Search results for: hyperparameter tuning

Authors: Naif Mohsen Alharbi

Abstract:

This paper present failure occurred on flexible coupling installed at oil anf gas operation. Also it presents maintenance ideas implemented on the flexible coupling installed to transmit high torque from gearbox to pump. Basically, the machine train is including steam turbine which drives the pump and there is gearbox located in between for speed reduction. investigation are identifying the root causes, solving and developing the technology designs or bad actor. This report provides the study intentionally for continues operation optimization, utilize the advanced opportunity and implement a improvement. Objective: The main objectives of the investigation are identifying the root causes, solving and developing the technology designs or bad actor. Ultimately, fulfilling the operation productivity, also ensuring better technology, quality and design by solutions. This report provides the study intentionally for continues operation optimization, utilize the advanced opportunity and implemet improvement. Method: The method used in this project was a very focused root cause analysis procedure that incorporated engineering analysis and measurements. The analysis method extensively covers the measuring of the complete coupling dimensions. Including the membranes thickness, hubs, bore diameter and total length, dismantle flexible coupling to diagnose how deep the coupling has been affected. Also, defining failure modes, so that the causes could be identified and verified. Moreover, Vibration analysis and metallurgy test. Lastly applying several solutions by advanced tools (will be mentioned in detail). Results and observation: Design capacity: Coupling capacity is an inadequate to fulfil 100% of operating conditions. Therefore, design modification of service factor to be at least 2.07 is crucial to address this issue and prevent recurrence of similar scenario, especially for the new upgrading project. Discharge fluctuation: High torque flexible coupling encountered during the operation. Therefore, discharge valve behaviour, tuning, set point and general conditions revaluated and modified subsequently, it can be used as baseline for upcoming Coupling design project. Metallurgy test: Material of flexible coupling membrane (discs) tested at the lab, for a detailed metallurgical investigation, better material grade has been selected for our operating conditions,

Keywords: high speed machine, reliabilty, flexible coupling, rotating equipment

Procedia PDF Downloads 38

Authors: Zunaira Zaman, David Bohannon

Abstract:

This study explores the application of machine learning (ML) for detecting security vulnerabilities in source code. The research aims to assist organizations with large application portfolios and limited security testing capabilities in prioritizing security activities. ML-based approaches offer benefits such as increased confidence scores, false positives and negatives tuning, and automated feedback. The initial approach using natural language processing techniques to extract features achieved 86% accuracy during the training phase but suffered from overfitting and performed poorly on unseen datasets during testing. To address these issues, the study proposes using the abstract syntax tree (AST) for Java and C++ codebases to capture code semantics and structure and generate path-context representations for each function. The Code2Vec model architecture is used to learn distributed representations of source code snippets for training a machine-learning classifier for vulnerability prediction. The study evaluates the performance of the proposed methodology using two datasets and compares the results with existing approaches. The Devign dataset yielded 60% accuracy in predicting vulnerable code snippets and helped resist overfitting, while the Juliet Test Suite predicted specific vulnerabilities such as OS-Command Injection, Cryptographic, and Cross-Site Scripting vulnerabilities. The Code2Vec model achieved 75% accuracy and a 98% recall rate in predicting OS-Command Injection vulnerabilities. The study concludes that even partial AST representations of source code can be useful for vulnerability prediction. The approach has the potential for automated intelligent analysis of source code, including vulnerability prediction on unseen source code. State-of-the-art models using natural language processing techniques and CNN models with ensemble modelling techniques did not generalize well on unseen data and faced overfitting issues. However, predicting vulnerabilities in source code using machine learning poses challenges such as high dimensionality and complexity of source code, imbalanced datasets, and identifying specific types of vulnerabilities. Future work will address these challenges and expand the scope of the research.

Keywords: code embeddings, neural networks, natural language processing, OS command injection, software security, code properties

Procedia PDF Downloads 74

Authors: Joni Asitashvili, Eka Chabashvili, Maya Virsaladze, Alexander Chokhonelidze

Abstract:

Studying the soundscape is a major challenge in many countries of the civilized world today. The sound environment and music itself are part of the Earth's ecosystem. Therefore, researching its positive or negative impact is important for a clean and healthy environment. The acoustics of nature gave people many musical ideas, and people enriched musical features and performance skills with the ability to imitate the surrounding sound. For example, a population surrounded by mountains invented the technique of antiphonal singing, which mimics the effect of an echo. Canadian composer Raymond Murray Schafer viewed the world as a kind of musical instrument with ever-renewing tuning. He coined the term "Soundscape" as a name of a natural environmental sound, including the sound field of the Earth. It can be said that from which the “music of nature” is constructed. In the 21st century, a new field–Ecomusicology–has emerged in the field of musical art to study the sound ecosystem and various issues related to it. Ecomusicology considers the interconnections between music, culture, and nature–According to the Aaron Allen. Eco-music is a field of ecomusicology concerning with the depiction and realization of practical processes using modern composition techniques. Finding an artificial sound source (instrumental or electronic) for the piece that will blend into the soundscape of Sound Oases. Creating a composition, which sounds in harmony with the vibrations of human, nature, environment, and micro- macrocosm as a whole; Currently, we are exploring the ambient sound of the Georgian urban and suburban environment to discover “Sound Oases" and compose Eco-music works. We called “Sound Oases" an environment with a specific sound of the ecosystem to use in the musical piece as an instrument. The most interesting examples of Eco-music are the round dances, which were already created in the BC era. In round dances people would feel the united energy. This urge to get united revealed itself in our age too, manifesting itself in a variety of social media. The virtual world, however, is not enough for a healthy interaction; we created plan of “contemporary round dance” in sound oasis, found during expedition in Georgian caves, where people interacted with cave's soundscape and eco-music, they feel each other sharing energy and listen to earth sound. This project could be considered a contemporary round dance, a long improvisation, particular type of art therapy, where everyone can participate in an artistic process. We would like to present research result of our eco-music experimental performance.

Keywords: eco-music, environment, sound, oasis

Procedia PDF Downloads 35

Authors: Emiliano Matta

Abstract:

Passive resonant vibration absorbers are among the most widely used dynamic control systems in civil engineering. They typically consist in a single-degree-of-freedom mechanical appendage of the main structure, tuned to one structural target mode through frequency and damping optimization. One classical scheme is the pendulum absorber, whose mass is constrained to move along a curved trajectory and is damped by viscous dashpots. Even though the principle is well known, the search for improved arrangements is still under way. In recent years this investigation inspired a type of bidirectional pendulum absorber (BPA), consisting of a mass constrained to move along an optimal three-dimensional (3D) concave surface. For such a BPA, the surface principal curvatures are designed to ensure a bidirectional tuning of the absorber to both principal modes of the main structure, while damping is produced either by horizontal viscous dashpots or by vertical friction dashpots, connecting the BPA to the main structure. In this paper, a variant of BPA is proposed, where damping originates from the variable tangential friction force which develops between the pendulum mass and the 3D surface as a result of a spatially-varying friction coefficient pattern. Namely, a friction coefficient is proposed that varies along the pendulum surface in proportion to the modulus of the 3D surface gradient. With such an assumption, the dissipative model of the absorber can be proven to be nonlinear homogeneous in the small displacement domain. The resulting homogeneous BPA (HBPA) has a fundamental advantage over conventional friction-type absorbers, because its equivalent damping ratio results independent on the amplitude of oscillations, and therefore its optimal performance does not depend on the excitation level. On the other hand, the HBPA is more compact than viscously damped BPAs because it does not need the installation of dampers. This paper presents the analytical model of the HBPA and an optimal methodology for its design. Numerical simulations of single- and multi-story building structures under wind and earthquake loads are presented to compare the HBPA with classical viscously damped BPAs. It is shown that the HBPA is a promising alternative to existing BPA types and that homogeneous tangential friction is an effective means to realize systems provided with amplitude-independent damping.

Keywords: amplitude-independent damping, homogeneous friction, pendulum nonlinear dynamics, structural control, vibration resonant absorbers

Procedia PDF Downloads 116

Authors: Andrei Bejan, Luminita Marin, Dalila Belei

Abstract:

Phenothiazine with electron-rich nitrogen and sulfur heteroatoms has a high electron-donating ability which promotes a good conjugation and therefore low band-gap with consequences upon charge carrier mobility improving and shifting of light emission in visible domain. Moreover, its non-planar butterfly conformation inhibits molecular aggregation and thus preserves quite well the fluorescence quantum yield in solid state compared to solution. Therefore phenothiazine and its derivatives are promising hole transport materials for use in organic electronic and optoelectronic devices as light emitting diodes, photovoltaic cells, integrated circuit sensors or driving circuits for large area display devices. The objective of this paper was to obtain a series of new phenothiazine derivatives by introduction of different electron withdrawing substituents as formyl, carboxyl and cyanoacryl units in order to create a push pull system which has potential to improve the electronic and optical properties. Bromine atom was used as electrono-donor moiety to extend furthermore the existing conjugation. The understudy compounds were structural characterized by FTIR and 1H-NMR spectroscopy and single crystal X-ray diffraction. Besides, the single crystal X-ray diffraction brought information regarding the supramolecular architecture of the compounds. Photophysical properties were monitored by UV-vis and photoluminescence spectroscopy, while the electrochemical behavior was established by cyclic voltammetry. The absorption maxima of the studied compounds vary in a large range (322-455 nm), reflecting the different electronic delocalization degree, depending by the substituent nature. In a similar manner, the emission spectra reveal different color of emitted light, a red shift being evident for the groups with higher electron withdrawing ability. The emitted light is pure and saturated for the compounds containing strong withdrawing formyl or cyanoacryl units and reach the highest quantum yield of 71% for the compound containing bromine and cyanoacrilic units. Electrochemical study show reversible oxidative and reduction processes for all the compounds and a close correlation of the HOMO-LUMO band gap with substituent nature. All these findings suggest the obtained compounds as promising materials for optoelectronic devices.

