Experimental Performance Evaluation Research Articles

DATran: Dual Attention Transformer for Multi-Label Image Classification

Multi-label image classification is a fundamental yet challenging task, which aims to predict the labels associated with a given image. Most of previous methods directly exploit the high-level features from the last layer of convolutional neural network for classification. However, these methods cannot obtain global features due to the limited size of convolutional kernels, and they fail to extract multi-scale features to effectively recognize small-scale objects in the images. Recent studies exploit the graph convolution network to model the label correlations for boosting the classification performance. Despite substantial progress, these methods rely on manually pre-defined graph structures. Besides, they ignore the associations between semantic labels and image regions, and do not fully explore the spatial context of images. To address above issues, we propose a novel Dual Attention Transformer (DATran) model, which adopts a dual-stream architecture that simultaneously learns spatial and channel correlations from multi-label images. Firstly, in order to solve the problem that current methods are difficult to recognize small-size objects, we develop a new multi-scale feature fusion (MSFF) module to generate multi-scale feature representation by jointly integrating both high-level semantics and low-level details. Secondly, we design a prior-enhanced spatial attention (PSA) module to learn the long-range correlation between objects from different spatial positions in images to enhance the model performance. Thirdly, we devise a prior-enhanced channel attention (PCA) module to capture the inter-dependencies between different channel maps, thus effectively improving the correlation between semantic categories. It is worth noting that PSA module and PCA module complement and promote each other to further augment the feature representations. Finally, the outputs of these two attention modules are fused to obtain the final features for classification. Performance evaluation experiments are conducted on MS-COCO 2014, PASCAL VOC 2007 and VG-500 datasets, demonstrating that DATran model achieves better performance than current state-of-the-art models.

Experimental and Numerical Evaluation of the Cavitation Performances of Self-Excited Oscillating Jets

Experimental and Numerical Evaluation of the Cavitation Performances of Self-Excited Oscillating Jets

MLGCN: an ultra efficient graph convolutional neural model for 3D point cloud analysis.

With the rapid advancement of 3D acquisition technologies, 3D sensors such as LiDARs, 3D scanners, and RGB-D cameras have become increasingly accessible and cost-effective. These sensors generate 3D point cloud data that require efficient algorithms for tasks such as 3D model classification and segmentation. While deep learning techniques have proven effective in these areas, existing models often rely on complex architectures, leading to high computational costs that are impractical for real-time applications like augmented reality and robotics. In this work, we propose the Multi-level Graph Convolutional Neural Network (MLGCN), an ultra-efficient model for 3D point cloud analysis. The MLGCN model utilizes shallow Graph Neural Network (GNN) blocks to extract features at various spatial locality levels, leveraging precomputed KNN graphs shared across GCN blocks. This approach significantly reduces computational overhead and memory usage, making the model well-suited for deployment on low-memory and low-CPU devices. Despite its efficiency, MLGCN achieves competitive performance in object classification and part segmentation tasks, demonstrating results comparable to state-of-the-art models while requiring up to a thousand times fewer floating-point operations and significantly less storage. The contributions of this paper include the introduction of a lightweight, multi-branch graph-based network for 3D shape analysis, the demonstration of the model's efficiency in both computation and storage, and a thorough theoretical and experimental evaluation of the model's performance. We also conduct ablation studies to assess the impact of different branches within the model, providing valuable insights into the role of specific components.

Experimental Investigation and Performance Evaluation of Jatropha Oil-Biodiesel Blending with Kerosene for Domestic Cooking and Lighting Applications.

