EISSN : 20763417
Current Publisher: MDPI (10.3390)
Total articles ≅ 7,839
Latest articles in this journal
Applied Sciences, Volume 9; doi:10.3390/app9153166
Abstract:This paper presents an experimental and numerical study to investigate the structural performance of a steel deck-plate system bolted with truss girder. This system has been proposed herein to resolve the issues caused by welding. Structural tests for six full-scale specimens were performed to ensure the structural safety of the proposed system based on design criteria for deflection. Local responses with an emphasis on the failure modes of the system were also assessed using the measured strains at the locations where stresses are localized. Numerical models for all test specimens were developed with the material test data and were validated based on the test results. The structural behaviors of the proposed system, not confirmed in the tests, were further examined using numerical simulations, with a focus on the failure mechanism between the numerical predictions and the test results.
Applied Sciences, Volume 9; doi:10.3390/app9153167
Abstract:In this paper, the basic composition and performance evaluation of the recycled asphalt pavement (RAP) materials were firstly analyzed, and two methods were proposed to evaluate strength characteristics of RAP materials, including a triaxial method for the residual strength and the mortar cementing method for the strength of RAP lump. Then, the cold recycling technology was applied on RAP materials with emulsified asphalt by using vibratory compaction and heavy-duty compaction methods (Proctor compaction method), and the results showed that the maximum dry density obtained by heavy-duty compaction was closer to the actual situation. Finally, the effects of wetting water, emulsified asphalt dosage and curing conditions on the performance of the specimens were investigated. It was found that when the emulsified asphalt was mixed uniformly, whether or not to add the wetting water have almost no effect on the strength of the molded specimens. When the matrix asphalt content of the emulsified asphalt was 30%~60%, the water could be directly added to the cold recycling mixture. The intensity of accelerated curing for two days at 40 °C was approximately equal to that of natural curing for three days, while that of accelerated curing for three days at 40 °C was approximately equal to that of natural curing for seven days, which provided a basis for the short-term laboratory test.
Applied Sciences, Volume 9; doi:10.3390/app9153168
Abstract:This work investigates the natural vibration characteristics of free-form shells when considering the influence of uncertainties, including initial geometric imperfection, shell thickness deviation, and elastic modulus deviation. Herein, free-form shell models are generated while using a self-coded optimization algorithm. The Latin hypercube sampling (LHS) method is used to draw the samplings of uncertainties with respect to their stochastic probability models. ANSYS finite element (FE) software is adopted to analyze the natural vibration characteristics and compute the natural frequencies. The mean values, standard deviations, and cumulative distributions functions (CDFs) of the first three natural frequencies are obtained. The partial correlation coefficient is adopted to rank the significances of uncertainty factors. The study reveals that, for the free-form shells that were investigated in this study, the natural frequencies is a random quantity with a normal distribution; elastic modulus deviation imposes the greatest effect on natural frequencies; shell thickness ranks the second; geometrical imperfection ranks the last, with a much lower weight than the other two factors, which illustrates that the shape of the studied free-form shells is robust in term of natural vibration characteristics; when the supported edges are fixed during the shape optimization, the stochastic characteristics do not significantly change during the shape optimization process.
Applied Sciences, Volume 9; doi:10.3390/app9153169
Abstract:This paper summarizes the top state-of-the-art contributions reported on the MNIST dataset for handwritten digit recognition. This dataset has been extensively used to validate novel techniques in computer vision, and in recent years, many authors have explored the performance of convolutional neural networks (CNNs) and other deep learning techniques over this dataset. To the best of our knowledge, this paper is the first exhaustive and updated review of this dataset; there are some online rankings, but they are outdated, and most published papers survey only closely related works, omitting most of the literature. This paper makes a distinction between those works using some kind of data augmentation and works using the original dataset out-of-the-box. Also, works using CNNs are reported separately; as they are becoming the state-of-the-art approach for solving this problem. Nowadays, a significant amount of works have attained a test error rate smaller than 1% on this dataset; which is becoming non-challenging. By mid-2017, a new dataset was introduced: EMNIST, which involves both digits and letters, with a larger amount of data acquired from a database different than MNIST’s. In this paper, EMNIST is explained and some results are surveyed.
