#### Applied Sciences

Journal Information
EISSN : 2076-3417
Total articles ≅ 32,449
Current Coverage
SCOPUS
INSPEC
DOAJ
SCIE
Archived in
SHERPA/ROMEO
Filter:

#### Latest articles in this journal

Published: 26 January 2022
by MDPI
Applied Sciences, Volume 12; https://doi.org/10.3390/app12031310

Abstract:
The lifetime of a cast-resin transformer mainly depends on the condition of insulation material. Partial discharge (PD) is an important reason for insulation deterioration in cast-resin transformers. Identifying the position of PD is very necessary for damage assessment while the transformer is still operating, and the transformer is covered by housing. This paper proposes the investigation of a cast-resin transformer using an AE sensor and HFCT sensor to specify the precise source of PD. In this study, four AE sensors were used to find PD sources, and the high-frequency current transducer (HFCT) technique was used to identify the PD source and the criteria level. The experiment, in the first two parts, identified the possibility of PD, which includes the position of PD. The final part of the experiment verified the position of the PD source of a cast-resin transformer and confirmed the inspection results. AE and HFCT sensors can be used to detect the location of PD sources, confirming the position of the PD source by sensor detection. In addition, the evident partial discharge picture on the insulator surface of high voltage side. The process successfully and accurately identifies and locates the PD source.
Published: 26 January 2022
by MDPI
Applied Sciences, Volume 12; https://doi.org/10.3390/app12031319

Abstract:
Multi-Object Tracking (MOT) techniques have been under continuous research and increasingly applied in a diverse range of tasks. One area in particular concerns its application in navigation tasks of assistive mobile robots, with the aim to increase the mobility and autonomy of people suffering from mobility decay, or severe motor impairments, due to muscular, neurological, or osteoarticular decay. Therefore, in this work, having in view navigation tasks for assistive mobile robots, an evaluation study of two MOTs by detection algorithms, SORT and Deep-SORT, is presented. To improve the data association of both methods, which are solved as a linear assignment problem with a generated cost matrix, a set of new object tracking data association cost matrices based on intersection over union, Euclidean distances, and bounding box metrics is proposed. For the evaluation of the MOT by detection in a real-time pipeline, the YOLOv3 is used to detect and classify the objects available on images. In addition, to perform the proposed evaluation aiming at assistive platforms, the ISR Tracking dataset, which represents the object conditions under which real robotic platforms may navigate, is presented. Experimental evaluations were also carried out on the MOT17 dataset. Promising results were achieved by the proposed object tracking data association cost matrices, showing an improvement in the majority of the MOT evaluation metrics compared to the default data association cost matrix. In addition, promising frame rate values were attained by the pipeline composed of the detector and the tracking module.
Published: 26 January 2022
by MDPI
Applied Sciences, Volume 12; https://doi.org/10.3390/app12031323

Abstract:
Coherent terahertz control of magnetization dynamics is an area of current interest due to its great potential for the realization of magnetization control on ultrafast timescales in commercial devices. Here we report on an experiment realized at the THz beamline of the free electron laser FLASH at DESY which offers a tunable terahertz radiation source and spontaneously synchronized free-electron laser X-ray pulses to resonantly probe the magnetization state of a ferromagnetic film. In this proof-of-principle experiment, we have excited a thin Permalloy film at different THz wavelengths and recorded the induced magnetization dynamics with photons resonantly tuned to the Ni ${M}_{2,3}$ absorption edge. For THz pump pulses including higher orders of the undulator source we observed demagnetization dynamics, which precise shape depended on the employed fundamental wavelength of the undulator source. Analyzing the shape in detail, we can reconstruct the temporal profile of the electric field of the THz pump pulse. This offers a new method for the realization of an in-situ terahertz beamline diagnostic which will help researchers to adjust the pulse characteristics as needed, for example, for future studies of THz induced coherent control of magnetization dynamics.
Published: 26 January 2022
by MDPI
Applied Sciences, Volume 12; https://doi.org/10.3390/app12031296