Reducing indoor air pollution and the related costs requires developing efficient cooking and lighting technologies as well as alternative energy sources. The appropriateness of virgin jatropha (Jatropha curcas) oil, its biodiesel, and the blending of these fuels with kerosene for wick stove lighting and cooking was examined in this study. To describe the fuel performance, a range of blending ratios were examined and characterizations were made of the fuel's density, calorific value, boiling point, viscosity, and rate of fuel consumption. To assess the fuels' thermal degradation behaviors, thermogravimetric and water boiling tests were performed. An air pollution meter was used to describe the levels of indoor air pollution. According to the findings, the virgin oil from Jatropha has a kinematic viscosity of 30.1 mm2/s. The 50% jatropha oil and 80% biodiesel blended with kerosene showed a reduction in viscosity of 72.6% and 46.8%, respectively. The thermal efficiency of the virgin oil, its biodiesel, and blending of these fuels with kerosene was in the range of 10-48%. Complete degradation of jatropha oil, its biodiesel, and 40 : 60 ratio of jatropha biodiesel and kerosene blend was conducted in the temperature range of 480-700 k, 185-280.6 K, and 100-300 K, respectively. The activation energies of jatropha oil, biodiesel, and kerosene blend (40 : 60) were 191.3, 73, and 25 kJ/mol, respectively. The average concentrations of particulate matter and carbon monoxide for pure jatropha oil biodiesel were 209.71 µg/m3 and 5.5 mg/kg, respectively. Thus, jatropha biodiesel and its blending with kerosene are suitable fuels for cooking and lighting operations in rural communities who are living far from the electrical grid compared with virgin oil fuel.

Knowledge Graph Based Large Scale Network Security Threat Detection Techniques

Abstract This paper constructs a detection technique for large-scale network security threats based on a knowledge graph, extracts the attack features of network security threats using feature template FT, and combines the CNN layer, BiLSTM layer and CRF layer to establish FT-CNN-BiLSTM-CRF large-scale network security threat detection technique. Network security threat performance evaluation experiments and multi-step attack experiments have verified the detection capability of this paper's method. The recall rate of the method built in this paper in detecting malicious data is about 62.39%, the average F1-Score for normal and malicious traffic detection is 0.7482, and the anomaly score for normal traffic detection is almost 0. The detection performance of this paper's method for multi-step network attacks is superior to that of other methods, and it is capable of detecting malicious attacks quickly. Experiments have proved that the method constructed in this paper can meet the requirements of detection capability and efficiency in large-scale network security threats and has high feasibility and application value.

Experimental evaluation of fatigue performance of recycled asphalt mixture using refined separation recycled aggregates

In order to improve the utilization of reclaimed asphalt pavement (RAP) in recycled asphalt mixtures (RAMs), RAP was firstly separated into recycled mortar and coarse refined separation recycled aggregates (CRSRAs) with different particle sizes based on a laboratory refined separation method. Meanwhile, the process parameters for RAP separation were optimized by using the orthogonal test method. The related properties of the aggregates before and after refined separation, including residual bitumen binder content, particle agglomeration, angularity, and sphericity, were characterized. The cumulative distribution of sphericity of CRSRAs was fitted with the Weibull function. Asphalt mixtures with 0, 30%, and 50% RAP content were prepared by the Marshall design method. Moreover, the indirect tensile dynamic modulus and fatigue life of these asphalt mixtures after different water immersion periods were evaluated. The results showed that the CRSRAs had a residual binder content less than 1.0% and fewer pseudo particles after refined separation. Furthermore, a bitumen-rich mortar with a particle size less than 2.36 mm and a bitumen content of 10–14% was obtained. Angularity and Weibull fitting results verified that refined separation could homogenize RAP and obtain CRSRAs with a low variability, which helps to prepare better RAMs with high RAP content. RAMs prepared with CRSRAs exhibit superior fatigue performance compared to the traditional RAMs. Their fatigue life was less sensitive to stress changes and fluctuated little with the increasing water immersion time.

Experimental and numerical evaluation of mechanical performances of hybrid Palmyra-Palm-Leaf-Stalk/Glass fiber reinforced polyester composites

In the present work, four layered Palmyra-Palm-Leaf-Stalk (PPLSF)/Glass hybrid composites laminates with six distinct stacking orders were made by following hand lay-up method. Different experiments were carried out to determine the tensile, flexural, impact and hardness of the laminates as per ASTM standard. The results reveal that the addition of glass fiber to PPLSF significantly enhanced the mechanical performance of the laminates. On successive replacement of PPLSF by glass fiber, an enhancement of properties such as 291.39% for tensile strength, 72.08% for flexural strength, 75.82% for impact strength, and 78.44% for micro-hardness was observed. Again, it is also noticed that the stacking order plays an important role in influencing the properties while comparing experimental finding of laminates with equal weight percentages PPLSF and glass fiber. Further to perform Finite Element Analysis (FEA), a 3D CAD model of tensile and flexural specimen was created in space-claim modeler of ANSYS R19.0 and subsequently imported to generate numerical predicted values. While comparing the numerical and experimental results, there was very good agreement among the results was found. Scanning Electron Microscope (SEM) was used for fractography analysis of the tested samples to examine the different mode of failures.

