Search

Micro ED milling is a micro machining technique for machining of high-quality micro parts on the electrically conductive material. Achieving a good surface finish with higher material removal is difficult in Micro ED (μED) Milling due to the short circuits and arcing attributed to partial removal of debris. In this research work, an attempt has been made to improve the quality of machined surfaces in the milling process without compromising the machining characteristics using nano powder mixed dielectrics such as Silicon carbide (SiC) and Carbon nano tubes (CNT). The milling trails were carried out using Tungsten carbide (WC) electrodes under adopted levels of powder concentration and discharge energy region in order to determine the best condition for achieving maximum material removal and excellent surface finish. Further, the morphology of the machined surface was examined using Scanning Electron Microscope (SEM) and found that ridges, crater size and overlapping of craters were considerably reduced due to addition of nano powders in dielectric resulted in high surface finish. The micro cracks are larger with higher discharge energy for all dielectrics used, whereas it is significantly reduced with nano powder mixed dielectrics as compared to plain dielectric. CNT powder offered high MRR and the minimum EWR as compared to other dielectrics used. Both the CNT and SiC powder mixed dielectrics have marginally improved the surface hardness. Results indicated that the electrical conductivity of dielectric has a significant effect on machining and surface characteristics. The surface texture achieved with CNT mixed dielectric looks better than that of other dielectrics: especially with the higher order of powder concentration of 0.9 g/l. Milling trails were designed using full factorial design for plain and nano powder mixed dielectrics.

Wear behavior of aluminum matrix composites (AMCs) has been studied. AMCs consist of LM6 aluminum alloy as the matrix material and 3%, 6% and 9% of particulate fly ash by weight as the reinforcement material. Stir casting process was used to produce the AMCs. Wear tests were performed on a pin on a disc machine according to ASTM G99 – 05 specifications. Taguchi's methodology was implemented and L27 orthogonal array was selected for conducting the wear tests. The main objective is to use S / N ratio and analysis of variance to investigate the effect of load, sliding distance, sliding speed and weight percentage of reinforcement on specific wear rate (SWR). Load has the largest statistical impact on SWR (24.31%) followed by sliding distance (16.43%). To validate the results of the experiments, confirmation tests were performed.

International audience; Single-image super-resolution (SISR) techniques attempt to reconstruct the finer resolution version of a given image from its coarser version. In the SISR of hyperspectral data sets, the simultaneous consideration of spectral bands is crucial for ensuring the spectral fidelity. However, the high spectral resolution of these data sets affects the performance of conventional approaches. This research proposes the design of 3-D convolutional neural network (CNN)-based SISR architectures that can map the spatial-spectral characteristics of hypercubes to a finer spatial resolution. The proposed approaches facilitate the simultaneous optimization of sparse codes and dictionaries with regard to the super-resolution objective. Our main hypothesis is that the consideration of spectral aspects is essential for the spatial enhancement of hyperspectral images. Also, we propose that the regularized deconvolution of a coarser-scale hypercube, using learned 3-D filters, yields the required high-resolution version. Based on these hypotheses, a convolution-deconvolution framework is proposed to super-resolve the hypercubes in parallel with the reconstruction of a set of regularizing features. Novel sparse code optimization sub-networks proposed in this article give better performance than the existing strategies. The endmember similarities and hyperspectral image prior are considered while designing the proposed loss functions. In order to improve the generalizability, a collaborative spectral unmixing strategy is employed to refine the spectral base of the super-resolved result. The spatial-spectral accuracy of the super-resolved hypercubes, in terms of the validity of regularizing features and endmembers, is explored to devise an optimal ensemble strategy. The experiments, over different data sets, confirm better accuracy of the proposed frameworks compared to the prominent approaches.

Micro ED milling is one of the widely applied processes to generate microfeatures on electrically conductive materials. However, the drawbacks associated with this process namely low MRR and high surface roughness limits its wide applicability. To enhance the rate of metal removal and surface finish an investigation is carried out by mixing SiC nano powders with dielectric media. Experimental trails were conducted at different settings of discharge energy and powder concentration to identify the optimum condition for responses namely MRR, Ra and Micro hardness. Full factorial design is used to design the experiments for both plain dielectric and powder mixed µEDM process. The morphology of milled surface with plain /nano powder mixed dielectric has also been analyzed. The findings revealed that the break down voltage of SiC nano powder mixed dielectric was 40% lower than plain dielectric media due to this enhancement in MRR and surface finish. Lower values of powder concentration (2 g/l) favors the MRR and surface finish but the higher value of powder concentration (6 g/l) favors the microhardness. Crater size, voids and cracks were reduced with SiC mixed dielectric than plain dielectric under considered settings of discharge energy due to increase in sparking area. Material removal efficiency is found to be high with medium range of discharge energy for both plain and SiC mixed dielectric used.