Keywords: electrochemical properties, phenothiazine derivatives, photoluminescence, quantum yield

Procedia PDF Downloads 302

Authors: Kunal Bhardwaj, Christine Browne

Abstract:

With the increase in global awareness of the environmental impacts of plastic-based products, there has been a massive drive to reduce our use of these products. Use of plastic-based substrates in electronic circuits has been a matter of concern recently. Plastics provide a very smooth and cheap surface for printing high-performance electronics due to their non-permeability to ink and easy mouldability. In this research, we explore the use of nano cellulose (NC) films in electronics as they provide an advantage of being 100% recyclable and eco-friendly. The main hindrance in the mass adoption of NC film as a substitute for plastic is its higher surface roughness which leads to ink penetration, and dispersion in the channels on the film. This research was conducted to tune the RMS roughness of NC films to a range where they can replace plastics in electronics(310-470nm). We studied the dependence of the surface roughness of the NC film on the following tunable aspects: 1) composition by weight of the NC suspension that is sprayed on a silicon wafer 2) the width and the depth of the channels on the silicon wafer used as a base. Various silicon wafers with channel depths ranging from 6 to 18 um and channel widths ranging from 5 to 500um were used as a base. Spray coating method for NC film production was used and two solutions namely, 1.5wt% NC and a 50-50 NC-CNC (cellulose nanocrystal) mixture in distilled water, were sprayed through a Wagner sprayer system model 117 at an angle of 90 degrees. The silicon wafer was kept on a conveyor moving at a velocity of 1.3+-0.1 cm/sec. Once the suspension was uniformly sprayed, the mould was left to dry in an oven at 50°C overnight. The images of the films were taken with the help of an optical profilometer, Olympus OLS 5000. These images were converted into a ‘.lext’ format and analyzed using Gwyddion, a data and image analysis software. Lowest measured RMS roughness of 291nm was with a 50-50 CNC-NC mixture, sprayed on a silicon wafer with a channel width of 5 µm and a channel depth of 12 µm. Surface roughness values of 320+-17nm were achieved at lower (5 to 10 µm) channel widths on a silicon wafer. This research opened the possibility of the usage of 100% recyclable NC films with an additive (50% CNC) in high-performance electronics. Possibility of using additives like Carboxymethyl Cellulose (CMC) is also being explored due to the hypothesis that CMC would reduce friction amongst fibers, which in turn would lead to better conformations amongst the NC fibers. CMC addition would thus be able to help tune the surface roughness of the NC film to an even greater extent in future.

Keywords: nano cellulose films, electronic circuits, nanocrystals and surface roughness

Procedia PDF Downloads 97

Authors: Rajan Goyal, S. Lamba, S. Annapoorni

Abstract:

The indirect exchange coupled system consists of two ferromagnetic layers separated by non-magnetic spacer layer. The type of exchange coupling may be either ferro or anti-ferro depending on the thickness of the spacer layer. In the present work, the strength of exchange coupling in FePt/Al₂O₃/Fe₃Pt has been investigated by varying the thickness of the spacer layer Al₂O₃. The FePt/Al₂O₃/Fe₃Pt trilayer structure is fabricated on Si <100> single crystal substrate using sputtering technique. The thickness of FePt and Fe₃Pt is fixed at 60 nm and 2 nm respectively. The thickness of spacer layer Al₂O₃ was varied from 0 to 16 nm. The normalized hysteresis loops recorded at room temperature both in the in-plane and out of plane configuration reveals that the orientation of easy axis lies along the plane of the film. It is observed that the hysteresis loop for ts=0 nm does not exhibit any knee around H=0 indicating that the hard FePt layer and soft Fe₃Pt layer are strongly exchange coupled. However, the insertion of Al₂O₃ spacer layer of thickness ts = 0.7 nm results in appearance of a minor knee around H=0 suggesting the weakening of exchange coupling between FePt and Fe₃Pt. The disappearance of knee in hysteresis loop with further increase in thickness of the spacer layer up to 8 nm predicts the co-existence of ferromagnetic (FM) and antiferromagnetic (AFM) exchange interaction between FePt and Fe₃Pt. In addition to this, the out of plane hysteresis loop also shows an asymmetry around H=0. The exchange field Hex = (Hc↑-HC↓)/2, where Hc↑ and Hc↓ are the coercivity estimated from lower and upper branch of hysteresis loop, increases from ~ 150 Oe to ~ 700 Oe respectively. This behavior may be attributed to the uncompensated moments in the hard FePt layer and soft Fe₃Pt layer at the interface. A better insight into the variation in indirect exchange coupling has been investigated using recoil curves. It is observed that the almost closed recoil curves are obtained for ts= 0 nm up to a reverse field of ~ 5 kOe. On the other hand, the appearance of appreciable open recoil curves at lower reverse field ~ 4 kOe for ts = 0.7 nm indicates that uncoupled soft phase undergoes irreversible magnetization reversal at lower reverse field suggesting the weakening of exchange coupling. The openness of recoil curves decreases with increase in thickness of the spacer layer up to 8 nm. This behavior may be attributed to the competition between FM and AFM exchange interactions. The FM exchange coupling between FePt and Fe₃Pt due to porous nature of Al₂O₃ decreases much slower than the weak AFM coupling due to interaction between Fe ions of FePt and Fe₃Pt via O ions of Al₂O₃. The hysteresis loop has been simulated using Monte Carlo based on Metropolis algorithm to investigate the variation in strength of exchange coupling in FePt/Al₂O₃/Fe₃Pt trilayer system.

Keywords: indirect exchange coupling, MH loop, Monte Carlo simulation, recoil curve

Procedia PDF Downloads 165

Authors: Mehrnaz Mostafavi, Jalaledin Ghanavi

Abstract:

This research investigates the production of nanoring structures through a chemical reduction approach, exploring gradual reduction processes assisted by reductant agents, leading to the formation of these specialized nanorings. The study focuses on the controlled reduction of metal atoms within these agents, crucial for shaping these nanoring structures over time. The paper commences by highlighting the wide-ranging applications of metal nanostructures across fields like Nanomedicine, Nanobiotechnology, and advanced spectroscopy methods such as Surface Enhanced Raman Spectroscopy (SERS) and Surface Enhanced Infrared Absorption Spectroscopy (SEIRA). Particularly, gold nanoparticles, especially in the nanoring configuration, have gained significant attention due to their distinctive properties, offering accessible spaces suitable for sensing and spectroscopic applications. The methodology involves utilizing human serum albumin as a reducing agent to create gold nanoparticles through a chemical reduction process. This process involves the transfer of electrons from albumin's carboxylic groups, converting them into carbonyl, while AuCl4− acquires electrons to form gold nanoparticles. Various characterization techniques like Ultraviolet–visible spectroscopy (UV-Vis), Atomic-force microscopy (AFM), and Transmission electron microscopy (TEM) were employed to examine and validate the creation and properties of the gold nanoparticles and nanorings. The findings suggest that precise and gradual reduction processes, in conjunction with optimal pH conditions, play a pivotal role in generating nanoring structures. Experiments manipulating optical properties revealed distinct responses in the visible and infrared spectrums, demonstrating the tunability of these nanorings. Detailed examinations of the morphology confirmed the formation of gold nanorings, elucidating their size, distribution, and structural characteristics. These nanorings, characterized by an empty volume enclosed by uniform walls, exhibit promising potential in the realms of Nanomedicine and Nanobiotechnology. In summary, this study presents a chemical synthesis approach using organic reducing agents to produce gold nanorings. The results underscore the significance of controlled and gradual reduction processes in crafting nanoring structures with unique optical traits, offering considerable value across diverse nanotechnological applications.