Leg Mechanism Design and Motion Performance Analysis for an Amphibious Crab-like Robot

Bionic-legged robots draw inspiration from animal locomotion methods and structures, demonstrating the potential to traverse irregular and unstructured environments. The ability of Portunus trituberculatus (Portunus) to run flexibly and quickly in amphibious environments inspires the design of systems and locomotion methods for amphibious robots. This research describes an amphibious crab-like robot based on Portunus and designs a parallel leg mechanism for the robot based on biological observations. The research creates the group and sequential gait commonly used in multiped robots combined with the form of the robot’s leg mechanism arrangement. This research designed the parallel leg mechanism and modeled its dynamics. Utilizing the outcomes of the dynamics modeling, we calculate the force and torque exerted on each joint of the leg mechanism during group gait and sequential gait when the robot is moving with a load. This analysis aims to assess the performance of the robot’s motion. Finally, a series of performance evaluation experiments are conducted on land and underwater, which show that the amphibious crab-like robot has good walking performance. The crab-like robot can perform forward, backward, left, and right walking well using group and sequential gaits. Simultaneously, the crab-like robot showcases faster movement in group gaits and a more substantial load capacity in sequential gaits.

A vector projection similarity-based method for miRNA-disease association prediction

[S U M M A R Y] Many miRNA-disease association prediction models incorporate Gaussian interaction profile kernel similarity (GIPS). However, the GIPS fails to consider the specificity of the miRNA-disease association matrix, where matrix elements with a value of 0 represent miRNA and disease relationships that have not been discovered yet. To address this issue and better account for the impact of known and unknown miRNA-disease associations on similarity, we propose a method called vector projection similarity-based method for miRNA-disease association prediction (VPSMDA). In VPSMDA, we introduce three projection rules and combined with logistic functions for the miRNA-disease association matrix and propose a vector projection similarity measure for miRNAs and diseases. By integrating the vector projection similarity matrix with the original one, we obtain the improved miRNA and disease similarity matrix. Additionally, we construct a weight matrix using different numbers of neighbors to reduce the noise in the similarity matrix. In performance evaluation, both LOOCV and 5-fold CV experiments demonstrate that VPSMDA outperforms seven other state-of-the-art methods in AUC. Furthermore, in a case study, VPSMDA successfully predicted 10, 9, and 10 out of the top 10 associations for three important human diseases, respectively, and these predictions were confirmed by recent biomedical resources.

1 experimental and numerical evaluation of residual mechanical performance of carbon/epoxy laminated coupons after integration of solid battery cells for aeronautical applications

The present study deals with a numerical and experimental evaluation of the residual mechanical performance of a quasi-isotropic carbon/epoxy laminated composite plate in which solid battery cells have been integrated. Firstly, the mechanical properties of the multifunctional Reinforced Multilayer Stack (RMS) battery cells have been determined thanks to dedicated multi-instrumented tests, while those of the AS4/8552 composite material have been extracted from the literature and validated by comparisons with tensile tests on plates without battery, considered as a reference case. Then, on the basis of the recommendations obtained from a virtual test campaign using FE simulations, a 16-ply laminated composite plate containing 6 solid battery cells was manufactured with a modified curing cycle that preserves the electrical properties of the batteries while ensuring the mechanical properties of the composite material. Various control methods, such as CT-scan and optical analysis, were used to demonstrate the promising quality of the manufactured composite coupon with embedded cells. Finally, a mechanical tensile test was carried out to experimentally determine the reduction in mechanical properties. The analysis of such a test with non-linear FE simulation has allowed a better understanding of the damage and failure process when solid battery cells are integrated into high performance laminated composite structures designed for the aeronautical industry.

