查看更多>>摘要：? 2022 Elsevier LtdBy different testing methods (electrochemical techniques, potential shift monitoring, and Environmental Scanning Electron Microscope), this research evaluates the stray current corrosion of steel rebar in different layouts. The more significant corrosion state is observed when the steel bar is parallel to stray current flow, compared to the situation as a steel bar is vertical to the stray current. These outcomes are further clarified by the recorded level of stray current picked-up by steel rebar. It is found that the level of current actually picked-up by the steel rebar is decreasing. At the instant when the stray current supply is just turned off, an opposite current flow (back flow) is recorded. Besides an expansion of the database for monitoring stray current interference on reinforced concrete structures, the recorded results can be the basis for better understanding the process of stray current interference.
查看更多>>摘要：? 2022 Elsevier LtdPavement crack detection methods based on deep learning and computer vision can greatly improve detection efficiency and accuracy, but in many cases the data in training set is lacking or uneven, making it insufficient to train an accurate detection model. This paper proposes a detection method under small samples, which is composed of two steps. First, with a generative adversarial network (GAN) constructed, the small sample data set of pavement cracks taken by unmanned aerial vehicle (UAV) is used as the training set and the GAN model is trained. The best trained model is used for generation of new images. Second, original small-sample data set is expanded by images generated by the GAN model, and a convolutional neural network (CNN) model is constructed at the same time. Then, data set before and after the expansion is trained and tested by the method of transfer learning to verify the effectiveness of expanded data separately. It has been proved that, compared with the unexpanded data set, CNN model trained after expansion improves the test set detection accuracy from 80.75% to 91.61%, which is regarded as a significant improvement. In addition, this paper also uses class activation map (CAM) to visually evaluate CNN model, and expands the detection ability of classification model.
查看更多>>摘要：? 2022 Elsevier LtdAn accurate aircraft dynamic model is essential for the flight control system design and flight characteristic analysis of tiltrotor unmanned aerial vehicle (UAV). The unique aerodynamic interaction between the rotor and wing and the installation misalignment angles of rotor could obviously degrade the control performance of flight. However, the traditional test-bed-based identification method cannot fully simulate the real flying conditions. Thus, a real-time method based on the Error-State Kalman Filter (ESKF) is proposed to estimate the model parameters of the tiltrotor UAV. Based on the navigation information from the onboard integrated navigation system, the model parameters are augmented into the aircraft dynamic model to be estimated. Then, the observability analysis is carried out to demonstrate the convergence of the proposed method and provide the proper measurements from which the model parameters possess the best observable degree. Finally, both simulations and real flight experiments are performed to validate the effectiveness of the proposed method.
查看更多>>摘要：? 2022 The Author(s)Guided waves recently have attracted significant interest as a very promising research area. The signals registered by a specially designed sensor network are processed to assess the state of the tested structure. Despite the constant development of novel damage detection algorithms employing guided waves, the phenomenon of wave propagation still needs detailed recognizing and understanding for the further progress of non-destructive wave-based methods. Special attention is paid to guided waves in plate-like structures, but the majority of considered cases concern plates with constant thickness. However, in the real world, we often deal with specimens with variable thickness. The thickness variability of the specimen is often forced to fulfill the construction requirements and optimize stress distribution or is the result of degradation i.e. corrosion. Thus, the development of NDT methods forces the need of considering specimens with complex geometry and the problem of wave propagation in waveguides with variable thickness is crucial for improving novel as well as so far proposed algorithms. The article presents the results of the analytical, numerical and experimental analysis of wave propagation in plates with variable thickness. The analysis concerns the influence of thickness distribution of plate structure on wave velocity, the time course wave packet and amplitude. Moreover, the novel approach based on constrained convex optimization for determining the plate thickness distribution has been proposed and verified during numerical and experimental campaigns.
查看更多>>摘要：? 2022 Elsevier LtdIn the low frequency range, the reverberation of a room should be characterized by the decay time, which can be determined experimentally by various methods. In order to make accurate and precise measurements, the differences between these methods and their advantages should be known and quantified. In this work, the performance of four linear and four adaptive time–frequency impulse response analysis methods was evaluated with the aim to find the most accurate method for characterizing the decays of the room modes. The methods were compared using generated Prony test signals with known modal properties and measured room impulse responses were also examined. The decay time using the selected time–frequency impulse response method was then estimated in three rooms and the measurement results were compared with interrupted sine tone methods. It has been shown that the choice of measurement setup and analysis parameters has an important influence on the estimation of the decay time and its modal frequency. The time–frequency impulse response method using the window width optimized Stockwell transform, the continuous wavelet transform, or the Morlet wave method were found to be the most robust methods, and their applicability is also enhanced by the fact that only a single measurement is required for the selected microphone position. The presented test methods could be used to evaluate the performance of other approaches for the evaluation of reverberation below the Schroeder frequency or the ability of measurement and signal processing techniques to characterize low frequency transient signals.