Keywords: nanoring structures, chemical reduction approach, gold nanoparticles, spectroscopy methods, nano medicine applications

Procedia PDF Downloads 71

Authors: Wael M. Rabeh, Cesar Carrasco-Lopez, Juliana C. Ferreira, Pance Naumov

Abstract:

Bioluminescence, the emission of light from a biological process, is found in various living organisms including bacteria, fireflies, beetles, fungus and different marine organisms. Luciferase is an enzyme that catalyzes a two steps oxidation of luciferin in the presence of Mg2+ and ATP to produce oxyluciferin and releases energy in the form of light. The luciferase assay is used in biological research and clinical applications for in vivo imaging, cell proliferation, and protein folding and secretion analysis. The luciferase enzyme consists of two domains, a large N-terminal domain (1-436 residues) that is connected to a small C-terminal domain (440-544) by a flexible loop that functions as a hinge for opening and closing the active site. The two domains are separated by a large cleft housing the active site that closes after binding the substrates, luciferin and ATP. Even though all insect luciferases catalyze the same chemical reaction and share 50% to 90% sequence homology and high structural similarity, they emit light of different colors from green at 560nm to red at 640 nm. Currently, the majority of the structural and biochemical studies have been conducted on green-emitting firefly luciferases. To address the color emission mechanism, we expressed and purified two luciferase enzymes with blue-shifted green and red emission from indigenous Brazilian species Amydetes fanestratus and Phrixothrix, respectively. The two enzymes naturally emit light of different colors and they are an excellent system to study the color-emission mechanism of luciferases, as the current proposed mechanisms are based on mutagenesis studies. Using a vapor-diffusion method and a high-throughput approach, we crystallized and solved the crystal structure of both enzymes, at 1.7 Å and 3.1 Å resolution respectively, using X-ray crystallography. The free enzyme adopted two open conformations in the crystallographic unit cell that are different from the previously characterized firefly luciferase. The blue-shifted green luciferase crystalized as a monomer similar to other luciferases reported in literature, while the red luciferases crystalized as an octamer and was also purified as an octomer in solution. The octomer conformation is the first of its kind for any insect’s luciferase, which might be relate to the red color emission. Structurally designed mutations confirmed the importance of the transition between the open and close conformations in the fine-tuning of the color and the characterization of other interesting mutants is underway.

Keywords: bioluminescence, enzymology, structural biology, x-ray crystallography

Procedia PDF Downloads 301

Authors: Schröde C., Rodriguez D., Sánchez A., Abdul Malak, Churchill J., Boksmati T., Alharbi, Alsulmi H., Maghrabi S., Mowalad, Mutwalli R., Abualnaja Y.

Abstract:

Formulating a strategy for sustainable development of the Kingdom of Saudi Arabia’s coastal and marine environment is at the core of the “Marine and Coastal Protection Assessment Study for the Kingdom of Saudi Arabia Coastline (MCEP)”; that was set up in the context of the Vision 2030 by the Saudi Arabian government and aimed at providing a first comprehensive ‘Status Quo Assessment’ of the Kingdom’s marine environment to inform a sustainable development strategy and serve as a baseline assessment for future monitoring activities. This baseline assessment relied on scientific evidence of the drivers, pressures and their impact on the environments of the Red Sea and Arabian Gulf. A key element of the assessment was the cumulative pressure hotspot analysis developed for both national waters of the Kingdom following the principles of the Driver-Pressure-State-Impact-Response (DPSIR) framework and using the cumulative pressure and impact assessment methodology. The ultimate goals of the analysis were to map and assess the main hotspots of environmental pressures, and identify priority areas for further field surveillance and for urgent management actions. The study identified maritime transport, fisheries, aquaculture, oil, gas, energy, coastal industry, coastal and maritime tourism, and urban development as the main drivers of pollution in the Saudi Arabian marine waters. For each of these drivers, pressure indicators were defined to spatially assess the potential influence of the drivers on the coastal and marine environment. A list of hotspots of 90 locations could be identified based on the assessment. Spatially grouped the list could be reduced to come up with of 10 hotspot areas, two in the Arabian Gulf, 8 in the Red Sea. The hotspot mapping revealed clear spatial patterns of drivers, pressures and hotspots within the marine environment of waters under KSA’s maritime jurisdiction in the Red Sea and Arabian Gulf. The cascading assessment approach based on the DPSIR framework ensured that the root causes of the hotspot patterns, i.e. the human activities and other drivers, can be identified. The adapted CPIA methodology allowed for the combination of the available data to spatially assess the cumulative pressure in a consistent manner, and to identify the most critical hotspots by determining the overlap of cumulative pressure with areas of sensitive biodiversity. Further improvements are expected by enhancing the data sources of drivers and pressure indicators, fine-tuning the decay factors and distances of the pressure indicators, as well as including trans-boundary pressures across the regional seas.

Keywords: Arabian Gulf, DPSIR, hotspot, red sea

Procedia PDF Downloads 100

Authors: Ahmad Haidar

Abstract:

The rapid adoption of Artificial Intelligence (AI) technology holds unforeseen risks like privacy violation, unemployment, and algorithmic bias, triggering research institutions, governments, and companies to develop principles of AI ethics. The extensive and diverse literature on AI lacks an analysis of the evolution of principles developed in recent years. There are two fundamental purposes of this paper. The first is to provide insights into how the principles of AI ethics have been changed recently, including concepts like risk management and public participation. In doing so, a NOISE (Needs, Opportunities, Improvements, Strengths, & Exceptions) analysis will be presented. Second, offering a framework for building Ethical AI linked to sustainability. This research adopts an explorative approach, more specifically, an inductive approach to address the theoretical gap. Consequently, this paper tracks the different efforts to have “trustworthy AI” and “ethical AI,” concluding a list of 12 documents released from 2017 to 2022. The analysis of this list unifies the different approaches toward trustworthy AI in two steps. First, splitting the principles into two categories, technical and net benefit, and second, testing the frequency of each principle, providing the different technical principles that may be useful for stakeholders considering the lifecycle of AI, or what is known as sustainable AI. Sustainable AI is the third wave of AI ethics and a movement to drive change throughout the entire lifecycle of AI products (i.e., idea generation, training, re-tuning, implementation, and governance) in the direction of greater ecological integrity and social fairness. In this vein, results suggest transparency, privacy, fairness, safety, autonomy, and accountability as recommended technical principles to include in the lifecycle of AI. Another contribution is to capture the different basis that aid the process of AI for sustainability (e.g., towards sustainable development goals). The results indicate data governance, do no harm, human well-being, and risk management as crucial AI for sustainability principles. This study’s last contribution clarifies how the principles evolved. To illustrate, in 2018, the Montreal declaration mentioned eight principles well-being, autonomy, privacy, solidarity, democratic participation, equity, and diversity. In 2021, notions emerged from the European Commission proposal, including public trust, public participation, scientific integrity, risk assessment, flexibility, benefit and cost, and interagency coordination. The study design will strengthen the validity of previous studies. Yet, we advance knowledge in trustworthy AI by considering recent documents, linking principles with sustainable AI and AI for sustainability, and shedding light on the evolution of guidelines over time.

Keywords: artificial intelligence, AI for sustainability, declarations, framework, regulations, risks, sustainable AI

Procedia PDF Downloads 65

Authors: Sh. Sousani, Z. Shadrokh, M. Hofbauerová, J. Kollár, M. Jergel, P. Nádaždy, M. Omastová, E. Majková

Abstract:

Perovskite quantum dots (PQDs) have the potential to increase the performance of the perovskite solar cell (PSCs). The integration of PQDs into PSCs can extend the absorption range and enhance photon harvesting and device efficiency. In addition, PQDs can stabilize the device structure by passivating surface defects and traps in the perovskite layer and enhance its stability. The integration of PQDs into PSCs is strongly affected by the type of ligands on the surface of PQDs. The ligands affect the charge transport properties of PQDs, as well as the formation of well-defined interfaces and stability of PSCs. In this work, the CsPbBr₃ QDs were synthesized by the conventional hot-injection method using cesium oleate, PbBr₂ and two different ligands, namely oleic acid (OA) oleylamine (OAm) and didodecyldimethylammonium bromide (DDAB). The STEM confirmed regular shape and relatively monodisperse cubic structure with an average size of about 10-14 nm of the prepared CsPbBr₃ QDs. Further, the photoluminescent (PL) properties of the PQDs/perovskite bilayer with the ligand OA, OAm and DDAB were studied. For this purpose, ITO/PQDs as well as ITO/PQDs/MAPI perovskite structures were prepared by spin coating and the effect of the ligand and oxygen plasma treatment was analyzed. The plasma treatment of the PQDs layer could be beneficial for the deposition of the MAPI perovskite layer and the formation of a well-defined PQDs/MAPI interface. The absorption edge in UV-Vis absorption spectra for OA, OAm CsPbBr₃ QDs is placed around 513 nm (the band gap 2.38 eV); for DDAB CsPbBr₃ QDs, it is located at 490 nm (the band gap 2.33 eV). The photoluminescence (PL) spectra of CsPbBr₃ QDs show two peaks located around 514 nm (503 nm) and 718 nm (708 nm) for OA, OAm (DDAB). The peak around 500 nm corresponds to the PL of PQDs, and the peak close to 710 nm belongs to the surface states of PQDs for both types of ligands. These surface states are strongly affected by the O₃ plasma treatment. For PQDs with DDAB ligand, the O₃ exposure (5, 10, 15 s) results in the blue shift of the PQDs peak and a non-monotonous change of the amplitude of the surface states' peak. For OA, OAm ligand, the O₃ exposition did not cause any shift of the PQDs peak, and the intensity of the PL peak related to the surface states is lower by one order of magnitude in comparison with DDAB, being affected by O₃ plasma treatment. The PL results indicate the possibility of tuning the position of the PL maximum by the ligand of the PQDs. Similar behavior of the PQDs layer was observed for the ITO/QDs/MAPI samples, where an additional strong PL peak at 770 nm coming from the perovskite layer was observed; for the sample with PQDs with DDAB ligands, a small blue shift of the perovskite PL maximum was observed independently of the plasma treatment. These results suggest the possibility of affecting the PL maximum position and the surface states of the PQDs by the combination of a suitable ligand and the O₃ plasma treatment.