This paper investigates the tribological behaviour of grey cast iron (GCI) and its counter face material (EN31 steel) under various environmental conditions. In many applications, GCI is subjected to different metal-to-metal contact under dry, lubricated and high temperature conditions. The main phenomenon in metal-to-metal sliding is wear; it may occur on either surface or any one of the surfaces due to the property of material or nature of contact or the sliding parameter of the test. Adhesive wear can occur on the surface due to the presence of localized bonding or oxide formation. As a result of this metal-to-metal contact, localized bonding may occur, leading to material transfer from GCI to EN31 steel or vice versa. This situation brings up the new era of positive or negative wear, at various environmental conditions in practical applications. If the material transfer is observed, then the surface could be subjected to adhesive wear. If oxidation takes place, then the material is subjected to rolling wear. Sometimes both surfaces may be subjected to material transfer due to severe sliding action. This work aims to determine the states of wear and sliding parameters under which slight and severe wear occurs. The states of wear include dry, lubricated and heated conditions. The sliding parameters are applied load, sliding speed and sliding distance. The combinations of sliding parameters were selected based on the principle of full factorial method, and L27 orthogonal array was employed.

Natural fibre reinforced plastics is one of the advanced polymer matrix composites developed in the area of engineering materials. The main reason behind its growth is the advantages that come along, such as being biodegradable, recyclable, lightweight, low cost of production, and abundantly available. Therefore, Natural fibre reinforced plastics are being used in many applications such as automotive components, aerospace parts, sporting goods and building industry. In this research polypropylene (PP) along with chopped banana fibre of approximately 1 to 1.5 mm in length is used to prepare a composite specimen with 85%:15% and 80%:20% polymer to fibre ratio. The natural fibres were made hydrophobic by treating it with 6% NaOH solution and then with 1% HCl. Series of mechanical properties like tensile strength in accordance with ASTM D638, flexural strength in accordance with ASTM D790 is tested. The specimen developed out of two different combinations of banana fibre reinforced plastics has been tested for its tensile and flexural strength and it has been found out that they exhibit good tensile and flexural strength.

Wind resource assessment is a challenging task where the availability of offshore Met buoys is limited. Satellite remote sensing provides the best alternative inexpensive data and fills the data gaps by providing a huge volume of data for extended periods. However, due to the availability of multiple satellite scatterometers, the assessment of such resources from any single scatterometer would lead to inconsistency. The present article provides an extensive comparison of four scatterometers namely QuikSCAT, OSCAT, ASCAT-A and ASCAT-B with long-term in-situ wind datasets over the North Indian Ocean. Results show that the QuikSCAT and the OSCAT wind data have a lesser bias with the range of 0.15 m/s (2.4%) to 0.83 m/s (15.1%) before adjustments. For adjusting the satellite wind speeds, linear regression was used. Further, the synergetic approach of linear regression adjustments and the combination of scatterometer data have resulted in smaller differences. The absolute deviation in mean wind speed between co-located buoy data and combined scatterometer data is less of the order 2.1%, and the mean power density showed a difference of 2.5%. The study revealed that the combined scatterometer datasets would help in the accurate assessments of wind resource rather than a single scatterometer.

In this paper, a Capsulenet-based framework is proposed for extracting spectral and spatial features for improving hyperspectral image classification. Unlike conventional strategies, the proposed framework simultaneously optimizes both feature extraction and classification. The spectral features/patterns derived at different levels of hierarchies are remodeled as spectral-feature capsules. Consequently, unlike conventional convolutional neural network-based approaches, the relative locations as well as other properties such as depth, width, and position of the spectral patterns are taken into consideration. In addition to learning spectral features/patterns, a convolutional long short-term memory (conv-LSTM) is employed for sequentially integrating the spatial features learned from each band. The integrated spatial-feature representation, thus obtained from the final hidden state of conv-LSTM, forms spatial-feature capsules. The capsule-level integration of spatial and spectral features/patterns yields better convergence and accuracy as compared to both ensemble-based and kernel-level integrations. Along with the margin loss, a spectral-angle-based reconstruction loss is also minimized to regularize the learning of network weights. Experiments over different standard datasets indicate that the proposed approach performs better than other prominent hyperspectral classifiers. Furthermore, in comparison with the recent deep learning models, our approach is found to be less sensitive to the network parameters and achieves better accuracy even with lesser network depth.