Experimental evaluation of cyclic performance of steel slag and grids of truck scrap tire as a novel reinforced ballast material

Introducing a novel ballast material based on reinforcing steel slag aggregates with strips of scrap tire can be beneficial due to decreasing the crushing and increasing the damping of the ballast materials. Moreover, recycling the two waste materials included in this novel ballast has a positive impact on environmental preservation. In this regard, the Waste Tire Reinforcement (WTR) consists of strips of truck scrap tire (385/80/R22.5) with a width of 70 mm utilized in grid shape (with various dimension grids of 21 ×7, 11.5 ×7, and 7 ×7 cm) as a reinforcement element. To evaluate the performance of suggested reinforcement with Steel Slag Ballast (SSB), a series of ballast box tests were performed up to 100,000 loading and unloading cycles with a magnitude and frequency of 30 kN and 3 Hz at three different levels of 5, 10, and 15 cm for WTR placement in ballast. The experimental results revealed that the WTR reinforcement with grids of 11.5 × 7 cm and a placement height of 5 cm (WTR11.5–5) has the best performance compared to other configurations. Compared to SSB without reinforcement, WTR11.5–5 may reduce breakage by more than 90% and increase the damaging ratio by 160%.

Cascading latent heat thermal energy storage in parabolic trough solar collector as a promising solution: An experimental investigation

Solar energy system can be considered as a reliable energy source if it connects to a latent heat thermal energy storage (LHTES) system using phase change materials (PCMs). To tackle the low thermal conductivity coefficient of PCMs, the cascaded latent heat storage (CLHS) configuration is a promising solution. This study aims to illustrate the improvement of CLHS setup comparing with non-cascaded arrangement, performing a whole-scale experimental investigation. A compact parabolic trough collector (PTC) unit equipped with a LHTES module is designed and constructed for a residential purpose. Experimental results reveal that the average thermal efficiency of PTC is 73.5 %. Three paraffin waxes with melting ranges of (64–68 ℃), (57–60 ℃) and (46–48 ℃) are injected in three same series of shell-tube heat exchangers. For comparing cascaded and non-cascaded PCM performance, three experimental series, namely (Exp.1, Exp.2 and Exp.3) are performed. The experimental findings show that the deployment of the CLHS configuration improves the latent heat storage of non-CLHS-Exps.1, 2, and 3 by more than 10, 47, and 5 %, respectively. In addition, CLHS system has a potential to increase the exergy storage of non-CLHS setup by up to 27.4 %. Furthermore, impact of increasing heat transfer fluid (HTF) flow rate is examined and it is observed that the melting process is shortened by almost 39 %. Accordingly, increasing flow rate improves energy stored in PCMs1, 2, and 3 by 60 %, 59.62 %, and 72.04 %, respectively. Consequently, the latent thermal exergy storage of the PCM experiences a twofold increase. Conducting an experimental evaluation of thermal performance of the storage module integrated with PTC setup can be valuable, particularly in the realm of residential applications.

Experimental investigation on comprehensive thermal-hydraulic performance of supercritical CO2 in a NACA 0020 airfoil fin printed circuit heat exchanger

Printed circuit heat exchanger (PCHE) is a promising compact heat exchanger for supercritical CO2 power conversion systems. Among the channel shapes of PCHE, an airfoil fin channel was considered for the high thermal-hydraulic performance of the PCHE. However, quantitative experimental evaluation of the thermal-hydraulic performance between the optimized airfoil fin and the wavy channel (commercially utilized channel shape) has not been performed. Therefore, in this study, to compare the performance according to channel shape, experiments on the optimized arrangement of airfoil fin PCHE (NACA 0020 airfoil fin with non-dimensional arrangement parameters, staggered number 1.0, horizontal number 2.0, and vertical number 1.8) and wavy PCHE (120° bending angle) were conducted. The experimental conditions of supercritical CO2 were temperatures ranging from 70°C (inlet) to 40.5°C (outlet) with pressures varying from 7.5 to 8.5 MPa. To reflect the real etched shape of airfoil fin channel, the real shape of airfoil fin was measured by using optical 3-Dimensional scanner. And hydraulic diameter was calculated based on measured information of real etched airfoil fin shape. The pressure drop of the airfoil fin PCHE was one-fifth of that of the wavy PCHE with the same total heat transfer rate per unit volume. New empirical correlations for the friction factor and Nusselt number of the airfoil fin PCHE were proposed based on the experimental data for the precooler design in the supercritical CO2 Brayton cycle. A friction factor correlation with maximum deviations of ±5.04 % and Nusselt number correlation with maximum deviations of ±14.0 % were proposed.

Time-domain heart rate variability features for automatic congestive heart failure prediction.