Keywords: perovskite quantum dots, photoluminescence, O₃ plasma., Perovskite Solar Cells

Procedia PDF Downloads 34

Authors: Sh. Sousani, Z. Shadrokh, M. Hofbauerová, J. Kollár, M. Jergel, P. Nádaždy, M. Omastová, E. Majková

Abstract:

Perovskite quantum dots (PQDs) have the potential to increase the performance of the perovskite solar cells (PSCs). The integration of PQDs into PSCs can extend the absorption range and enhance photon harvesting and device efficiency. In addition, PQDs can stabilize the device structure by passivating surface defects and traps in the perovskite layer and enhance its stability. The integration of PQDs into PSCs is strongly affected by the type of ligands on the surface of PQDs. The ligands affect the charge transport properties of PQDs, as well as the formation of well-defined interfaces and stability of PSCs. In this work, the CsPbBr₃ QDs were synthesized by the conventional hot-injection method using cesium oleate, PbBr₂, and two different ligands, namely oleic acid (OA)@oleylamine (OAm) and didodecyldimethylammonium bromide (DDAB). The STEM confirmed regular shape and relatively monodisperse cubic structure with an average size of about 10-14 nm of the prepared CsPbBr₃ QDs. Further, the photoluminescent (PL) properties of the PQDs/perovskite bilayer with the ligand OA@OAm and DDAB were studied. For this purpose, ITO/PQDs, as well as ITO/PQDs/MAPI perovskite structures, were prepared by spin coating, and the effect of the ligand and oxygen plasma treatment was analysed. The plasma treatment of the PQDs layer could be beneficial for the deposition of the MAPI perovskite layer and the formation of a well-defined PQDs/MAPI interface. The absorption edge in UV-Vis absorption spectra for OA@OAm CsPbBr₃ QDs is placed around 513 nm (the band gap 2.38 eV); for DDAB CsPbBr₃ QDs, it is located at 490 nm (the band gap 2.33 eV). The photoluminescence (PL) spectra of CsPbBr₃ QDs show two peaks located around 514 nm (503 nm) and 718 nm (708 nm) for OA@OAm (DDAB). The peak around 500 nm corresponds to the PL of PQDs, and the peak close to 710 nm belongs to the surface states of PQDs for both types of ligands. These surface states are strongly affected by the O₃ plasma treatment. For PQDs with DDAB ligand, the O₃ exposure (5, 10, 15 s) results in the blue shift of the PQDs peak and a non-monotonous change of the amplitude of the surface states' peak. For OA@OAm ligand, the O₃ exposition did not cause any shift of the PQDs peak, and the intensity of the PL peak related to the surface states is lower by one order of magnitude in comparison with DDAB, being affected by O₃ plasma treatment. The PL results indicate the possibility of tuning the position of the PL maximum by the ligand of the PQDs. Similar behaviour of the PQDs layer was observed for the ITO/QDs/MAPI samples, where an additional strong PL peak at 770 nm coming from the perovskite layer was observed; for the sample with PQDs with DDAB ligands, a small blue shift of the perovskite PL maximum was observed independently of the plasma treatment. These results suggest the possibility of affecting the PL maximum position and the surface states of the PQDs by the combination of a suitable ligand and the O₃ plasma treatment.

Keywords: perovskite quantum dots, photoluminescence, O₃ plasma., perovskite solar cells

Procedia PDF Downloads 35

Authors: Shayan Mohajer Hamidi

Abstract:

Federated learning has emerged as a promising approach for distributed model training while preserving data privacy. However, the challenges of communication overhead, limited network resources, and slow convergence hinder its widespread adoption. On the other hand, knowledge distillation has shown great potential in compressing large models into smaller ones without significant loss in performance. In this paper, we propose an innovative framework that combines federated learning and knowledge distillation to address these challenges and enhance the efficiency of distributed learning. Our approach, called Federated Knowledge Distillation (FKD), enables multiple clients in a federated learning setting to collaboratively distill knowledge from a teacher model. By leveraging the collaborative nature of federated learning, FKD aims to improve model compression while maintaining privacy. The proposed framework utilizes a coded teacher model that acts as a reference for distilling knowledge to the client models. To demonstrate the effectiveness of FKD, we conduct extensive experiments on various datasets and models. We compare FKD with baseline federated learning methods and standalone knowledge distillation techniques. The results show that FKD achieves superior model compression, faster convergence, and improved performance compared to traditional federated learning approaches. Furthermore, FKD effectively preserves privacy by ensuring that sensitive data remains on the client devices and only distilled knowledge is shared during the training process. In our experiments, we explore different knowledge transfer methods within the FKD framework, including Fine-Tuning (FT), FitNet, Correlation Congruence (CC), Similarity-Preserving (SP), and Relational Knowledge Distillation (RKD). We analyze the impact of these methods on model compression and convergence speed, shedding light on the trade-offs between size reduction and performance. Moreover, we address the challenges of communication efficiency and network resource utilization in federated learning by leveraging the knowledge distillation process. FKD reduces the amount of data transmitted across the network, minimizing communication overhead and improving resource utilization. This makes FKD particularly suitable for resource-constrained environments such as edge computing and IoT devices. The proposed FKD framework opens up new avenues for collaborative and privacy-preserving distributed learning. By combining the strengths of federated learning and knowledge distillation, it offers an efficient solution for model compression and convergence speed enhancement. Future research can explore further extensions and optimizations of FKD, as well as its applications in domains such as healthcare, finance, and smart cities, where privacy and distributed learning are of paramount importance.

Keywords: federated learning, knowledge distillation, knowledge transfer, deep learning

Procedia PDF Downloads 42

Authors: Mehmet Sinan Erguven

Abstract:

A number of authorities argue that radio is an outdated medium for advertising and does not have the same impact on consumers as it did in the past. This grim outlook on the future of radio has its basis in the audio-visual world that consumers now live in and the popularity of Internet-based marketing tools among advertising professionals. Nonetheless, consumers still appear to overwhelmingly prefer radio as an entertainment tool. Today, in Canada, 90% of all adults (18+) tune into the radio on a weekly basis, and they listen for 17 hours. Teens are the most challenging group for radio to capture as an audience, but still, almost 75% tune in weekly. One online radio station reaches more than 250 million registered listeners worldwide, and revenues from radio advertising in Australia are expected to grow at an annual rate of 3% for the foreseeable future. Radio is also starting to become popular again in Turkey, with a 5% increase in the listening rates compared to 2014. A major matter of concern always affecting radio advertising is creativity. As radio generally serves as a background medium for listeners, the creativity of the radio commercials is important in terms of attracting the attention of the listener and directing their focus on the advertising message. This cannot simply be done by using audio tools like sound effects and jingles. This study aims to identify the creative elements (execution formats appeals and approaches) and creativity factors of radio commercials in Turkey. As part of the study, all of the award winning radio commercials produced throughout the history of the Kristal Elma Advertising Festival were analyzed using the content analysis technique. Two judges (an advertising agency copywriter and an academic) coded the commercials. The reliability was measured according to the proportional agreement. The results showed that sound effects, jingles, testimonials, slices of life and announcements were the most common execution formats in creative Turkish radio ads. Humor and excitement were the most commonly used creative appeals while award-winning ads featured various approaches, such as surprise musical performances, audio wallpaper, product voice, and theater of the mind. Some ads, however, were found to not contain any creativity factors. In order to be accepted as creative, an ad must have at least one divergence factor, such as originality, flexibility, unusual/empathic perspective, and provocative questions. These findings, as well as others from the study, hold great value for the history of creative radio advertising in Turkey. Today, the nature of radio and its listeners is changing. As more and more people are tuning into online radio channels, brands will need to focus more on this relatively cheap advertising medium in the very near future. This new development will require that advertising agencies focus their attention on creativity in order to produce radio commercials for their customers that will differentiate them from their competitors.

Keywords: advertising, creativity, radio, Turkey

Procedia PDF Downloads 353

Authors: Shayla He

Abstract:

Background and Purpose: According to Chamie (2017), it’s estimated that no less than 150 million people, or about 2 percent of the world’s population, are homeless. The homeless population in the United States has grown rapidly in the past four decades. In New York City, the sheltered homeless population has increased from 12,830 in 1983 to 62,679 in 2020. Knowing the trend on the homeless population is crucial at helping the states and the cities make affordable housing plans, and other community service plans ahead of time to better prepare for the situation. This study utilized the data from New York City, examined the key factors associated with the homelessness, and developed systematic modeling to predict homeless populations of the future. Using the best model developed, named HP-RNN, an analysis on the homeless population change during the months of 2020 and 2021, which were impacted by the COVID-19 pandemic, was conducted. Moreover, HP-RNN was tested on the data from Seattle. Methods: The methodology involves four phases in developing robust prediction methods. Phase 1 gathered and analyzed raw data of homeless population and demographic conditions from five urban centers. Phase 2 identified the key factors that contribute to the rate of homelessness. In Phase 3, three models were built using Linear Regression, Random Forest, and Recurrent Neural Network (RNN), respectively, to predict the future trend of society's homeless population. Each model was trained and tuned based on the dataset from New York City for its accuracy measured by Mean Squared Error (MSE). In Phase 4, the final phase, the best model from Phase 3 was evaluated using the data from Seattle that was not part of the model training and tuning process in Phase 3. Results: Compared to the Linear Regression based model used by HUD et al (2019), HP-RNN significantly improved the prediction metrics of Coefficient of Determination (R2) from -11.73 to 0.88 and MSE by 99%. HP-RNN was then validated on the data from Seattle, WA, which showed a peak %error of 14.5% between the actual and the predicted count. Finally, the modeling results were collected to predict the trend during the COVID-19 pandemic. It shows a good correlation between the actual and the predicted homeless population, with the peak %error less than 8.6%. Conclusions and Implications: This work is the first work to apply RNN to model the time series of the homeless related data. The Model shows a close correlation between the actual and the predicted homeless population. There are two major implications of this result. First, the model can be used to predict the homeless population for the next several years, and the prediction can help the states and the cities plan ahead on affordable housing allocation and other community service to better prepare for the future. Moreover, this prediction can serve as a reference to policy makers and legislators as they seek to make changes that may impact the factors closely associated with the future homeless population trend.

