass concentration of V in the leachate is 1 order magnitude higher than Fe and the mass ratio of V to Fe reaches 18.34. The LR of V, Fe and Cr are 91% 1.39% and 0.28%, respectively. The leaching residue can be reused as ironmaking raw material. More importantly, the (Fex,Cr1-x)2O3 and Fe2TiO5 can be separated from the leaching residue and recycled as raw materials for black ceramic pigments and titanium dioxide production by mineral processing technology, respectively.

Pitting corrosion of 7075 and LY12 aluminum alloys in Cl−-containing atmosphere was investigated by utilizing the electrochemical noise (EN) analyses. Theoretical predictions based on equivalent circuit (EC) approach and experimental tests were compared to quantify atmospheric corrosion rate. The reliability of EN parameters (including noise resistance, spectral noise resistance) were further checked by electrochemical impedance spectroscopy (EIS) results. The localized corrosion rate can be successfully quantified by using noise resistance and spectral noise resistance, however impact of electrolyte resistance should be checked prior to EN measurement because it may lead to an error in calculating the corrosion rate.

This work aims to develop field corrosion detection method for identifying the defect levels in organic coatings by using electrochemical noise (EN) measured at Singe Cell (SC) mode. An electrochemical sensor was used for defects identification, the results indicate that, as the defect levels rise, the amplitudes of the electrochemical current noise (ECN) increase. Further data analysis based on power spectral density (PSD), and discrete wavelet analysis confirmed that white noise level WL, and wavelet energy distribution can successfully discriminate the levels of protectivity of the organic coating.

International audience; Electrochemical noise (EN) can be used in situ to investigate corrosion processes and to detect and monitor the corrosion of metallic materials. EN data are largely influenced by the measurement mode, the surface area of the working electrodes, the electrolyte resistance, and the symmetry of the electrode system. Herein, the advantages and limitations of electrochemical kinetics, equivalent circuit, and shot noise methods for quantifying corrosion rates with EN are discussed.; Le bruit électrochimique (EN) peut être utilisé in situ pour étudier les processus de corrosion et pour détecter et surveiller la corrosion des matériaux métalliques. Les données EN sont largement influencées par le mode de mesure, la surface des électrodes de travail, la résistance électrolytique et la symétrie du système d'électrodes. Ici, les avantages et les limites de la cinétique électrochimique, du circuit équivalent et des méthodes de bruit de grenaille pour quantifier les taux de corrosion avec EN sont discutés.

In this paper, a crack parameter identification method based on the coordinate of 1x intersections and Kriging surrogate model with enhanced samples is proposed to solve the crack parameter identification problem of train axles in start-stop condition. The location of the crack is identified in the first step by the x-coordinate of the linear intersection of 1x frequency components. Then, the intersecting coordinates and rigid constraint points were used to solve response curves of equal crack depth, and numerous enhanced samples of same depth at different positions were obtained to establish a Kriging model. Finally, the crack depth was directly obtained by the Kriging model. In this paper, response curve s of equal crack depth is proposed according to the symmetry of the train structure, which greatly reduces the number of samples needed to construct the Kriging model. Numerical simulation and experimental results demonstrate the effectiveness and robustness of the proposed method.

Focusing on the difficulty in vibration control of internal combustion power plant because of its diverse substructures and complex excitation, an abnormal vibration diagnosis and attenuation method based on the system’s whole-body vibration signals is proposed. The characteristics of the exciting forces, such as toppling torque, inertia force (or moment) and coupling centrifugal force, are analyzed. The resonance frequency distribution of the power unit is obtained by modal analysis. Then an expression that could reveal the relationship between the whole-body vibration intensity and the exciting force amplitude is established, and conclusion is drawn that different excitations would drive the system to vibrate at different frequencies. Based on this basic relation, a fault diagnosis method relying on comprehensive analysis of vibration signals in both time and frequency domain is proposed. Vibration attenuation methods targeted at specific faults are also put forward. Finally, the validity of the proposed method is verified by applying it to a 12V280 diesel generator set.

Reduced sulphur induced corrosion of steam generator (SG) tubing in pressurized water reactor (PWR) is assumed to occur at low temperature. Corrosion of UNS N08800 in chloride and thiosulphate solutions at 20, 40 and 90 ℃ is systemically investigated by using polarization technique, electrochemical impedance spectroscopy (EIS), and X-ray photoelectron spectroscopy (XPS). It is concluded that thiosulphate can be reduced to S°, S22− and S2−, and incorporate into the passive film, which significantly depends on temperature. The evolution on pitting potentials in chloride + thiosulphate solution can be ascribed to the competitive adsorption of these two ions on the passive surface.