Abstract:
Deep Neural Networks (DNNs) have shown superior accuracy at the expense of high memory and computation requirements. Optimizing DNN models regarding energy and hardware resource requirements is extremely important for applications with resource-constrained embedded environments. Although using binary neural networks (BNNs), one of the recent promising approaches, significantly reduces the design’s complexity, accuracy degradation is inevitable when reducing the precision of parameters and output activations. To balance between implementation cost and accuracy, in addition to proposing specialized hardware accelerators for corresponding specific network models, most recent software binary neural networks have been optimized based on generalized metrics, such as FLOPs or MAC operation requirements. However, with the wide range of hardware available today, independently evaluating software network structures is not good enough to determine the final network model for typical devices. In this paper, an architecture search algorithm based on estimating the hardware performance at the design time is proposed to achieve the best binary neural network models for hardware implementation on target platforms. With the XNOR-net used as a base architecture and target platforms, including Field Programmable Gate Array (FPGA), Graphic Processing Unit (GPU), and Resistive Random Access Memory (RRAM), the proposed algorithm shows its efficiency by giving more accurate estimation for the hardware performance at the design time than FLOPs or MAC operations.
Published: 26 January 2022
by MDPI
Applied Sciences, Volume 12; https://doi.org/10.3390/app12031304

Abstract:
In the AECO field, there have been many efforts to transform the design, construction, and maintenance methods that have been carried out in the traditional way based on document-based data into digital ones. BIM makes it possible for the information necessary for the entire construction process to be transformed into a language that computers can understand. In the field of architecture, BIM has been actively used throughout the entire construction life cycle through steady research since the 2000s, while COBie was developed in order to deliver information generated in the design and construction phase to the maintenance phase based on the BIM environment. In the field of infrastructure, however, digital data-based maintenance information management has not been actively studied. In particular, in the case of a port, once a facility is built, it has to be maintained for a longer period than other facilities, so facility maintenance is of utmost importance. Therefore, this study applied COBie, which is being used in the architecture field (building), to the port field. To this end, the COBie standard format in the port field was developed based on the port facility object breakdown structure and property breakdown structure. The port COBie developed through this study is included when the BIM model is converted to the IFC model so that facility management can be performed through the IFC-based viewer in the future. The port COBie schema developed in this study can be expanded to other fields such as roads and bridges in the future.
Published: 26 January 2022
by MDPI
Applied Sciences, Volume 12; https://doi.org/10.3390/app12031288

Abstract:
In recent years, CVD diamond-coated tungsten carbide (WC-Co) tools have been widely utilized due to their benefits in the machining of non-ferrous alloys and polymer composite materials, especially carbon-fiber-reinforced plastics (CFRPs). The reconditioning of such coated tools is economically attractive due to their high cost and short tool life. The decoating of the remaining diamond film from the used tools and the subsequent surface preparation by wet chemical pretreatment are essential steps for new CVD diamond film formation. Previously, it was shown that reactive ion beam etching (RIBE) could effectively remove CVD diamond films. However, some degree of WC-Co tool substrate damage is expected due to the high ion energy in RIBE and the chemical activity in wet etching. This study addresses the effects of RIBE decoating and surface pretreatment steps on WC-Co tools with a complex shape in terms of the ion-induced surface damage, geometry alteration, and adhesion of a subsequently re-applied CVD diamond film. Moreover, the cutting performance of the tools subjected to the RIBE decoating and repeated film deposition was studied via CFRP cutting tests. It has been shown that the RIBE decoated and recoated tools had a high level of cutting performance comparable to the new tools.
Published: 26 January 2022
by MDPI
Applied Sciences, Volume 12; https://doi.org/10.3390/app12031333

Abstract:
The combustion process in diesel engines is controlled by the injection rate shape. The stricter emission regulations requiring simultaneous reduction of nitrogen oxides and particulate matter imposes intense research and development activity for achieving clean and robust combustion. This work describes the experimental investigation made for calibration of an engine model and the numerical investigation performed to assess the influences of different injection rate shapes on performances of a diesel engine fuelled with diesel and rapeseed biodiesel B20. The engine model was developed with the AVL-BOOST code using the AVL-MCC combustion mode. The model was calibrated for the reference Top-Hat injection rate shape using experimental data registered for maximum brake torque and maximum brake power speed conditions. Other injection rate shapes such as triangular, trapezoidal, and boot having the same area, start, and duration of injection were investigated in terms of combustion characteristics, performance, and pollutant emissions. The link existing between the injection characteristics and the NOx and Soot emissions highlights that, for the optimal rate of injection shape, a simultaneous reduction of NOx and Soot by 11%, respectively 4% for maximum brake torque and by 22%, respectively 7% for maximum brake power, can be obtained using biodiesel B20.
Published: 26 January 2022
by MDPI
Applied Sciences, Volume 12; https://doi.org/10.3390/app12031292

