Sensor Fault Diagnosis and Unknown Disturbances Estimation of High Switching Frequency Single-phase PWM Rectifier

International Journal of Control, Automation and Systems - The fault and disturbances estimation has an important role in the modern traction railway system. This paper proposes a unique method for...     

Effect of expandable graphite and ammonium polyphosphate on the thermal degradation and weathering of intumescent fire-retardant coating

This research aims to study the effect of ammonium polyphosphate and expandable graphite on the intumescent coating formulations (ICF). The coating presented in this research article is based on carbon source expandable graphite (EG), blowing agent melamine, acid source ammonium polyphosphate (APP), epoxy resin as a binder with polyamide amine. The stability of the developed coating was verified at 950°C for 1-hour fire test. The results showed that the coating is stable and well bond with the steel substrate. The char was characterized by using FESEM, XRD, FTIR, DTA, TGA, XPS, Py-GCMS and Weathering Test. The morphology of the char was studied by SEM of the coating after furnace fire test. XRD and FTIR show the presence of graphite, borophosphate; boron oxide and sassolite in the residual char. TGA and DTG disclosed that EG improved the residual mass of coating. XPS analysis showed the char residue of IF5-APP-EG contains carbon and oxygen contents 47.50 and 40.70, respectively. Py-GCMS analysis described that the IF5-APP-EG released less gaseous compounds. The weathering test illustrated that's the char expansion of coatings samples was decreased due to the presence of a humid environment and UV light. The IF5-APP-EG showed the maximum char expansion, lower substrate temperature and high residual weight among the studied formulations.     

Performance tuning of glass fiber/epoxy composites through interfacial modification upon integrating with dendrimer functionalized graphene oxide

In this study, glass fiber/epoxy composites were interfacially tailored by introducing polyamidoamine (PAM) dendrimer functionalized graphene oxide (GO) into epoxy matrix. Two different composites each containing varying loading fraction (0.5, 1.0, and 1.5 wt%) of GO and GO-PAM were fabricated via hot press processing. Composites were evaluated for interlaminar shear strength (ILSS), dynamic mechanical properties and thermal conductivity. The inclusion of 1.5 wt% GO-PAM resulted ~57.3%, ~42.7%, and ~54% enhancement in ILSS, storage modulus and thermal conductivity, respectively. Almost, ~71% reduction in coefficient of thermal expansion was also observed at same GO-PAM loading. Moreover, higher glass transition temperature was observed with GO-PAM addition. GO-PAM substantially improved fiber/matrix interfacial adhesion, which was witnessed through scanning electron microscopy. The enhanced thermo-mechanical performance was attributed to interfacial covalent interactions engendered by ring opening reaction between epoxy and amine moieties of PAM dendrimers. These multiscale composites with extraordinary functional properties can outperform conventional counterparts with improved reliability and performance.     

Your comments matter: incorporating viewers’ comments for ranking online video content using bibliometrics

ABSTRACT

The quality of user-generated content over World Wide Web media is a matter of serious concern for both creators and users. To measure the quality of content, webometric techniques are commonly used. In recent times, bibliometric techniques have been introduced to good effect for evaluation of the quality of user-generated content, which were originally used for scholarly data. However, the application of bibliometric techniques to evaluate the quality of YouTube content is limited to h-index and g-index considering only views. This paper advocates for and demonstrates the adaptation of existing Bibliometric indices including h-index, g-index and M-index exploiting both views and comments and proposes three indices h_vc, g_vc and m_vc for YouTube video channel ranking. The empirical results prove that the proposed indices using views along with the comments outperform the existing approaches on a real-world dataset of YouTube.     

Alpha Fusion Adversarial Attack Analysis Using Deep Learning

The deep learning model encompasses a powerful learning ability that integrates the feature extraction, and classification method to improve accuracy. Convolutional Neural Networks (CNN) perform well in machine learning and image processing tasks like segmentation, classification, detection, identification, etc. The CNN models are still sensitive to noise and attack. The smallest change in training images as in an adversarial attack can greatly decrease the accuracy of the CNN model. This paper presents an alpha fusion attack analysis and generates defense against adversarial attacks. The proposed work is divided into three phases: firstly, an MLSTM-based CNN classification model is developed for classifying COVID-CT images. Secondly, an alpha fusion attack is generated to fool the classification model. The alpha fusion attack is tested in the last phase on a modified LSTM-based CNN (CNN-MLSTM) model and other pre-trained models. The results of CNN models show that the accuracy of these models dropped greatly after the alpha-fusion attack. The highest F1 score before the attack was achieved is 97.45 And after the attack lowest F1 score recorded is 22%. Results elucidate the performance in terms of accuracy, precision, F1 score and Recall.     