查看更多>>摘要：? 2022 Elsevier LtdNon-separation measurement is a research difficulty and hotspot in multiphase flow. In this paper, the multimodal ultrasonic sensors and differential pressure transmitter combined with two-fluid model are used to measure the wet gas flow rates under non-separation state. Firstly, a modeling measurement system is built. The liquid film thickness and pressure drop are obtained by using ultrasonic sensor and differential pressure transmitter. Based on the steady-state one-dimensional two-fluid model and the limitations of existing models, a new prediction model is proposed. Comparing the pressure drop data obtained by new two-fluid model with the experimental data, the errors are within ±30%. Then, wet gas flow rates measurement system is constructed. The liquid film thickness and gas velocity are measured by using multimodal ultrasonic sensors. Combined this new two-fluid model, the wet gas flow rates are obtained. The measurement errors of gas and liquid flow rate are within ±10% and ±15%.
查看更多>>摘要：? 2022 Elsevier LtdTo solve the problem that position information obtained by SINS/DVL integrated navigation system is divergent in the underwater navigation applications, this paper introduces hydroacoustic single beacon (HSB) into the underwater integrated system and proposes a novel SINS/HSB/DVL fault-tolerant federated variational Bayesian robust adaptive Kalman filter (FVBRAKF). In the FVBRAKF, the traditional KF is replaced by VBRAKF, which not only suppress the adverse influence of outliers based on Mahalanobis distance (MD) effectively, but also estimate the unknown measurement noise covariance based on variational Bayesian (VB) approximation adaptively. Meanwhile, a novel adaptive information sharing factor (ISF) method is proposed during the information fusion process to form an improved FVBRAKF（IFVBRAKF）, which can adjust the fusion weights in real-time according to the accuracy of local filter. The semi-physical simulation experiments for SINS/HSB/DVL underwater integrated navigation system based on the test data are carried out to verify the adaptive ability of the ISF and the robust adaptive ability of the method, respectively. Experimental results demonstrate that the proposed algorithm can not only improve the estimation accuracy, but also enhance the fault-tolerant performance.
查看更多>>摘要：? 2022 Elsevier LtdRecent advancements in machine learning and deep learning models find them helpful in designing effective complex measurement systems. At the same time, examining the brain's activities using Electroencephalography (EEG) is essential to determine the mental state or thought of a person. It is essential in several application areas, such as Brain-Computer Interface (BCI), emotion recognition, and mental disease diagnosis. The proper brain signal classification using EEG finds helpful diagnose epileptic seizures. Since the traditional seizure detection process is a lengthy and challenging task, the automated identification of epilepsy is a significant problem. In order to resolve the issues that exist in the traditional brain signal classification models, this study designs Automated Deep Learning-Enabled Brain Signal Classification for Epileptic Seizure Detection (ADLBSC-ESD). The proposed ADLBSC-ESD technique aims to classify the brain signals to determine the existence of seizures or not. In addition, the presented model involves the design of the Improved Teaching and Learning-Enabled Optimization (ITLBO) technique for selecting features from EEG signals. Moreover, the Deep Belief Network (DBN) model is used for an effectual classification of EEG signals, and the hyperparameters of the DBN model are optimally tuned using the Swallow Swarm Optimization Algorithm (SSA). In order to ensure the improved brain signal classification performance of the ADLBSC-ESD technique, a series of simulations take place, and the outcomes are investigated concerning different measures. The experimental values highlighted the better performance of the ADLBSC-ESD technique over the current state of art techniques with maximum accuracy of 0.8316 and 0.8609 under binary and multiple classes, respectively.
查看更多>>摘要：? 2022 Elsevier LtdAccuracy of machinery fault diagnosis and interpretability of diagnosis methods are fundamental to safe operation of machinery and help to improve the universality of the model. Mechanical vibration signals can reflect the operating state of the machine. Therefore, to improve the accuracy of fault diagnosis, this paper constructs a 6-layer residual neural network (ResNet06), which embeds two residual blocks to fully extract features of the mechanical vibration signals. Then, we use the gradient-based class activation map (Grad-CAM) and eigenvector-based class activation map (Eigen-CAM) to interpret the ResNet06 visually and to verify the ResNet06 correctness. Experimental results indicate that the fault diagnosis accuracy of our proposed model can reach almost 100%, and it can be seen that the model can accurately capture the fault points by the visualization of the model.
查看更多>>摘要：? 2022 Elsevier LtdThe occurrence of coal and rock dynamic disasters threatens and restricts the safe production of underground coal mining activities seriously. For the prevention and control of the dynamic disasters, it is the foundation and premise to monitor the damage evolution and forecast the failure of coal mass under joint action of loading stress and gas pressure. For this, the stress loading and electric potential (EP) measuring experimental system of gas-bearing coal specimen was established, to test and analyze the precursor information for the failure of gas-bearing coal from EP response. The results show that the EP intensity of gas-bearing coal is positively correlated with loading stress level. The EP signals show step-like changes once the specimen is severely damaged at local moment. Based on mathematical statistics, the multifractal and critical slowing down paratemers of temporal EP data were analyzed, to reveal the nonlinear characteristics of gas-bearing coal during the loading and damaging process. In the critical failure stage, the multifractal parameter Δα rapidly increases to the peak and another parameter Δfm changes significantly to a negative value. Furthermore, the variance (V), as a critical slowing down parameter, rapidly rises to the peak value very shortly (0.3 s) before the failure occurs. Simultaneously, the autocorrelation coefficient (ac) also exhibits abnormal responses. Analysis combining the above indexes enables comprehensive monitoring and identification of precursor information for the failure of gas-bearing coal from temporal EP data.
NSTL
Elsevier