Spatial resolution enhancement is a pre-requisite for integrating unmanned aerial vehicle (UAV) datasets with the data from other sources. However, the mobility of UAV platforms, along with radiometric and atmospheric distortions, makes the task difficult. In this paper, various convolutional neural network (CNN) architectures are explored for resolving the issues related to sub-pixel classification and super-resolution of drone-derived datasets. The main contributions of this work are: 1) network-inversion based architectures for super-resolution and sub-pixel mapping of drone-derived images taking into account their spectral-spatial characteristics and the distortions prevalent in them 2) a feature-guided transformation for regularizing the inversion problem 3) loss functions for improving the spectral fidelity and inter-label compatibility of coarser to finer-scale mapping 4) use of multi-size kernel units for avoiding over-fitting. The proposed approach is the first of its kind in using neural network inversion for super-resolution and sub-pixel mapping. Experiments indicate that the proposed super-resolution approach gives better results in comparison with the sparse-code based approaches which generally result in corrupted dictionaries and sparse codes for multispectral aerial images. Also, the proposed use of neural network inversion, for projecting spatial affinities to sub-pixel maps, facilitates the consideration of coarser-scale texture and color information in modeling the finer-scale spatial-correlation. The simultaneous consideration of spectral bands, as proposed in this study, gives better super-resolution results when compared to the individual band enhancements. The proposed use of different data-augmentation strategies, for emulating the distortions, improves the generalization capability of the framework. Sensitivity of the proposed super-resolution and sub-pixel mapping frameworks with regard to the network parameters is thoroughly analyzed. The experiments over various standard datasets as well as those collected from known locations indicate that the proposed frameworks perform better when compared to the prominent published approaches.

The continuum fluctuations of river water quality and the metal contamination within the sediments of Netravati river basin were analysed using multivariate techniques and environmental indices. Water and sediment samples were collected from ten locations along the flow path of river in the basin and the hydrogeochemical features were investigated. The results were compared with maximum permissible limit values recommended by World Health Organization. A comprehensive Water Quality Index methodology was applied to come up with transparent understanding of overall excellence in water quality status. The dominant water cations and anions were within the order of Na+ > Mg2+ > Ca2+ > K+ and HCO3− > Cl− > SO42− > NO3−. The average abundance of sediment heavy metals was in the order of Pb > Mn > Ni > Zn > Cr > Cu > Co. An approach of exploiting environmental indices: enrichment factor, geoaccumulation index, contamination factor, degree of contamination, modified degree of contamination and Pollution Load Index were accustomed to assess the pollutant level. Geochemical investigation and spatial distribution of heavy metals were represented by geographic information system based maps. The results indicate that lead contamination of surficial sediments might pose a significant threat to aquatic biota and the persistent exposure of pollutants even in low concentration causes changes in metabolic activities and alterations in the community structure of river biota.

Micro-electric discharge milling (μED milling) has the potential to create micro-channels in metals. Since μED milling is a slow machining process, the requirement of strategies that can enhance the material removal rate (MRR) is always crucial in the field of micro-machining. Nano-powder-mixed micro-electric discharge milling (NPMμED milling) is least explored by researchers. Also, the behaviour of μED milling at various discharge energy regime (D.E.R) was not analysed extensively. The present study investigates the combined effects of nano-powder-mixed dielectric at three different regimes of discharge energies with respect to MRR and tool wear rate (TWR) while micro-machining of Ti-6Al-4V using cryogenically treated electrodes. There is significant difference in the machining behaviour at low, medium and high D.E.Rs. Also, the effects of transformation in cathodic streamers due to addition of nano-graphene powders in the dielectric fluid were extensively analysed. Cryogenic treatment of the tool electrodes gave better MRR. The addition of nano-powders in the dielectric has reduced the crater diameter significantly however increased the overall 3D roughness of the machined surface.

As per Indian Standards, laterally loaded piles are usually analysed using the method adopted by IS 2911-2010 (Part 1/Section 2). But the practising engineers are of the opinion that the IS method is very conservative in design. This work aims at determining the extent to which the conventional IS design approach is conservative. This is done through a comparative study between IS approach and the theoretical model based on Vesic’s equation. Bore log details for six different bridges were collected from the Kerala Public Works Department. Cast in situ fixed head piles embedded in three soil conditions both end bearing as well as friction piles were considered and analyzed separately. Piles were also modelled in STAAD.Pro software based on IS approach and the results were validated using Matlock and Reese (In Proceedings of fifth international conference on soil mechanics and foundation engineering, 1961) equation. The results were presented as the percentage variation in values of bending moment and deflection obtained by different methods. The results obtained from the mathematical model based on Vesic’s equation and that obtained as per the IS approach were compared and the IS method was found to be uneconomical and conservative.

An algorithm has been developed to retrieve ocean surface wind speed from Synthetic Aperture Radar (SAR) on-board Radar Imaging Satellite-1 (RISAT-1). The retrieved wind speed is subsequently valid...