Keywords: homeless, prediction, model, RNN

Procedia PDF Downloads 92

Authors: Andrey A. Lipovskii, Alexey V. Redkov, Vyacheslav V. Rusan, Dmitry K. Tagantsev, Valentina V. Zhurikhina

Abstract:

The presented research is related to the development of recently proposed technique of the formation of composite materials, like optical glass-ceramics, with predetermined structure and properties of the crystalline component. The technique is based on the control of the size and concentration of the crystalline grains using the phenomenon of glass-ceramics decrystallization (vitrification) induced by ion-exchange. This phenomenon was discovered and explained in the beginning of the 2000s, while related theoretical description was given in 2016 only. In general, the developed theory enables one to model the process and optimize the conditions of ion-exchange processing of glass-ceramics, which provide given properties of crystalline component, in particular, profile of the average size of the crystalline grains. The optimization is possible if one knows two dimensionless parameters of the theoretical model. One of them (β) is the value which is directly related to the solubility of crystalline component of the glass-ceramics in the glass matrix, and another (γ) is equal to the ratio of characteristic times of ion-exchange diffusion and crystalline grain dissolution. The presented study is dedicated to the development of experimental technique and simulation which allow determining these parameters. It is shown that these parameters can be deduced from the data on the space distributions of diffusant concentrations and average size of crystalline grains in the glass-ceramics samples subjected to ion-exchange treatment. Measurements at least at two temperatures and two processing times at each temperature are necessary. The composite material used was a silica-based glass-ceramics with crystalline grains of Li2OSiO2. Cubical samples of the glass-ceramics (6x6x6 mm3) underwent the ion exchange process in NaNO3 salt melt at 520 oC (for 16 and 48 h), 540 oC (for 8 and 24 h), 560 oC (for 4 and 12 h), and 580 oC (for 2 and 8 h). The ion exchange processing resulted in the glass-ceramics vitrification in the subsurface layers where ion-exchange diffusion took place. Slabs about 1 mm thick were cut from the central part of the samples and their big facets were polished. These slabs were used to find profiles of diffusant concentrations and average size of the crystalline grains. The concentration profiles were determined from refractive index profiles measured with Max-Zender interferometer, and profiles of the average size of the crystalline grains were determined with micro-Raman spectroscopy. Numerical simulation were based on the developed theoretical model of the glass-ceramics decrystallization induced by ion exchange. The simulation of the processes was carried out for different values of β and γ parameters under all above-mentioned ion exchange conditions. As a result, the temperature dependences of the parameters, which provided a reliable coincidence of the simulation and experimental data, were found. This ensured the adequate modeling of the process of the glass-ceramics decrystallization in 520-580 oC temperature interval. Developed approach provides a powerful tool for fine tuning of the glass-ceramics structure, namely, concentration and average size of crystalline grains.

Keywords: diffusion, glass-ceramics, ion exchange, vitrification

Procedia PDF Downloads 244

Authors: Roy B. Clariana, Ryan L. Solnosky

Abstract:

Structural feedback is any form of feedback that aims to improve the quality of students’ domain-normative conceptual interrelationships. Research with structural feedback points to the potential mediating role of network graphs as feedback for tuning students’ conceptual understanding; for example, improved content knowledge and motivation were observed for undergraduate students who accessed the instructor’s networks of course content. This exploratory study uses a one-group pretest-posttest design to examine the effects of instructor-provided network feedback during lectures on students’ knowledge structure measured using a concept sorting task at the pretest and posttest. Undergraduate students in an architectural engineering course (n = 32) completed a lesson module and then an end-of-unit quiz on building with wood and wood framing. Three weeks later, as a review, students completed a sorting task that used 26 terms from that lesson, then a week later, the sorting task data were used to create a group-average network, this network along with the instructor’s expert network were added to that week’s lecture slides and were compared and discussed during class time. A week later, students completed the sorting task again. The pre and post-sorting data were rendered into pathfinder networks, and then these students’ networks were compared to five referent networks, specifically the textbook chapter network, the lecture slides network, a network of the end-of-unit quiz, the actual expert network that served as the feedback intervention, and the group-average network. Inspection of means shows that knowledge structure measures improved for all five measures from pre-to-post, becoming more like the lesson content and like the expert. Repeated measures analysis with follow-up paired samples t-tests showed pre-to-post significant increases for both the end-of-unit quiz and the expert network referents. The findings show that instructor presentation of structural feedback as networks improved or ‘tuned’ students’ knowledge structure of the lesson content. This approach only takes a few extra minutes of class time and is fairly simple to implement in ordinary classrooms, and so it has wide potential to support classroom instruction and student learning. Further research is needed to determine how critical it is to present both the group-average network along with the expert network for comparison in order to highlight group-level misconceptions, or is presenting only the expert network sufficient? If a group-level network is necessary, then a simple clicker-like classroom tool could be developed to collect sorting task data during lectures that could then immediately provide the group-average network for class discussion and reflection.

Keywords: classroom instruction, engineering education, knowledge structure, pathfinder networks, structural feedback

Procedia PDF Downloads 40

Authors: N. Renuka, R. Ramesh Babu, N. Vijayan

Abstract:

Ion beams play a significant role in the process of tuning the properties of materials. Based on the radiation behavior, the engineering materials are categorized into two different types. The first one comprises organic solids which are sensitive to the energy deposited in their electronic system and the second one comprises metals which are insensitive to the energy deposited in their electronic system. However, exposure to swift heavy ions alters this general behavior. Depending on the mass, kinetic energy and nuclear charge, an ion can produce modifications within a thin surface layer or it can penetrate deeply to produce long and narrow distorted area along its path. When a high energetic ion beam impinges on a material, it causes two different types of changes in the material due to the columbic interaction between the target atom and the energetic ion beam: (i) inelastic collisions of the energetic ion with the atomic electrons of the material; and (ii) elastic scattering from the nuclei of the atoms of the material, which is extremely responsible for relocating the atoms of matter from their lattice position. The exposure of the heavy ions renders the material return to equilibrium state during which the material undergoes surface and bulk modifications which depends on the mass of the projectile ion, physical properties of the target material, its energy, and beam dimension. It is well established that electronic stopping power plays a major role in the defect creation mechanism provided it exceeds a threshold which strongly depends on the nature of the target material. There are reports available on heavy ion irradiation especially on crystalline materials to tune their physical and chemical properties. L-Arginine Acetate [LAA] is a potential semi-organic nonlinear optical crystal and its optical, mechanical and thermal properties have already been reported The main objective of the present work is to enhance or tune the structural and optical properties of LAA single crystals by heavy ion irradiation. In the present study, a potential nonlinear optical single crystal, L-arginine acetate (LAA) was grown by slow evaporation solution growth technique. The grown LAA single crystal was irradiated with oxygen ions at the dose rate of 600 krad and 1M rad in order to tune the structural and optical properties. The structural properties of pristine and oxygen ions irradiated LAA single crystals were studied using Powder X- ray diffraction and Fourier Transform Infrared spectral studies which reveal the structural changes that are generated due to irradiation. Optical behavior of pristine and oxygen ions irradiated crystals is studied by UV-Vis-NIR and photoluminescence analyses. From this investigation we can concluded that oxygen ions irradiation modifies the structural and optical properties of LAA single crystals.