Heart failure is a serious condition that often goes undiagnosed in primary care due to the lack of reliable diagnostic tools and the similarity of its symptoms with other diseases. Non-invasive monitoring of heart rate variability (HRV), which reflects the activity of the autonomic nervous system, could offer a novel and accurate way to detect and manage heart failure patients. This study aimed to assess the feasibility of using machine learning techniques on HRV data as a non-invasive biomarker to classify healthy adults and those with heart failure. We used digitized electrocardiogram recordings from 54 adults with normal sinus rhythm and 44 adults categorized into New York Heart Association classes 1, 2, and 3, suffering from congestive heart failure. All recordings were sourced from the PhysioNet database. Following data pre-processing, we performed time-domain HRV analysis on all individual recordings, including root mean square of the successive difference in adjacent RR interval (RRi) (RMSSD), the standard deviation of RRi (SDNN, the NN stands for natural or sinus intervals), the standard deviation of the successive differences between successive RRi (SDSD), the number or percentage of RRi longer than 50ms (NN50 and pNN50), and the average value of RRi [mean RR interval (mRRi)]. In our experimental classification performance evaluation, on the computed HRV parameters, we optimized hyperparameters and performed five-fold cross-validation using four machine learning classification algorithms: support vector machine, k-nearest neighbour (KNN), naïve Bayes, and decision tree (DT). We evaluated the prediction accuracy of these models using performance criteria, namely, precision, recall, specificity, F1 score, and overall accuracy. For added insight, we also presented receiver operating characteristic (ROC) plots and area under the ROC curve (AUC) values. The overall best performance accuracy of 77% was achieved when KNN and DT were trained on computed HRV parameters with a 5min time window. KNN obtained an AUC of 0.77, while DT attained 0.78. Additionally, in the classification of severe congestive heart failure, KNN and DT had the best accuracy of 91%, with KNN achieving an AUC of 0.88 and DT obtaining 0.92. The results show that HRV can accurately predict severe congestive heart failure. The findings of this study could inform the use of machine learning approaches on non-invasive HRV, to screen congestive heart failure individuals in primary care.

Tongue image datasets and performance evaluation

With the development of computer medicine, data has become an extremely important resource, and the construction and research of dataset is an important means to promote the development and innovation of medicine. Because there are few tongue image datasets and most tongue body image datasets are in a non-public state, the research of medical related algorithms is hindered to some extent, especially in computer vision and medical artificial intelligence. The constructed tongue image data set provides a large number of tongue images and related clinical information, which can be used to train computer medical models such as tongue image detection and tongue segmentation. A Tongue dataset based on deep learning was established by selecting and labeling the collected tongue images manually. The datasets contain 2,021 tongue images, which are randomly divided into training sets, verification sets and test sets according to the proportion, and the dataset performance evaluation experiments are carried out on the existing algorithms. Testing on convolutional neural network models such as YOLO detection and tongue segmentation, the average accuracy of detection and segmentation Miou value can reach 97.0% and 99.95%, respectively. The experimental results show that the data set can obtain high accuracy in detection and segmentation algorithms, and can be further applied to related research in the medical field. Tongue dataset data set can be downloaded at the following url: https://github.com/xiaohuomiao12/datasets.git.

Enhanced CO2 capture potential of UiO-66-NH2 synthesized by sonochemical method: experimental findings and performance evaluation

The excessive release of greenhouse gases, especially carbon dioxide (CO2) pollution, has resulted in significant environmental problems all over the world. CO2 capture technologies offer a very effective means of combating global warming, climate change, and promoting sustainable economic growth. In this work, UiO-66-NH2 was synthesized by the novel sonochemical method in only one hour. This material was characterized through PXRD, FT-IR, FE-SEM, EDX, BET, and TGA methods. The CO2 capture potential of the presented material was investigated through the analysis of gas isotherms under varying pressure conditions, encompassing both low and high-pressure regions. Remarkably, this adsorbent manifested a notable augmentation in CO2 adsorption capacity (3.2 mmol/g), achieving an approximate enhancement of 0.9 mmol/g, when compared to conventional solvothermal techniques (2.3 mmol/g) at 25 °C and 1 bar. To accurately represent the experimental findings, three isotherm, and kinetic models were used to fit the experimental data in which the Langmuir model and the Elovich model exhibited the best fit with R2 values of 0.999 and 0.981, respectively. Isosteric heat evaluation showed values higher than 80 kJ/mol which indicates chemisorption between the adsorbent surface and the adsorbate. Furthermore, the selectivity of the adsorbent was examined using the Ideal Adsorbed Solution Theory (IAST), which showed a high value of 202 towards CO2 adsorption under simulated flue gas conditions. To evaluate the durability and performance of the material over consecutive adsorption–desorption processes, cyclic tests were conducted. Interestingly, these tests demonstrated only 0.6 mmol/g capacity decrease for sonochemical UiO-66-NH2 throughout 8 consecutive cycles.