This article proposes a completely mechanized calculation for the exact division of high-resolution computerized tomography regions (HRCTs). A portrayal in light of the power of the locale permits considering the picture for the most part comprising of three districts: the base of the CT, the lungs and the chest area encompassing the lungs. Another technique in view of the surge fill calculation is utilized to viably recognize the encompassing district. This progression encourages the utilization of another quick strategy to evacuate the CT foundation utilizing direct outputs beginning from installed pixels. The related segments speaking to the tracheal parts are expelled by watching the partition of the mean and standard deviation of the power esteems between the trachea and the lungs. Divided aspiratory pictures are additionally improved to reestablish pixel force esteems to bronchi and lung restrict. The proposed system isn't just computational easily, yet in addition powerful and precise in distinguishing lung limits. This article introduces a total picture, including illustrations and test comes about.

The rapid pace of advancements in Artificial Intelligence (AI) is delivering unprecedented opportunities to upgrade and enhance the performance of various technologies and organizations, mainly in the transport area. AI needs a good understanding of the information and connection among AI on one hand, and transportation system characteristics and factors, on the other hand, to use the AI successfully. In addition, it is promising for every transport specialists to determine which is the best approach or method to utilize these advances to predict a quick improvement in diminishing congestion in the transport system. And also for making travel time progressively solid which is very useful to their customers and improve their financial status and productivity of their fundamental resources. Artificial Intelligence systems which mainly involves addressing the transportation issues mostly in traffic the board, traffic safety, problems during travelling in public transportation, urban portability. It is too difficult to handle in the transportation system when the transport issues become a challenge to model and anticipate the transport patterns. In this paper, we have a perfect view of the Artificial intelligence that will help the future transport scenarios thereby making it easy for people to travel using smart cars. We have also discussed about the different ways by which a road, rail track, airway can be modernized using artificial intelligence and its impact on people’s life style.

Clustering algorithm is frequently used for extending a distance metric or a similarity evaluation for the separation of data from database. The divided data points are more similar and they are categorized significantly to cluster in high dimensional data spaces. Then, clustered data points are projected with different diverse set of proportions. Clustering high dimensional data is a proficient research field. High-dimensional data are wide-ranging in numerous areas of forest scenario, machine learning, signal and image processing, computer vision, pattern recognition, bioinformatics and so on. Let us consider the forest scenario for clustering the information about the trees among dissimilar tree structure. Based on the clustering of forest information, they combine diverse areas of capability and equipment.

Classification of crops using time-series vegetation index (VI) curves requires appropriate modeling of phenological events and their characteristics. The current study explores the use of capsules, a group of neurons having an activation vector, to learn the characteristic features of the phenological curves. In addition, joint optimization of denoising and classification is adopted to improve the generalizability of the approach and to make it resilient to noise. The proposed approach employs reconstruction loss as a regularizer for classification, whereas the crop-type label is used as prior information for denoising. The activity vector of the class capsule is applied to sample the latent space conditioned on the cell state of a Long Short-Term Memory (LSTM) that integrates the sequences of the phenological events. Learning of significant phenological characteristics is facilitated by adversarial variational encoding in conjunction with constraints to regulate latent representations and embed label information. The proposed architecture, called the variational capsule network (VCapsNet), significantly improves the classification and denoising results. The performance of VCapsNet can be attributed to the suitable modeling of phenological events and the resilience to outliers and noise. The maxpooling-based capsule implementation yields better results, particularly with limited training samples, compared to the conventional implementations. In addition to the confusion matrix-based accuracy measures, this study illustrates the use of interpretability-based evaluation measures. Moreover, the proposed approach is less sensitive to noise and yields good results, even at shallower depths, compared to the main existing approaches. The performance of VCapsNet in accurately classifying wheat and barley crops indicates that the approach addresses the issues in crop-type classification. The approach is generic and effectively models the crop-specific phenological features and events. The interpretability-based evaluation measures further indicate that the approach successfully identifies the crop transitions, in addition to the planting, heading, and harvesting dates. Due to its effectiveness in crop-type classification, the proposed approach is applicable to acreage estimation and other applications in different scales.

Molecular Dynamic simulations and experiments have been performed to investigate the evaporative behavior of a liquid droplet in the presence of nanoparticles on a smooth, heated flat plate. Nanofluids of different volume fractions with CuO and Al2O3 nanoparticles were prepared to obtain liquid droplets. Experiments were conducted for droplets of both nanofluids and base fluid (DI water). The results showed that in the pinned region of evaporation, droplets of nanofluids with higher volume fractions of dense nanoparticles evaporated by keeping their base pinned at the triple point contact line. It is also observed that the evaporation rate of nanofluid droplets with nanoparticles having higher density is slower and promoted a typical pattern formation. Molecular Dynamic simulations were further performed to analyse the influence of various nanoparticle-liquid interaction strengths ( e np − l ) on evaporative behavior. The simulation results suggest that in the pinned evaporation stage, for higher values of e np − l , the droplet undergoes a delay on evaporation while subjected to higher heat flux on the substrate.