Keywords: heavy ion irradiation, NLO single crystal, photoluminescence, X-ray diffractometer

Procedia PDF Downloads 222

Authors: Dorottya Guban, Irina Borbath, Istvan Bakos, Peter Nemeth, Andras Tompos

Abstract:

One of the greatest challenges of the implementation of polymer electrolyte membrane fuel cells (PEMFCs) is to find active and durable electrocatalysts. The cell performance is always limited by the oxygen reduction reaction (ORR) on the cathode since it is at least 6 orders of magnitude slower than the hydrogen oxidation on the anode. Therefore high loading of Pt is required. Catalyst corrosion is also more significant on the cathode, especially in case of mobile applications, where rapid changes of loading have to be tolerated. Pt-Sn bulk alloys and SnO2-decorated Pt3Sn nanostructures are among the most studied bimetallic systems for fuel cell applications. Exclusive formation of supported Sn-Pt alloy phases with different Pt/Sn ratios can be achieved by using controlled surface reactions (CSRs) between hydrogen adsorbed on Pt sites and tetraethyl tin. In this contribution our results for commercial and a home-made 20 wt.% Pt/C catalysts modified by tin anchoring via CSRs are presented. The parent Pt/C catalysts were synthesized by modified NaBH4-assisted ethylene-glycol reduction method using ethanol as a solvent, which resulted either in dispersed and highly stable Pt nanoparticles or evenly distributed raspberry-like agglomerates according to the chosen synthesis parameters. The 20 wt.% Pt/C catalysts prepared that way showed improved electrocatalytic performance in the ORR and stability in comparison to the commercial 20 wt.% Pt/C catalysts. Then, in order to obtain Sn-Pt/C catalysts with Pt/Sn= 3 ratio, the Pt/C catalysts were modified with tetraethyl tin (SnEt4) using three and five consecutive tin anchoring periods. According to in situ XPS studies in case of catalysts with highly dispersed Pt nanoparticles, pre-treatment in hydrogen even at 170°C resulted in complete reduction of the ionic tin to Sn0. No evidence of the presence of SnO2 phase was found by means of the XRD and EDS analysis. These results demonstrate that the method of CSRs is a powerful tool to create Pt-Sn bimetallic nanoparticles exclusively, without tin deposition onto the carbon support. On the contrary, the XPS results revealed that the tin-modified catalysts with raspberry-like Pt agglomerates always contained a fraction of non-reducible tin oxide. At the same time, they showed increased activity and long-term stability in the ORR than Pt/C, which was assigned to the presence of SnO2 in close proximity/contact with Pt-Sn alloy phase. It has been demonstrated that the content and dispersion of the fcc Pt3Sn phase within the electrocatalysts can be controlled by tuning the reaction conditions of CSRs. The bimetallic catalysts displayed an outstanding performance in the ORR. The preparation of a highly dispersed 20Pt/C catalyst permits to decrease the Pt content without relevant decline in the electrocatalytic performance of the catalysts.

Keywords: anode catalyst, cathode catalyst, controlled surface reactions, oxygen reduction reaction, PtSn/C electrocatalyst

Procedia PDF Downloads 204

Authors: M. Lehtihet, S. Rosado, C. Pradère, J. Leng

Abstract:

Poly(3,4-ethylene dioxythiophene) polystyrene sulfonate (PEDOT: PSS), is a polymer mixture well-known for its semiconducting properties and is widely used in the coating industry for its visible transparency and high electronic conductivity (up to 4600 S/cm) as a transparent non-metallic electrode and in organic light-emitting diodes (OLED). It also possesses strong absorption properties in the Near Infra-Red (NIR) range (λ ranging between 900 nm to 2.5 µm). In the present work, we take advantage of this absorption to explore its potential use as a transparent light-to-heat converter. PEDOT: PSS aqueous dispersions are deposited onto a glass substrate using a blade-coating technique in order to produce uniform coatings with controlled thicknesses ranging in ≈ 400 nm to 2 µm. Blade-coating technique allows us good control of the deposit thickness and uniformity by the tuning of several experimental conditions (blade velocity, evaporation rate, temperature, etc…). This liquid coating technique is a well-known, non-expensive technique to realize thin film coatings on various substrates. For coatings on glass substrates destined to solar insulation applications, the ideal coating would be made of a material able to transmit all the visible range while reflecting the NIR range perfectly, but materials possessing similar properties still have unsatisfactory opacity in the visible too (for example, titanium dioxide nanoparticles). NIR absorbing thin films is a more realistic alternative for such an application. Under solar illumination, PEDOT: PSS thin films heat up due to absorption of NIR light and thus act as planar heaters while maintaining good transparency in the visible range. Whereas they screen some NIR radiation, they also generate heat which is then conducted into the substrate that re-emits this energy by thermal emission in every direction. In order to quantify the heating power of these coatings, a sample (coating on glass) is placed in a black enclosure and illuminated with a solar simulator, a lamp emitting a calibrated radiation very similar to the solar spectrum. The temperature of the rear face of the substrate is measured in real-time using thermocouples and a black-painted Peltier sensor measures the total entering flux (sum of transmitted and re-emitted fluxes). The heating power density of the thin films is estimated from a model of the thin film/glass substrate describing the system, and we estimate the Solar Heat Gain Coefficient (SHGC) to quantify the light-to-heat conversion efficiency of such systems. Eventually, the effect of additives such as dimethyl sulfoxide (DMSO) or optical scatterers (particles) on the performances are also studied, as the first one can alter the IR absorption properties of PEDOT: PSS drastically and the second one can increase the apparent optical path of light within the thin film material.

Keywords: PEDOT: PSS, blade-coating, heat, thin-film, Solar spectrum

Procedia PDF Downloads 131

Authors: Dechao Meng, Yongqi Dong, Qiyuan Feng, Zhangzhang Cui, Xiang Hu, Haoliang Huang, Genhao Liang, Huanhua Wang, Hua Zhou, Hawoong Hong, Jinghua Guo, Qingyou Lu, Xiaofang Zhai, Yalin Lu

Abstract:

Great efforts have been taken to reveal the intrinsic origins of emerging ferromagnetism (FM) in strained LaCoO₃ (LCO) films. However, some macro magnetic performances of LCO are still not well understood and even controversial, such as magnetic anisotropy. Determining and understanding magnetic anisotropy might help to find the true causes of FM in turn. Perpendicular magnetic anisotropy (PMA) was the first time to be directly observed in high-quality LCO films with different thickness. The in-plane (IP) and out of plane (OOP) remnant magnetic moment ratio of 30 unit cell (u.c.) films is as large as 20. The easy axis lays in the OOP direction with an IP/OOP coercive field ratio of 10. What's more, the PMA could be simply tuned by changing the thickness. With the thickness increases, the IP/OOP magnetic moment ratio remarkably decrease with magnetic easy axis changing from OOP to IP. Such a huge and tunable PMA performance exhibit strong potentials in fundamental researches or applications. What causes PMA is the first concern. More OOP orbitals occupation may be one of the micro reasons of PMA. A cluster-like magnetic domain pattern was found in 30 u.c. with no obvious color contrasts, similar to that of LaAlO₃/SrTiO₃ films. And the nanosize domains could not be totally switched even at a large OOP magnetic field of 23 T. It indicates strong IP characters or none OOP magnetism of some clusters. The IP magnetic domains might influence the magnetic performance and help to form PMA. Meanwhile some possible nonmagnetic clusters might be the reason why the measured moments of LCO films are smaller than the calculated values 2 μB/Co, one of the biggest confusions in LCO films.What tunes PMA seems much more interesting. Totally different magnetic domain patterns were found in 180 u.c. films with cluster magnetic domains surrounded by < 110 > cross-hatch lines. These lines were regarded as structure domain walls (DWs) determined by 3D reciprocal space mapping (RSM). Two groups of in-plane features with fourfold symmetry were observed near the film diffraction peaks in (002) 3D-RSM. One is along < 110 > directions with a larger intensity, which is well match the lines on the surfaces. The other is much weaker and along < 100 > directions, which is from the normal lattice titling of films deposited on cubic substrates. The < 110 > domain features obtained from (103) and (113) 3D-RSMs exhibit similar evolution of the DWs percentages and magnetic behavior. Structure domains and domain walls are believed to tune PMA performances by transform more IP magnetic moments to OOP. Last but not the least, thick films with lots of structure domains exhibit different electrical transport behaviors. A metal-to-insulator transition (MIT) and an angular dependent negative magnetic resistivity were observed near 150 K, higher than FM transition temperature but similar to that of spin-orbital coupling related 1/4 order diffraction peaks.

Keywords: structure domain, magnetic anisotropy, magnetic domain, domain wall, 3D-RSM, strain

Procedia PDF Downloads 125

Authors: Burak Oksuzoglu, Savas Yildirim, Ferhat Kutlu

Abstract:

Opinion mining (OM) is one of the natural language processing (NLP) problems to determine the polarity of opinions, mostly represented on a positive-neutral-negative axis. The data for OM is usually collected from various social media platforms. In an era where social media has considerable control over companies’ futures, it’s worth understanding social media and taking actions accordingly. OM comes to the fore here as the scale of the discussion about companies increases, and it becomes unfeasible to gauge opinion on individual levels. Thus, the companies opt to automize this process by applying machine learning (ML) approaches to their data. For the last two decades, OM or sentiment analysis (SA) has been mainly performed by applying ML classification algorithms such as support vector machines (SVM) and Naïve Bayes to a bag of n-gram representations of textual data. With the advent of deep learning and its apparent success in NLP, traditional methods have become obsolete. Transfer learning paradigm that has been commonly used in computer vision (CV) problems started to shape NLP approaches and language models (LM) lately. This gave a sudden rise to the usage of the pretrained language model (PTM), which contains language representations that are obtained by training it on the large datasets using self-supervised learning objectives. The PTMs are further fine-tuned by a specialized downstream task dataset to produce efficient models for various NLP tasks such as OM, NER (Named-Entity Recognition), Question Answering (QA), and so forth. In this study, the traditional and modern NLP approaches have been evaluated for OM by using a sizable corpus belonging to a large private company containing about 76,000 comments in Turkish: SVM with a bag of n-grams, and two chosen pre-trained models, multilingual universal sentence encoder (MUSE) and bidirectional encoder representations from transformers (BERT). The MUSE model is a multilingual model that supports 16 languages, including Turkish, and it is based on convolutional neural networks. The BERT is a monolingual model in our case and transformers-based neural networks. It uses a masked language model and next sentence prediction tasks that allow the bidirectional training of the transformers. During the training phase of the architecture, pre-processing operations such as morphological parsing, stemming, and spelling correction was not used since the experiments showed that their contribution to the model performance was found insignificant even though Turkish is a highly agglutinative and inflective language. The results show that usage of deep learning methods with pre-trained models and fine-tuning achieve about 11% improvement over SVM for OM. The BERT model achieved around 94% prediction accuracy while the MUSE model achieved around 88% and SVM did around 83%. The MUSE multilingual model shows better results than SVM, but it still performs worse than the monolingual BERT model.