Applied Sciences, Volume 9; doi:10.3390/app9153170
Abstract:Bicyclists can be subjected to crashes, which can cause injuries over the whole body, especially the head. Head injuries can be prevented by wearing bicycle helmets; however, bicycle helmets are frequently not worn due to a variety of reasons. One of the most common complaints about wearing bicycle helmets relates to thermal discomfort. So far, insufficient attention has been given to the thermal performance of helmets. This paper aimed to introduce and develop an adaptive model for the online monitoring of head thermal comfort based on easily measured variables, which can be measured continuously using impeded sensors in the helmet. During the course of this work, 22 participants in total were subjected to different levels of environmental conditions (air temperature, air velocity, mechanical work and helmet thermal resistance) to develop a general model to predict head thermal comfort. A reduced-order general linear regression model with three input variables, namely, temperature difference between ambient temperature and average under-helmet temperature, cyclist’s heart rate and the interaction between ambient temperature and helmet thermal resistance, was the most suitable to predict the cyclist’s head thermal comfort and showed maximum mean absolute percentage error (MAPE) of 8.4%. Based on the selected model variables, a smart helmet prototype (SmartHelmet) was developed using impeded sensing technology, which was used to validate the developed general model. Finally, we introduced a framework of calculation for an adaptive personalised model to predict head thermal comfort based on streaming data from the SmartHelmet prototype.
Applied Sciences, Volume 9; doi:10.3390/app9153173
Abstract:Rockburst intensity prediction is one of the basic works of underground engineering disaster prevention and mitigation. Considering the dynamic variability and fuzziness in rockburst intensity prediction, variable fuzzy sets (VFS) are selected for evaluation and prediction. Here, there are two problems in the application of traditional VFS: (i) the relative membership degree (RMD) calculation process is complex and time-consuming, and the RMD matrix of all indexes can be only obtained by using the RMD function repeatedly; (ii) unreasonable weights of indicators have great impact on the synthetic relative membership degree (SRMD), so it is difficult to guarantee the correctness of the final prediction result. In view of the above problem, this paper established three simplified feature relationship expressions of RMD based on VFS principle and used the SRMD function to establish a BP neural network model to optimize SRMD. The improved VFS method is more efficient and the prediction results are more stable and reliable than the traditional VFS method. The main advantages are as follows: (1) the improved VFS method has higher computational efficiency; (2) the improved VFS method can verify the correctness of RMD at all times; (3) the improved VFS method has higher prediction accuracy; and (4) the improved VFS method has higher fault tolerance and practicability.
Applied Sciences, Volume 9; doi:10.3390/app9153172
Abstract:The main objective of this study is to develop and compare hybrid Artificial Intelligence (AI) approaches, namely Adaptive Network-based Fuzzy Inference System (ANFIS) optimized by Genetic Algorithm (GAANFIS) and Particle Swarm Optimization (PSOANFIS) and Support Vector Machine (SVM) for predicting the Marshall Stability (MS) of Stone Matrix Asphalt (SMA) materials. Other important properties of the SMA, namely Marshall Flow (MF) and Marshall Quotient (MQ) were also predicted using the best model found. With that goal, the SMA samples were fabricated in a local laboratory and used to generate datasets for the modeling. The considered input parameters were coarse and fine aggregates, bitumen content and cellulose. The predicted targets were Marshall Parameters such as MS, MF and MQ. Models performance assessment was evaluated thanks to criteria such as Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) and correlation coefficient (R). A Monte Carlo approach with 1000 simulations was used to deduce the statistical results to assess the performance of the three proposed AI models. The results showed that the SVM is the best predictor regarding the converged statistical criteria and probability density functions of RMSE, MAE and R. The results of this study represent a contribution towards the selection of a suitable AI approach to quickly and accurately determine the Marshall Parameters of SMA mixtures.