Abstract:
The sub-tropical broadleaved forests in Pakistan are the main constituents of the ecosystem services playing a vital role in the global carbon cycle. Monotheca buxifolia (Falc.) A. DC. is an important constituent of these forests, encompassing a variety of ecological and commercial uses. To our best knowledge, no quantitative studies have been conducted in these forests across the landscape to establish a baseline for future monitoring. We investigated the forest structural attributes, growing stock characteristics and total biomass carbon stock and established relationships among them in the phytocoenosis of Monotheca forests along an altitudinal gradient in Pakistan to expand an eco-systemic model for assessment of the originally-implemented conservation strategies. A floristic survey recorded 4986 individuals of 27 species in overstory and 59 species in the understory stratum. Species richness (ANOVA; F = 3.239; p = 0.045) and Simpson’s diversity (ANOVA; F = 2.802; p = 0.043) differed significantly in three altitudinal zones, with a maximum value for lower elevations, followed by middle and higher elevations. Based on the importance values, Acacia modesta and Olea ferruginea are strong companions of M. buxifolia at lower and higher altitudes, whereas forests at mid elevation represent pure crop of M. buxifolia (IVI = ≥85.85%). A similar pattern in stem density, volume and Basal area were also recorded. The carbon stock in trees stratum (51.81 T ha−1) and understory vegetation (0.148 T ha−1) contributes high values in the lower elevation forests. In contrast, soil carbon had maximum values at higher elevation (36.21 T ha−1) and minimum at lower elevation (16.69 T ha−1) zones. Aboveground biomass carbon stock (AGB BMC) of woody trees, understory vegetation and soil organic carbon (SOC) were estimated higher (77.72 T ha−1) at higher and lower (68.65 T ha−1) elevations. Likewise, the AGB BMC exhibited a significant (p< 0.05) negative correlation with elevation and positive correlation with soil carbon. We concluded that lower elevation forests are more diverse and floristically rich in comparison to higher altitudinal forests. Similarly, the biomass carbon of Monotheca forests were recorded maximum at low altitudes followed by high and middle ranges, respectively.
Published: 26 January 2022
by MDPI
Applied Sciences, Volume 12; https://doi.org/10.3390/app12031316

Abstract:
Sentiment Analysis is an essential research topic in the field of natural language processing (NLP) and has attracted the attention of many researchers in the last few years. Recently, deep neural network (DNN) models have been used for sentiment analysis tasks, achieving promising results. Although these models can analyze sequences of arbitrary length, utilizing them in the feature extraction layer of a DNN increases the dimensionality of the feature space. More recently, graph neural networks (GNNs) have achieved a promising performance in different NLP tasks. However, previous models cannot be transferred to a large corpus and neglect the heterogeneity of textual graphs. To overcome these difficulties, we propose a new Transformer-based graph convolutional network for heterogeneous graphs called Sentiment Transformer Graph Convolutional Network (ST-GCN). To the best of our knowledge, this is the first study to model the sentiment corpus as a heterogeneous graph and learn document and word embeddings using the proposed sentiment graph transformer neural network. In addition, our model offers an easy mechanism to fuse node positional information for graph datasets using Laplacian eigenvectors. Extensive experiments on four standard datasets show that our model outperforms the existing state-of-the-art models.
Published: 26 January 2022
by MDPI
Applied Sciences, Volume 12; https://doi.org/10.3390/app12031309

Abstract:
The aim of this study was to evaluate the potential use of remote and proximal sensing techniques to identify homogeneous zones in a high density irrigated olive (Olea europaea L.) orchard subjected to three irrigation regimes (full irrigation, deficit irrigation and rainfed conditions). An unmanned aerial vehicle equipped with a multispectral camera was used to measure the canopy NDVI and two different proximal soil sensors to map soil spatial variability at high resolution. We identified two clusters of trees showing differences in fruit yield (17.259 and 14.003 kg per tree in Cluster 1 and 2, respectively) and annual TCSA increment (0.26 and 0.24 dm2, respectively). The higher tree productivity measured in Cluster 1 also resulted in a higher water use efficiency for fruit (WUEf of 0.90 g dry weight L−1 H2O) and oil (WUEo of 0.32 g oil L−1 H2O) compared to Cluster 2 (0.67 and 0.27 for WUEf and WUEo, respectively). Remote and proximal sensing technologies allowed to determine that: (i) the effect of different irrigation regimes on tree performance and WUE depended on the location within the orchard; (ii) tree vigour played a major role in determining the final fruit yield under optimal soil water availability, whereas soil features prevailed under rainfed conditions.