Keywords: BERT, MUSE, opinion mining, pretrained language model, SVM, Turkish

Procedia PDF Downloads 112

Authors: Saheli Mitra, Ramesh Karri, Praveen K. Mylapalli, Arka. B. Dey, Gourav Bhattacharya, Gouriprasanna Roy, Syed M. Kamil, Surajit Dhara, Sunil K. Sinha, Sajal K. Ghosh

Abstract:

The most well-known structures of lyotropic liquid crystalline systems are the two dimensional hexagonal phase of cylindrical micelles with a positive interfacial curvature and the lamellar phase of flat bilayers with zero interfacial curvature. In aqueous solution of surfactants, the concentration dependent phase transitions have been investigated extensively. However, instead of changing the surfactant concentrations, the local curvature of an aggregate can be altered by tuning the electrostatic interactions among the constituent molecules. Intermediate phases with non-uniform interfacial curvature are still unexplored steps to understand the route of phase transition from hexagonal to lamellar. Understanding such structural evolution in lyotropic liquid crystalline systems is important as it decides the complex rheological behavior of the system, which is one of the main interests of the soft matter industry. Sodium dodecyl sulfate (SDS) is an anionic surfactant and can be considered as a unique system to tune the electrostatics by cationic additives. In present study, imidazolium-based ionic liquids (ILs) with different number of carbon atoms in their single hydrocarbon chain were used as the additive in the aqueous solution of SDS. At a fixed concentration of total non-aqueous components (SDS and IL), the molar ratio of these components was changed, which effectively altered the electrostatic interactions between the SDS molecules. As a result, the local curvature is observed to modify, and correspondingly, the structure of the hexagonal liquid crystalline phases are transformed into other phases. Polarizing optical microscopy of SDS and imidazole-based-IL systems have exhibited different textures of the liquid crystalline phases as a function of increasing concentration of the ILs. The small angle synchrotron x-ray diffraction (SAXD) study has indicated the hexagonal phase of direct cylindrical micelles to transform to a rectangular phase at the presence of short (two hydrocarbons) chain IL. However, the hexagonal phase is transformed to a lamellar phase at the presence of long (ten hydrocarbons) chain IL. Interestingly, at the presence of a medium (four hydrocarbons) chain IL, the hexagonal phase is transformed to another hexagonal phase of direct cylindrical micelles through the lamellar phase. To the best of our knowledge, such a phase sequence has not been reported earlier. Even though the small angle x-ray diffraction study has revealed the lattice parameters of these phases to be similar to each other, their rheological behavior has been distinctly different. These rheological studies have shed lights on how these phases differ in their viscoelastic behavior. Finally, the packing parameters, calculated for these phases based on the geometry of the aggregates, have explained the formation of the self-assembled aggregates.

Keywords: lyotropic liquid crystals, polarizing optical microscopy, rheology, surfactants, small angle x-ray diffraction

Procedia PDF Downloads 112

Authors: Nutthan Inkhong, Sutthiphong Ruangchante

Abstract:

Chiang Rai province is located at the northern border of Thailand. Most of the geography there is the northern continental highlands, and the population has many types of inhabitants, including Thai people, immigrants and ethnic groups such as Akha, Lahu, Lisu, Yao, etc. Most of these ethnic groups migrated from neighbouring countries such as Myanmar, Laos, China, etc. and settled in the mountains. Each ethnic group has their unique traditions, culture, and ways of life, including the musical culture that the ancestors of each ethnic group brought with them. In the present, the Akha have the largest population in the region and still live together in numerous villages in many districts. Thus, Akha musical culture still appears in the community traditions and cultural events of Chiang Rai province regularly. This article presents the situations of Akha musical culture in the present and the predictions for the future. The study method involves the analysis of music information and the related social contexts, which were collected from the fieldwork of ethnomusicological methodology by in-depth interviews, observations, audio and visual recordings, and related documents. The results found that the important persons who are related with Akha musical culture include (1) a musical instrument maker (lives in Mae Chan district) who produces various Akha musical instruments, including gourd mouth organs, Akha drums, two-way flutes, three-hole flutes, Jew’s harps (the sound of teenage love), buffalo horns (the sound symbol of hunting) and bird call instruments (the imitation of bird sounds), (2) a folk philosopher (lives in Mae Pha Luang district) who can teach music to the new generation of Akha people as well as lecture and demonstrate music to academics and tourists, and (3) a community leader (lives in Mae Chan district) who conserves Akha performances, singing and music through various activities of the students in an informal school. Because of the changes to the social contexts and ways of life of the Akha people, such as the educational system, religion, social media, etc., including the popularity of both Thai and international popular music among the new generation of Akha people, changes to and the fading away of Akha musical culture in the future may likely occur. Therefore, the conservation and inheritance of Akha music is an issue that should be resolved quickly. This primary study leads to the next step of the ethnomusicological work and plays a part in preventing or reducing the problems impacting Akha musical culture survival by the recording of Akha music in all of its dimensions, such as producing musical instruments, playing musical instruments, analysis of tuning systems, recording Akha music as musical notation using symbols, researching related social contexts, etc. and the transcription of this information to create lessons that can be returned to the Akha community.

Keywords: Akha music, Chiang Rai, ethnic music in Thailand, ethnomusicology

Procedia PDF Downloads 126

Authors: Colin Schmidt, Adrien Lecossier, Pascal Crubleau, Simon Richir

Abstract:

Introduction:The widespread availability of artificial intelligence, exemplified by Generative Pre-trained Transformers (GPT) relying on large language models (LLM), has caused a seismic shift in the realm of knowledge. Everyone now has the capacity to swiftly learn how these models can either serve them well or not. Today, conversational AI like ChatGPT is grounded in neural transformer models, a significant advance in natural language processing facilitated by the emergence of renowned LLMs constructed using neural transformer architecture. Inventiveness of an LLM : OpenAI's GPT-3 stands as a premier LLM, capable of handling a broad spectrum of natural language processing tasks without requiring fine-tuning, reliably producing text that reads as if authored by humans. However, even with an understanding of how LLMs respond to questions asked, there may be lurking behind OpenAI’s seemingly endless responses an inventive model yet to be uncovered. There may be some unforeseen reasoning emerging from the interconnection of neural networks here. Just as a Soviet researcher in the 1940s questioned the existence of Common factors in inventions, enabling an Under standing of how and according to what principles humans create them, it is equally legitimate today to explore whether solutions provided by LLMs to complex problems also share common denominators. Theory of Inventive Problem Solving (TRIZ) : We will revisit some fundamentals of TRIZ and how Genrich ALTSHULLER was inspired by the idea that inventions and innovations are essential means to solve societal problems. It's crucial to note that traditional problem-solving methods often fall short in discovering innovative solutions. The design team is frequently hampered by psychological barriers stemming from confinement within a highly specialized knowledge domain that is difficult to question. We presume ChatGPT Utilizes TRIZ 40. Hence, the objective of this research is to decipher the inventive model of LLMs, particularly that of ChatGPT, through a comparative study. This will enhance the efficiency of sustainable innovation processes and shed light on how the construction of a solution to a complex problem was devised. Description of the Experimental Protocol : To confirm or reject our main hypothesis that is to determine whether ChatGPT uses TRIZ, we will follow a stringent protocol that we will detail, drawing on insights from a panel of two TRIZ experts. Conclusion and Future Directions : In this endeavor, we sought to comprehend how an LLM like GPT addresses complex challenges. Our goal was to analyze the inventive model of responses provided by an LLM, specifically ChatGPT, by comparing it to an existing standard model: TRIZ 40. Of course, problem solving is our main focus in our endeavours.