Applied Sciences, Volume 9; doi:10.3390/app9153171
Abstract:Smart glasses for wearable augmented reality (AR) are widely used in various applications, such as training and task assistance. However, as the field of view (FOV) in the current AR smart glasses is narrow, it is difficult to visualize all the information on the AR display. Besides, only simple interactions are supported. This paper presents a comparative and substantial evaluation of user interactions for wearable AR concerning visual contexts and gesture interactions using AR smart glasses. Based on the evaluation, it suggests new guidelines for visual augmentation focused on task assistance. Three different types of visual contexts for wearable AR were implemented and evaluated: stereo rendering and direct augmentation, and non-stereo rendering and indirect augmentation with/without video background. Also, gesture interactions, such as multi-touch interaction and hand gesture-based interaction, were implemented and evaluated. We performed quantitative and qualitative analyses, including performance measurement and questionnaire evaluation. The experimental assessment proves that both FOV and visual registration between virtual and physical artifacts are important, and they can complement each other. Hand gesture-based interaction can be more intuitive and useful. Therefore, by analyzing the advantages and disadvantages of the visual context and gesture interaction in wearable AR, this study suggests more effective and user-centric guidance for task assistance.
Applied Sciences, Volume 9; doi:10.3390/app9153174
Abstract:The in-vehicle controller area network (CAN) bus is one of the essential components for autonomous vehicles, and its safety will be one of the greatest challenges in the field of intelligent vehicles in the future. In this paper, we propose a novel system that uses a deep neural network (DNN) to detect anomalous CAN bus messages. We treat anomaly detection as a cross-domain modelling problem, in which three CAN bus data packets as a group are directly imported into the DNN architecture for parallel training with shared weights. After that, three data packets are represented as three independent feature vectors, which corresponds to three different types of data sequences, namely anchor, positive and negative. The proposed DNN architecture is an embedded triplet loss network that optimizes the distance between the anchor example and the positive example, makes it smaller than the distance between the anchor example and the negative example, and realizes the similarity calculation of samples, which were originally used in face detection. Compared to traditional anomaly detection methods, the proposed method to learn the parameters with shared-weight could improve detection efficiency and detection accuracy. The whole detection system is composed of the front-end and the back-end, which correspond to deep network and triplet loss network, respectively, and are trainable in an end-to-end fashion. Experimental results demonstrate that the proposed technology can make real-time responses to anomalies and attacks to the CAN bus, and significantly improve the detection ratio. To the best of our knowledge, the proposed method is the first used for anomaly detection in the in-vehicle CAN bus.
Applied Sciences, Volume 9; doi:10.3390/app9153158
Abstract:Monitoring critical temperatures in permanent magnet synchronous motors is crucial for improving working reliability. Aiming at resolving the difficulty in online temperature estimation, an accurate and simple five-node lumped parameter thermal network (LPTN) is proposed and the mathematical model of the LPTN is built. Both radial and axial heat transfer paths inside the motor are considered to model the complete thermal circuit. In addition, an innovative parameter identification method based on multiple linear regression is applied to identify the parameters of the LPTN model. The parameters in the state equation are identified instead of the data of the motor, which are strongly dependent on the material and geometrical parameters. Finally, an open-loop estimation scheme based on the state equation and Kalman filter algorithm is adopted to predict the motor temperature online. The model performances are validated by extensive experiments under varying speed and torque conditions in terms of the accuracy and robustness. The results indicate that the temperature estimation error is within the range of ±5 °C in most cases and the proposed model can quickly follow the load variation. Besides, the online temperature estimation scheme and parameter identification method are easy and convenient to implement in an embedded system, which is feasible in automobile applications.