Increase in energy demand and scarcity of fossil fuels have increased concerns over rudimentary energy security. The adverse impact on the environment by fossil fuel based power generation has supervened the focus on renewable energy. To ascertain the reliability of power, diesel generator (DG-set) based backup system is predominant. DG-set based power generation has become very costly and has an adverse impact on the environment. Renewable energy (RE) based power generation (solar/wind) proves to be an effective alternative but is intermittent and variable in nature. One of the key solutions is by providing energy storage in multi-generation systems (MGS). Energy storage plays a vital role in MGS by smoothening the variations in the intermittent generation and arresting the net load variations. However, an approach for obtaining optimal asset configuration of MGS for mitigating DG-set along with operating strategy for a battery energy storage system (BESS) is the real challenge. A generic adaptable model for techno-economic optimization has been developed and presented. This paper attempts to find the optimal sizes for a battery bank with solar PV system for eliminating DG-set for microgrids. The problem has been solved using mixed integer linear programming (MILP) based optimization. The economic viability along with carbon mitigation for different microgrids have been presented. The evaluation of the optimization model is carried out with six different real case studies (Educational, micro industrial, residential, medical institutions) of India. The techno-economic analysis has been presented for different scenarios of each case study and compared with the existing solution.

Nowadays reconfigurable antennae are finding their importance in a variety of applications. This research proposes to design a novel micro controller based frequency reconfigurable patch antenna using RF P---I---N diodes. This proposed design consists of three RF P---I---N diodes and by adjusting the ON and OFF state of the diodes using the microcontroller unit, the radiating frequency of the antenna can be changed and hence it covers different wireless bands. Eight different switching states were achieved through these three P---I---N diode combinations. A prototype antenna was fabricated with a length and width of 50 mm × 60 mm respectively and a substrate thickness of 1.6 mm. The simulations were carried out through HFSS and the results were obtained using vector network analyzer. Upon comparing the simulation results with the actual prototype antenna, the return loss was observed to be below ź10 dB and VSWR value was

The electrode materials titanium dioxide and copper oxide doped titanium dioxide are synthesized. Modified electrodes are characterized by SEM and cyclic voltammetry (CV). We present a systematic study of the electrochemistry of azines at bare and modified electrodes such as Glassy carbon electrode (GCE), Nafion coated titanium dioxide GCE (Naf/TiO 2 @GCE), Nafion coated copper oxide doped titanium dioxide GCE (Naf/CuO–TiO 2 @GCE), Platinum electrode (Pt), Nafion coated titanium dioxide Pt (Naf/TiO 2 @Pt) and Nafion coated copper oxide doped titanium dioxide Pt (Naf/CuO–TiO 2 @Pt) in cyclic voltammetry. Platinum, glassy carbon and its modified working electrodes were compared. In all azines better electrochemical responses were observed on GCE and its modified electrodes, which yielded higher currents in the cyclic voltammograms. Electrocatalytic reduction of azine on the surface modified electrodes was investigated in cyclic voltammetry and the results showed that the Naf/CuO–TiO 2 @GCE has excellent electrochemical catalytic activities toward azine reduction. The electrocatalytic investigation reveals electrocatalytic effect not only depends up on nature and porosity of electrode material but it also depends on nature of electrolyte and electrode material acts as screen for electrochemical response of electrolyte. In this paper also better explanation about how electron donating and withdrawing groups is enhanced and decreased the reduction potential in cyclic voltammetry.

The present study investigates the feasibility of wave energy exploitation along the Indian coast using 19-year (2000–2018) hindcasted wave data generated by WAVEWATCH-III with a fine spatial resolution (0.1° × 0.1°) and temporal resolution (6 h). The wave model is validated against 12-month measured wave data collected at multiple buoy sites and compared with 19-year ERA5 data across the study area. The results show good agreement between model data and the other two datasets. The spatio-temporal variability analysis of wave energy flux reveals that southern coast of India has the mean wave energy flux ranging from 6 to 10 kW/m with a minimum monthly (

In this work, SnO2 loaded with a minor amount of corn cob activated carbon (CCAC) was synthesized through a simple chemical precipitation method. The product materials were characterized by XRD, XPS, UV-DRS, PL, FTIR, FE-SEM with EDAX, TEM, BET, EIS analysis and its photocatalytic activity for methylene blue (MB) degradation under sunlight irradiation was investigated. XRD pattern showed that the average crystalline size is reducing (17.25 nm) upon loading CCAC compared to the SnO2. The band gap is reduced (3.52 eV) upon loading on CCAC than the SnO2. Furthermore, lower PL emission intensity of the SnO2/CCAC enhanced separation of the photogenerated charge carriers. It increased surface area and pore volume is due to the adding CCAC is 293.12 m2/g and 0.345 cm3/g are confirmed by BET analysis. Therefore, SnO2/CCAC composite of the maximum removal efficiency (90.86 %) of MB dye shows that the improve SnO2 (photocatalysis) and CCAC (adsorption) under sunlight irradiation within 120 min. Further, it also possible mechanism and pseudo-first-order model is used to analyze the kinetics studies.