Keywords: artificial intelligence, Triz, ChatGPT, inventiveness, problem-solving

Procedia PDF Downloads 31

Authors: P. Dubcek, B. Pivac, J. Dasovic, V. Janicki, S. Bernstorff

Abstract:

When grown only to nanometric sizes, metallic particles (e.g. Ag, Au and Cu) exhibit specific optical properties caused by the presence of plasmon band. The plasmon band represents collective oscillation of the conduction electrons, and causes a narrow band absorption of light in the visible range. When the nanoparticles are embedded in a dielectric, they also cause modifications of dielectrics optical properties. This can be fine-tuned by tuning the particle size. We investigated Cu nanoparticle growth with and without surrounding dielectric (SiO2 capping layer). The morphology and crystallinity were investigated by GISAXS and GIWAXS, respectively. Samples were produced by high vacuum thermal evaporation of Cu onto monocrystalline silicon substrate held at room temperature, 100°C or 180°C. One series was in situ capped by 10nm SiO2 layer. Additionally, samples were annealed at different temperatures up to 550°C, also in high vacuum. The room temperature deposited samples annealed at lower temperatures exhibit continuous film structure: strong oscillations in the GISAXS intensity are present especially in the capped samples. At higher temperatures enhanced surface dewetting and Cu nanoparticles (nanoislands) formation partially destroy the flatness of the interface. Therefore the particle type of scattering is enhanced, while the film fringes are depleted. However, capping layer hinders particle formation, and continuous film structure is preserved up to higher annealing temperatures (visible as strong and persistent fringes in GISAXS), compared to the non- capped samples. According to GISAXS, lateral particle sizes are reduced at higher temperatures, while particle height is increasing. This is ascribed to close packing of the formed particles at lower temperatures, and GISAXS deduced sizes are partially the result of the particle agglomerate dimensions. Lateral maxima in GISAXS are an indication of good positional correlation, and the particle to particle distance is increased as the particles grow with temperature elevation. This coordination is much stronger in the capped and lower temperature deposited samples. The dewetting is much more vigorous in the non-capped sample, and since nanoparticles are formed in a range of sizes, correlation is receding both with deposition and annealing temperature. Surface topology was checked by atomic force microscopy (AFM). Capped sample's surfaces were smoother and lateral size of the surface features were larger compared to the non-capped samples. Altogether, AFM results suggest somewhat larger particles and wider size distribution, and this can be attributed to the difference in probe size. Finally, the plasmonic effect was monitored by UV-Vis reflectance spectroscopy, and relative weak plasmonic effect could be explained by uncomplete dewetting or partial interconnection of the formed particles.

Keywords: coper, GISAXS, nanoparticles, plasmonics

Procedia PDF Downloads 99

Authors: Federico Capra, Emanuele Martelli, Matteo Gazzani, Marco Mazzotti, Maurizio Notaro

Abstract:

Natural gas is a major energy source in the current global economy, contributing to roughly 21% of the total primary energy consumption. Production of natural gas starting from renewable energy sources is key to limit the related CO2 emissions, especially for those sectors that heavily rely on natural gas use. In this context, biomethane produced via biogas upgrading represents a good candidate for partial substitution of fossil natural gas. The upgrading process of biogas to biomethane consists in (i) the removal of pollutants and impurities (e.g. H2S, siloxanes, ammonia, water), and (ii) the separation of carbon dioxide from methane. Focusing on the CO2 removal process, several technologies can be considered: chemical or physical absorption with solvents (e.g. water, amines), membranes, adsorption-based systems (PSA). However, none emerged as the leading technology, because of (i) the heterogeneity in plant size, ii) the heterogeneity in biogas composition, which is strongly related to the feedstock type (animal manure, sewage treatment, landfill products), (iii) the case-sensitive optimal tradeoff between purity and recovery of biomethane, and iv) the destination of the produced biomethane (grid injection, CHP applications, transportation sector). With this contribution, we explore the use of a technology for biogas upgrading and we compare the resulting performance with benchmark technologies. The proposed technology makes use of a chemical sorbent, which is engineered by RSE and consists of Di-Ethanol-Amine deposited on a solid support made of γ-Alumina, to chemically adsorb the CO2 contained in the gas. The material is packed into fixed beds that cyclically undergo adsorption and regeneration steps. CO2 is adsorbed at low temperature and ambient pressure (or slightly above) while the regeneration is carried out by pulling vacuum and increasing the temperature of the bed (vacuum-temperature swing adsorption - VTSA). Dynamic adsorption tests were performed by RSE and were used to tune the mathematical model of the process, including material and transport parameters (i.e. Langmuir isotherms data and heat and mass transport). Based on this set of data, an optimal VTSA cycle was designed. The results enabled a better understanding of the interplay between material and cycle tuning. As exemplary application, the upgrading of biogas for grid injection, produced by an anaerobic digester (60-70% CO2, 30-40% CH4), for an equivalent size of 1 MWel was selected. A plant configuration is proposed to maximize heat recovery and minimize the energy consumption of the process. The resulting performances are very promising compared to benchmark solutions, which make the VTSA configuration a valuable alternative for biomethane production starting from biogas.

Keywords: biogas upgrading, biogas upgrading energetic cost, CO2 adsorption, VTSA process modelling

Procedia PDF Downloads 243

Authors: Poojan Gharat, Haridas Pal, Sharmistha Dutta Choudhury

Abstract:

Tuning photophysical properties of guest dyes through host-guest interactions involving macrocyclic hosts are the attractive research areas since past few decades, as these changes can directly be implemented in chemical sensing, molecular recognition, fluorescence imaging and dye laser applications. Excited state intramolecular proton transfer (ESIPT) is an intramolecular prototautomerization process display by some specific dyes. The process is quite amenable to tunability by the presence of different macrocyclic hosts. The present study explores the interesting effect of p-sulfonatocalix[n]arene (SCXn) and cyclodextrin (CD) hosts on the excited-state prototautomeric equilibrium of Chrysazine (CZ), a model antitumour drug. CZ exists exclusively in its normal form (N) in the ground state. However, in the excited state, the excited N* form undergoes ESIPT along with its pre-existing intramolecular hydrogen bonds, giving the excited state prototautomer (T*). Accordingly, CZ shows a single absorption band due to N form, but two emission bands due to N* and T* forms. Facile prototautomerization of CZ is considerably inhibited when the dye gets bound to SCXn hosts. However, in spite of lower binding affinity, the inhibition is more profound with SCX6 host as compared to SCX4 host. For CD-CZ system, while prototautomerization process is hindered by the presence of β-CD, it remains unaffected in the presence of γCD. Reduction in the prototautomerization process of CZ by SCXn and βCD hosts is unusual, because T* form is less dipolar in nature than the N*, hence binding of CZ within relatively hydrophobic hosts cavities should have enhanced the prototautomerization process. At the same time, considering the similar chemical nature of two CD hosts, their effect on prototautomerization process of CZ would have also been similar. The atypical effects on the prototautomerization process of CZ by the studied hosts are suggested to arise due to the partial inclusion or external binding of CZ with the hosts. As a result, there is a strong possibility of intermolecular H-bonding interaction between CZ dye and the functional groups present at the portals of SCXn and βCD hosts. Formation of these intermolecular H-bonds effectively causes the pre-existing intramolecular H-bonding network within CZ molecule to become weak, and this consequently reduces the prototautomerization process for the dye. Our results suggest that rather than the binding affinity between the dye and host, it is the orientation of CZ in the case of SCXn-CZ complexes and the binding stoichiometry in the case of CD-CZ complexes that play the predominant role in influencing the prototautomeric equilibrium of the dye CZ. In the case of SCXn-CZ complexes, the results obtained through experimental findings are well supported by quantum chemical calculations. Similarly for CD-CZ systems, binding stoichiometries obtained through geometry optimization studies on the complexes between CZ and CD hosts correlate nicely with the experimental results. Formation of βCD-CZ complexes with 1:1 stoichiometry while formation of γCD-CZ complexes with 1:1, 1:2 and 2:2 stoichiometries are revealed from geometry optimization studies and these results are in good accordance with the observed effects by the βCD and γCD hosts on the ESIPT process of CZ dye.

Keywords: intermolecular proton transfer, macrocyclic hosts, quantum chemical studies, photophysical studies

Procedia PDF Downloads 84

Authors: Alexandre Leclerc, Christian Gout, Carole Le Guyader, Pierre Besset, Carlos Miranda, Olivier Gibaru

Abstract:

In this work, we address the challenges faced by human operators in visually inspecting production lines in factories, especially in automotive manufacturing involving numerous checkpoints to be handled every minute for extended hours. This tedious and error-prone process calls for a solution, and we propose an AI-powered assistant to achieve this goal. The objective is to design a system that can efficiently perform visual inspections on moving items and assemblies, in a structured manner, such as on conveyors and production lines. The system task is to visually verify specific part conformity for each assembly or class of assembly. For instance, in an automotive assembly line, the assistant may need to check the presence of headlights and ensure proper rim mounting. To meet these requirements, the assistant must detect the passage of relevant assemblies and perform anomaly detection on relevent parts. Notably, the system copes with the a prior knowledge deficit challenge of which assemblies and parts the users will choose for inspection. Therefore, it needs to be versatile enough to work across a wide range of industrial use cases. By incorporating these considerations, the proposed AI-powered assistant aims to reduce the difficulty, tedium and unreliability of human visual checks on production lines. Constraints specific to the industrial sector underpin the proposal: • Easy and fast setup (< 1 hour). • Strong hardware constraints on GPU power and memory. • Short inference time (< 100 ms). The proposed approach involves leveraging recent advances in mathematical models and computer science tools complying with the above requirements. It can be summarized as follows: 1) Implementing state-of-the-art deep neural networks trained on vast and diverse datasets (known as foundation models) to assist with annotation, facilitating rapid preparation for detector training. 2) Utilizing the latest models capable of handling object detection and instance segmentation tasks to track the objects needing analysis. Result refinement is then achieved through trajectory analysis of the examined areas. 3) Incorporating a pre-trained classifier and fine-tuning it using detector-acquired data. 4) Inferring the classification model on the detections obtained from the trackers to warn the operator quickly when necessary. Furthermore, we subjected these methods to continuous testing in real industrial environments to ensure their practical applicability and effectiveness.

Keywords: anomaly detection, computer vision, deep learning, image and video processing, image segmentation, industrial inspection, production line

Procedia PDF Downloads 52