Green-synthetized β-amino-α-carbethoxy ethyl acrylates as corrosion inhibitors for mild steel in acid media: Experimental performance evaluation and atomic/molecular-level modeling

Two β-amino acrylate derivatives were synthesized through a solvent- and heating-free mechanochemical procedure. Those substances were tested as corrosion inhibitors for mild steel in acidic environment (1 mol/L HCl). Weight Loss Study demonstrated that both molecules mitigate corrosion with 85 and 90 % efficiencies. Electrochemical experiments proved that the corrosion mechanism is charge transfer and that the presence of EOAB and DBMM at any tested concentration in the electrolyte significantly increases the polarization resistance and lowers the corrosion density current. This phenomenon is due to the adsorption of the molecules on the mild steel surface, leading to the formation of a protection film that prevents the corrosion reaction, as suggested by Atomic Force Microscopy topographic images. The self-consistent-charge density-functional tight-binding method (SCC-DFTB) revealed that their adsorption strength on the iron surface followed well the corrosion inhibition performance. Both inhibitor molecules formed covalent bonds with iron atoms as evidenced from stable adsorption geometries and projected density of states. Quantum chemical parameters seem to fail in predicting the experimental performance. Outcomes from the present study confirmed the fact that an additional functional group would lead to an increased inhibition performance is not a granted correlation.

Experimental Evaluation of the Mechanical Performance of Geo Polymer Concrete

Experimental Evaluation of the Mechanical Performance of Geo Polymer Concrete

Detection of backside coupled propagating surface plasmon resonance on the sidewall of a wafer

We proposed a surface plasmon resonance (SPR) sensor structure that utilized a glass wafer with a diffraction grating and an n-type silicon piece bonded near the SPR coupling site. This configuration enabled surface plasmon excitation from the back of the substrate without the unwanted interaction between the excitation light and the sample, and electrical detection of the SPR response by a 0.7-eV Schottky barrier at the Au/n-Si interface formed on the sidewall of the silicon piece was achieved. Experimental evaluation of the surface plasmon coupling performance was conducted, showing clear peaks in the photocurrent for various wavelengths in the NIR-II window, ranging from 1100 to 1300 nm. The device’s ability to detect propagating surface plasmons as a photocurrent was confirmed; the results indicated a consistent trend with theoretical and numerical calculations. Since the device was composed of a glass substrate, the use of wavelengths shorter than the near-infrared wavelength was possible, including visible wavelengths where the optical absorption by water is negligible. Thus, our proposed sensor provides a compact and efficient solution for SPR sensing in aqueous solutions.

Investigation of tribo-mechanical performance of alkali treated rice-husk and polypropylene-random-copolymer based biocomposites

This study was based on the experimental performance evaluation of a wood polymer composite (WPC) that was synthesized by incorporating untreated and treated rice husk (RH) fibers into a polypropylene random copolymer matrix. The submicron-scale RH fibers were alkali-treated to modify the surface and introduce new functional groups in the WPC. A compatibilizer (maleic anhydride) and a thermos-mechanical properties modifier (polypropylene grafted with 30 % glass fiber) were used in the WPC. The effects of untreated and treated RH on the WPC panels were studied using FESEM, FTIR, and microscope images. A pin-on-disk setup was used to investigate the bulk tribological properties of PPRC and WPC. The complex relationship between the friction coefficient of different loading of RH fibers in the WPC, as a function of sliding distance, was analyzed along with the temperature and morphology of the surface. It was observed that untreated RH acted as a friction modifier, while treated RH acted as a solid lubricant. Microhardness was calculated using the QCSM module on nanoindentation. It was found that untreated RH led to an increase in microhardness, while treated RH caused a decrease in hardness compared to PPRC.