A favorable application to use Mn-doped ZnO NPs and loaded with a (0.5 g) small amount of cotton stalk activated carbon (CSAC) in photocatalytic degradation of BG dye. The impact of Mn-ZnO/CSAC has been studied by XRD, XPS, UV–Vis, FTIR, FE-SEM with EDAX, BET and EIS analysis. We found that just, Mn-ZnO/CSAC showed that the particle size, band gap is also reduced and increase surface area. There is an optimum Mn-doped ZnO/CSAC for the degradation efficiency of BG showed improved photocatalytic activity (97.47%) under sunlight irradiation at a catalytic dose of 0.20 g/L than Mn doped ZnO and pure ZnO NPs.

Bayesian optimisation is a popular method in optimising complex, unknown and expensive objective functions. In complex design optimisation problems, the additional information about the smoothness, monotonicity or the modality of the unknown objective functions can be obtained either from the domain expertise or from the problem environment. Incorporating such additional information can potentially enhance the performance of the optimisation. We propose a methodology to incorporate the aforesaid extra information to have a better fitted surrogate model of the unknown objective function. Specifically, for Gaussian Process regression, we propose a covariance function to encompass varying smoothness across the input space through a parametric function whose parameters are tuned from the observations. Our experiments on both synthetic benchmark functions and real-world applications demonstrate that embodying such additional knowledge accelerates the convergence.

The spectral super-resolution techniques attempt to re-project spectrally coarse images to a set of finer wavelength bands. However, complexity of the mapping between coarser and finer scale spectra, large variability of spectral signatures, and the difficulty in simultaneously modeling spatial and spectral contexts make the problem highly ill-posed. Our main hypothesis is that the consideration of spatial as well as spectral aspects is essential for spectral enhancement of remote sensing images. In this regard, this paper proposes a framework consisting of sparse-coding-based pixel-spectra enhancement, collaborative unmixing, and spatial-spectral prior based transformation. Two sparse-coding-based architectures are proposed to project the coarser scale pixel-spectra to the target scale. These models facilitate simultaneous optimization of sparse codes and dictionaries with regard to the spectral super-resolution objective. A CNN based encoding-decoding architecture is explored to model the spatial-spectral prior for improving fidelity of the reconstructions. The endmember similarities and spectral image prior are considered while designing the proposed loss functions. The experiments, over standard as well as AVIRIS-NG and drone-derived datasets, confirm better accuracy of the proposed frameworks as compared to the prominent approaches. In addition, the proposed CNN models for spectral upscaling and spatial-spectral transformation are found to be less sensitive to the variation in network parameter values.

The deep network features are being widely explored for improving the classification of remote sensing images. However, for hyperspectral datasets, the spectral features are found to be more significant as compared to their spatial counterparts. In this study, a deep learning framework is proposed for modelling the spectral features. Unlike the conventional strategies, the approach simultaneously optimizes both the feature extraction and the classification stages. In this approach, the spectral features derived from different levels of hierarchies, re-modelled as capsules, are used to label the given spectrum based on an iterative dynamic routing process. Consequently, unlike the regular convolutional architectures, here the relative locations of the spectral artefacts are also taken into consideration. Along with the margin loss, a spectral-angle-based reconstruction loss is also employed to facilitate proper regularization. Experiments over different standard datasets indicate that the proposed approach performs better when compared to the prominent approaches. Furthermore, in comparison with the former deep learning models, our approach is found to be less sensitive to the network parameters and achieves better accuracy even with lesser network depth.

This paper presents different multi-carrier unipolar trapezoidal pulse width modulation strategies for a reduced switch asymmetrical multilevel inverter. The different strategies are phase disposition, alternative phase opposition and disposition, and carrier overlapping and variable frequency that involve triangular waves as carriers with a unipolar trapezoidal wave as a reference. The reduced switch, asymmetrical multilevel inverter operation was examined for generating the seven-level output voltage using Matlab/Simulink 2009b and the results were verified with a real-time laboratory-based experimental setup using a field-programmable gate array. Different parameter analyses, such as total harmonic distortion, fundamental root mean square voltage, and distortion factor, were analyzed with different modulation indices to investigate the performance of the selected topology. Unipolar trapezoidal pulse width modulation provides a higher root mean square voltage value.

In this paper, we propose the use of network inversion for modeling the variation of class distributions with scale. Unlike the state of the art methods that predict the mapping between coarser and finer scale patches without considering the distributions at coarser scale, our approach uses coarser scale features for effective reconstruction. This is the pioneer work of using network inversion for the purpose. Analysis over the proposed framework reveals that both the computational performance and accuracy varies with the depth of the network as well as the size and number of filters in each layer. Also the performance of the approach has been found to improve with the increase in the number of input feature maps. Investigations over standard datasets indicate that the proposed approach performs much better than the recent sub-pixel classification as well as super resolution techniques.