In this work, an electrochemical system based on electrochemical noise (EN) technique for online detection and monitoring of atmospheric corrosion of LY12CZ aluminium alloys has been established. A detecting probe and a monitoring instrument with a software have been developed to perform the electrochemical noise measurements with zero resistance ammeter (ZRA) mode. Experimental results show that the atmospheric corrosion behaviour of aluminium could be effectively detected and monitored by the analysis of the electrochemical potential and current noise, also by the noise resistance variation.

Some critical issues when applying electrochemical noise (EN) in field corrosion detection of nuclear materials are solved. In this work, a portable EN measurement system for field test is established by using Compact RIO modular instrument and zero resistance ammeter (ZRA) module. The electrode system for EN measurement is built by using platinum wire or stainless steels coated with thermal sprayed ceramic coatings as counter electrode. Two EN sensors are developed based on the electrode system mentioned above, and field corrosion detection is successfully achieved. The corrosion of stainless steel pipeline surface in nuclear power plant and internal-surface corrosion of continuous blowdown piping are successfully evaluated by the established new EN measurement system.

The diesel multiple unit (DMU) has been widely used in high-speed railway service due to its flexibility and economy. Considering the broadband and complex vibration generated by DMU power package, the advanced semi-active suspension with magnetorheological (MR) dampers is introduced to promote anti-vibration performance. In this work, a comprehensive optimal design approach for MR damper used in DMU power package is proposed. Quasi-static modeling process is conducted to obtain MR damper's low-frequency outputs, while its high-frequency damping forces are calculated by physical modeling considering the fluid compressibility and piston assembly inertia. Then the objective functions and optimization variables are determined. Based on response surface and linear correlation analysis, the influence of the optimal variables on the objective functions is discussed. Using reference-point based nondominated sorting approach (NSGA-III), the evolutionary many-objective optimization is conducted. In addition, magnetic design is incorporated into the optimal process to ensure the magnetic flux density in the effective working area. Finally, an optimized MR damper prototype is manufactured and tested. By comparing the experimental damping force with calculated results in both low-frequency and high-frequency ranges, the effectiveness of the presented optimal method for MR dampers is validated.

Three electrochemical probes were designed to study the effects of weather on the atmospheric corrosion of carbon steel and low-alloy steel in an urban atmosphere by using the electrochemical noise (EN) technique. The fabricated probes can be successfully applied to investigate instant atmospheric corrosion resistance in various weather conditions. Continuous 15-day EN monitoring showed that, under the same weather conditions, the corrosion resistance of Q235 carbon steel is higher than that of low-alloy steels.

In this work, a novel electrochemical sensor combining with a thin insulating net is designed for monitoring atmospheric corrosion of metals by using electrochemical noise (EN) technique. Experimental results reveal that the sensor can successfully in-situ detect the atmospheric process and the corresponding corrosion form, which has promising applications in the field of atmospheric corrosion monitoring of metal structures in the future.

The early stage atmospheric corrosion of T91 and Q235B steels exposed to Tianjin’s urban atmosphere over 20 days was studied using two electrochemical probes via an electrochemical noise (EN) technique. To identify the corrosion process and the corrosion form of the two steels, EN data were analysed by statistics and wavelet transform. The results revealed that the wavelet energy of decomposed EN mainly located at high-frequency level for Q235B steel, whereas it mainly located at the low-frequency level for T91 steel. Analyses of surface images confirmed that Q235B steel underwent uniform corrosion whereas T91 steel suffered from localised corrosion. The obtained noise resistance correlated well with weight loss data.

Purpose This paper aims to investigate the stress corrosion cracking (SCC) process of sensitized 304 stainless steel during the slow strain rate test by using the electrochemical noise (EN) technique. Design/methodology/approach EN data are interpreted based on chaos and wavelet analyses, and correlation dimension and wavelet energy distribution are used as indicators for SCC process identification. Findings Experimental results reveal that the corrosion potential abruptly decreases from 180 to 100 mV at 6,300 s and the current increases from 10 to 100 nA accordingly, which is attributed to passive film breakdown and crack initiation. Chaos and wavelet analyses results reveal that, as crack initiates, the correlation dimensions increase from 1.2 to 1.9, and the corresponding distribution frequencies of maximum relative wavelet energy change from high frequency to low frequency. Originality/value SCC is monitored in lab, and crack initiation and propagation are identified by chaos and wavelet analyses. This work lays the foundation for SCC detection in field using EN.

Due to the advantage in deactivating fluorine and enhancing the decomposition of rare earth (RE) minerals, MgO was applied to the magnetizing roasting of Bayan Obo tailings in this work. The effects of MgO dosages, roasting temperature, and holding time on the decomposition rate of RE minerals were experimentally studied. With a MgO dosage of 10 wt.%, the decomposition rate of RE minerals reached 98.09% at 750 °C. The phase composition of roasted samples was characterized by XRD and SEM-EDS. The incomplete decomposition rate was investigated with the observation of leaching residual by SEM-EDS. The decomposition kinetics of the RE minerals with the addition of MgO was analyzed with the Ginstling-Brundshtein model, where the reaction rate was controlled by chemical reaction.