Investigating the effect of varying block size on power and energy consumption of GPU kernels

The Journal of Supercomputing - Power consumption is likely to remain a significant concern for exascale performance in the foreseeable future. In addition, graphics processing units (GPUs) have...     

Dynamic and scalable multi-level trust management model for Social Internet of Things

The Internet of Things (IoT) is a paradigm that has made everyday objects intelligent by offering them the ability to connect to the Internet and communicate. Integrating the social component into IoT gave rise to the Social Internet of Things (SIoT), which has helped overcome various issues such as heterogeneity and navigability. In this kind of environment, participants compete to offer a variety of attractive services. Nevertheless, some of them resort to malicious behaviour to spread poor-quality services. They perform so-called Trust-Attacks and break the basic functionality of the system. Trust management mechanisms aim to counter these attacks and provide the user with an estimate of the trust degree they can place in other users, thus ensuring reliable and qualified exchanges and interactions. Several works in literature have interfered with this problem and have proposed different Trust-Models. The majority tried to adapt and reapply Trust-Models designed for common social networks or peer-to-peer ones. That is, despite the similarities between these types of networks, SIoT ones present specific peculiarities. In SIoT, users, devices and services are collaborating. Devices entities can present constrained computing and storage capabilities, and their number can reach some millions. The resulting network is complex, constrained and highly dynamic, and the attacks-implications can be more significant. In this paper, we propose DSL-STM a new dynamic and scalable multi-level Trust-Model, specifically designed for SIoT environments. We propose multidimensional metrics to describe and SIoT entities behaviours. The latter are aggregated via a Machine Learning-based method, allowing classifying users, detecting attack types and countering them. Finally, a hybrid propagation method is suggested to spread trust values in the network, while minimizing resource consumption and preserving scalability and dynamism. Experimentation made on various simulated scenarios allows us to prove the resilience and performance of DSL-STM.

     

The Double Edge Sword Based Distributed Executor Service

Scalability is one of the most important quality attribute of software-intensive systems, because it maintains an effective performance parallel to the large fluctuating and sometimes unpredictable workload. In order to achieve scalability, thread pool system (TPS) (which is also known as executor service) has been used extensively as a middleware service in software-intensive systems. TPS optimization is a challenging problem that determines the optimal size of thread pool dynamically on runtime. In case of distributed-TPS (DTPS), another issue is the load balancing b/w available set of TPSs running at backend servers. Existing DTPSs are overloaded either due to an inappropriate TPS optimization strategy at backend servers or improper load balancing scheme that cannot quickly recover an overload. Consequently, the performance of software-intensive system is suffered. Thus, in this paper, we propose a new DTPS that follows the collaborative round robin load balancing that has the effect of a double-edge sword. On the one hand, it effectively performs the load balancing (in case of overload situation) among available TPSs by a fast overload recovery procedure that decelerates the load on the overloaded TPSs up to their capacities and shifts the remaining load towards other gracefully running TPSs. And on the other hand, its robust load deceleration technique which is applied to an overloaded TPS sets an appropriate upper bound of thread pool size, because the pool size in each TPS is kept equal to the request rate on it, hence dynamically optimizes TPS. We evaluated the results of the proposed system against state of the art DTPSs by a client-server based simulator and found that our system outperformed by sustaining smaller response times.     

Trimetallic PtCoFe Alloy Monolayer Superlattices as Bifunctional Oxygen‐Reduction and Ethanol‐Oxidation Electrocatalysts

A synthesis strategy for the preparation of trimetallic PtCoFe alloy nanoparticle superlattices is reported. Trimetallic PtCoFe alloy monolayer array of nanoparticle superlattices with a large density of high index facets and platinum‐rich surface are successfully prepared by coreduction of metal precursors in formamide solvent. The concentration of cetyl trimethyl ammonium bromide plays a vital role for the formation of a monolayer array of nanoparticle superlattices, while the size of nanoparticles is determined by NaI. The prepared monolayer array of nanoparticle superlattices is the superior catalyst for oxygen reduction reaction as well as for ethanol oxidation owing to their specific structural and compositional characteristics.