This experimental investigation deals with the evaluation of abrasive wear behavior of Al alloy and Al2O3 particle reinforced Al alloy metal matrix composites on a pin-on-disc test rig, where the specimens were slided against a 600 grade SiCP emery paper. The specimen preparation and experimentation have been carried out according to the ASTM G-99 standard. A plan of experiments based on the Taguchi design of experiments was developed to acquire data in controlled way. An orthogonal array and the analysis of variance were employed to investigate the percentage of contribution of various process parameters like sliding speed, applied load, sliding distance and their interactions affecting the abrasive wear volume loss of the materials. The correlations between the various factors affecting the abrasive wear behavior of composites were obtained by using multiple linear regression equations. The experimental results showed that the sliding distance has more contribution on the abrasive wear followed by applied load for both materials. The predicted volume loss of the specimens was found to lie close to that of the experimentally observed ones.

Foil bearings are self-acting (aerodynamic) hydrodynamic bearings supports lightly loaded shafts with major advantages like physically non-contacting high speed operation using process fluid as medium in hostile environments. However, the lack of load carrying capacity at relatively lower speeds limits their applications in heavy turbo machinery and as such are highly suitable in lightly loaded, high speed turbo machinery like small gas turbines. This paper discusses the design and assessment of dynamic characteristics in terms of load carrying capabilities as a function of speed, gap between the bearing and the runner as well as shape of foils for an air foil thrust bearing. The effects of various bearing parameters like foil thickness, number of foils fixed circumferentially, foil geometry configuration and position of foils along the circumference such that inner edge and outer edge of the same foil are at different height and same height. Characteristics of performance defined essentially in terms of load carrying capabilities and static stiffness have been used for evaluation. Experiments were conducted for sector foils (with inner edge height less than outer edge height) and sector foils (with inner edge height is equal to outer edge height) by varying number of foils. Along with the above two types the load carrying capabilities of air foil thrust bearing using newly conceptualized cascaded foils (Patented) with increased number of wedge films essentially from the point view of increased pressure profile culminating in relatively higher load carrying capabilities as compared to other two types (conventional) Air foil thrust bearing. The experimental results show that the effect of foil assembly, number and improved formation of wedge film which enhances the load carrying capabilities of air foil thrust bearing. Along with the above experimental work an COMSOL model of the thrust bearing have been analysed to know that the pressure distribution profile on the foil which clearly indicates that the pressure developed by the conventional type of foils configuration is less than the newly configured foils in which inner edge height is equal to outer edge along the radial direction i.e the height of the foil with respect to the runner is equal in the radial direction.

Sub-pixel mapping techniques predict the spatial distribution of endmember abundances which are estimated through spectral unmixing. The sub-pixel mapping and spectral unmixing approaches are mostly unsupervised, and both are generally treated as independent optimization problems. This study explores convolutional encoder-decoder as well as recurrent and deconvolution networks for jointly optimizing the unmixing and mapping stages in a supervised manner. In the proposed approach, finer scale classified maps are used for training the network, thereby avoiding the requirement of fractional abundance ground truths. It is observed that the class compatibility based loss minimization yields better convergence and accuracy when compared to the conventional mean squared error (MSE) or cross entropy based approaches. An ensemble of the proposed framework is devised so as to avoid the possible convergence of stochastic gradient minimization towards local optima. The proposed approaches are evaluated using simulated and standard hyperspectral datasets. Experiments indicate that the joint optimization of spectral unmixing and sub-pixel mapping stages improves the accuracy as well as the convergence time. Also, the proposed LSTM approach gives better results, especially for linear mixtures, in comparison with the encoder-decoder based approaches. Although the LSTM approach is relatively less sensitive to the network parameters, the size and number of filters need to be tuned considering the required trade-off between accuracy and running time. (C) 2018 Elsevier B.V. All rights reserved.

This letter investigates the use of coarse-image features for predicting class labels at a given finer spatial scale. In this regard, two unsupervised subpixel mapping approaches, a semivariogram method, and a pixel-affinity based method are proposed. Furthermore, segmentation-based spectral unmixing is explored so as to address the spectral variability and nonconvexity of classes. In addition, the gradient information is employed to resolve uncertainties in the unmixing process. The proposed modifications based on pixel-affinity and semivariogram have produced an accuracy improvement of 5% or more over the state-of-the-art approaches.

Software Defined Network (SDN) is a new networking paradigm which has emerged recently. Research in SDN is flourishing with new architectures, methods and solutions. However, minimal efforts have been made for creating a vendor independent open source SDN solution; which can be readily deployed in Cloud networks, Internet of Things (IoT) etc. For betterment of open source SDN, suitable improvisations with thorough validation in real networks is necessary. In this paper, we have conducted evaluation of open source Open Flow (OF) based SDN network using OpenDaylight (ODL) controller. Both Mininet simulations and Openstack cloud network deployments have been used. As per the observations, certain functional drawbacks persist in ODL which affects the SDN control plane management. A new module called “Network Provider Module” (NPM) has been developed in ODL controller. NPM addresses the functional drawbacks in ODL and brings out comprehensive SDN capabilities for cloud provisioning as well.