The stress corrosion crack (SCC) of 304 nuclear grade (NG) stainless steel (SS) in 0.5 mol/L NaCl+1.5 mol/L H2SO4 was monitored using electrochemical noise (EN) based on chaos theory, statistics and wavelet analysis. The results indicated that the SCC process was divided into three stages according to the transient features in the EN. In the beginning, compared with the sample without applied stress, the enhanced fluctuation amplitude in the electrochemical current noise (ECN) of the stressed samples was attributed to stress-enhanced pitting corrosion and uniform corrosion; then the fluctuations of ECN for all the samples decreased due to a coverage by the corrosion products; however, the ECN fluctuations of stressed sample were larger than the unstressed sample, suggesting that the stress enhanced the SCC initiation and propagation. Chaos analysis revealed that the correlation dimensions increase from 2.1 to 2.5 during the corrosion process, and the applied stress seems increase the complexity and uncertainty of the ECN signal.

Electrochemical noise (EN), as one of the most promising in situ electrochemical methods in corrosion and electrochemical science, has been developing rapidly in recent years with the advancements in instrumentation and signal processing methods. One advantage of EN is its application in long-term or early stage corrosion process monitoring because it instantly detects corrosion rate and corrosion forms. Investigators have applied various mathematical methods to extract characteristic parameters from EN. In this paper, identifying corrosion forms using parameters obtained from time domain, frequency domain and time–frequency domain is reviewed, and the correlation between parameters and corrosion forms is discussed. Finally, other in situ techniques are recommended to be employed synchronously with EN measurement in order to obtain reliable analyses.

To evaluate the residual life and safe operation of metallic materials, it is crucial to accurately identify the intensity and depth of corrosion pits. Scanning acoustic microscopy (SAM) using tomographic acoustic microimaging (TAMI) under C-mode was employed to determine corrosion pit morphology and depth on aluminum alloy 7050, and the reults were cross-checked by optical microscopy. Maximum and average pitting depths of 204 μm and 80 μm, respectively, were determined by multilayer scannning, and a pitted area of 30 ∼ 40 % was calculated using binary image processing.

The goal of this work was to prepare a layer of dodecyl phosphate (DDPO4) film on LY12 aluminum alloy substrate for corrosion protection using self-assembly technique. The prepared DDPO4 Film properties were characterized by X-ray photoelectron spectroscopy (XPS), water contact angle measurement, and electrochemical measurement. Experimental results indicated that DDPO4 was successfully assembled on aluminum alloy substrate via covalent bond attachment. The modified surface was hydrophobic due to the DDPO4 attached to the oxide surface and a hydrocarbon tail-up orientation. Accelerated corrosion test showed that the DDPO4 modified LY12 aluminum alloy substrate exhibited excellent corrosion resistance in an electrolyte containing 0.01 M Cl− + 0.01 M SO42−, and no apparent corrosion pits were observed after an exposure of 96 h. The charge transfer resistance of the DDPO4 covered aluminum alloy was ∼100 times larger than that of the blank aluminum alloy.

Electrochemical probes and sensors have been developed to detect and monitor atmospheric corrosion of metallic materials in the past 40 decades. Depending on the measurement methods, the electrodes and structures of probes and sensors can be different. Various mathematical methods and models have been developed to determine the time-dependent corrosion rate of metal under thin electrolyte film. Polarization techniques such as electrochemical impedance spectroscopy (EIS) and linear polarization resistance (LRP) have the advantage of easy data interpretation but have a tendency to interfere with the corrosion system under investigation. Nonpolarized techniques such as electrochemical noise (EN) do not disturb the corrosion system but data interpretation can be problematic. To achieve long term and reliable corrosion monitoring, optimized electrode design and a multichannel electrochemical instrument are required. New corrosion models and novel data interpretation methods are needed in future work.

This paper aims to monitoring the atmospheric corrosion of Q235B steel by electrochemical noise (EN) technique, and both statistical parameters and frequency domain analysis are used to quantify the corrosion rate. The reliability of EN measurement was evaluated by electrochemical impedance spectroscopy (EIS). Experimental results indicate that electrolyte resistance have an impact on EN measurement but it can be discarded if it is much lower than the impedance of working electrode (WE). Theoretical predictions using Thevenin equivalent circuit model reveal that spectral noise impedance is not equal to impedance module as estimated by EIS, and a relationship between them is derived. Power spectral density (PSD) obtained from experimental data further verified the theoretical predictions.