This article introduces a new reduced switch multilevel inverter topology that operates in asymmetric state. The advantage of proposed multilevel inverter is utilizing the lower device count when compared to traditional configuration. This paper depicts a hybrid pulse width modulation which includes three different carriers and unipolar references. The three different carriers are saw-tooth, triangular, inverted unipolar sine which is arranged in phase disposition technique. The effectiveness of proposed hybrid pulse width modulation (HPWM) technique is tested with a new reduced switch topology using MATLAB/Simulink. The different parameters are calculated with various modulation indices.

Banana bunchy top virus (BBTV) affects all varieties of banana plants and causes heavy economic loss in most of the banana cultivating areas. The BBTV genome comprises of six DNA components; in this study, we have cloned the six BBTV-DNA components from one of the BBTV-infected plants (Tri-8) and were submitted to GenBank. Analysis of the BBTV DNA-R component showed that it belonged to south Pacific group. Resistance against BBTV has not been observed so far in banana plants and removal and killing of the infected plants has been routinely practiced. Hence, early detection of BBTV infection would be desirable and various detection methods routinely employed include enzyme linked immunosorbent assay (antigen–antibody based) and molecular-based methods such as polymerase chain reaction (PCR), qPCR, or LAMP PCR. Most of these methods require enzymes or antibodies for detection and hence are expensive. Here, we report a visual detection method (AuNP probe assay) using gold nanoparticles (AuNPs) functionalized with an ssDNA-thiolated probe (CR1). This method is based on the hybridization of the functionalized AuNPs with the target DNA (BBTV). In the AuNP probe assay, the functionalized AuNPs retains red colour when BBTV DNA is present, and in the absence of BBTV DNA, the colour of the functionalized AuNPs changes to purple when salt is added. The AuNP probe assay was compared with PCR for the detection of banana plants and it was found that AuNP probe assay was better than PCR in detecting BBTV infection (86.5% for AuNP probe assay and 65% for PCR). The AuNP probe assay was found to be highly specific to BBTV and was found to detect up to 1 pg/μl of the plasmid (pTZBBTri 4, BBTV DNA) mixed with healthy banana DNA.

ZnO is widely used as a photocatalytic process due to high photosensitivity, strongly oxidizing nature, non-toxic, favorable band gap energy and the excellent chemical and mechanical stabilities of the material. The present work deals with the synthesis of ZnO doped with transition metal (Sn) by using a simple chemical precipitation method. XRD analysis confirms the substitution of Sn4+ ions into the ZnO matrix by increased lattice spacing and a reduced 2θ angle. Optical properties like reflectance and band gap were studied using UV–Vis–DRS spectroscopy. The morphology of the samples was observed using FESEM and TEM which explains the growth rate and shape of the prepared samples. Methylene Blue (MB) dye was used to evaluate the photocatalytic activity of synthesized ZnO nanoparticles and their performance with respect time were analyzed and presented.

The synergetic offshore wind dataset derived from a combination of scatterometers (QuikSCAT, OSCAT and ASCAT-A, B) have been utilised and validated with available buoy data. It is observed that, the mean deviation in wind speed between the co-located buoy and synergetic scatterometer data is small of the order 2% and the power density showed a difference of 2.5%. The combined synergetic maps of wind speed, wind power density and practical power production from two offshore wind turbines V112 3M and G128 5M have been analysed and evaluated on a seasonal and annual basis. Further, considering the available area after exclusions, conflict regions due to shipping, visual and avian exclusions, India has a net power production of 364 GW (3190 TWh/yr) for V112 3M and 516 GW (4522 TWh/yr) for G128 5M turbines at 80 m hub height. This would reduce overall carbon emissions of 3117 and 4418 million tons respectively per year. Finally, the Levelised cost of energy is computed for two optimum sites in shallow waters off Gujarat and Tamil Nadu coasts and found to be lowest of around 106.87 €/MWh in Tamil Nadu site, suggesting an optimum site for wind farm development.

The sparse coding approaches, which learn over-complete bases for efficiently representing a given set of data, are being extensively studied for various image analysis tasks. However, since sparse analysis of individual bands does not consider inter-spectral characteristics, adaptation of these approaches to hyperspectral domain has been quite limited. In this regard, we investigate convolutional frameworks for simultaneously learning sparse codes and dictionaries so as to efficiently represent the spectral–spatial characteristics of hypercubes. This study also reviews prominent convolutional sparse coding algorithms relevant to hyperspectral datasets. Unlike conventional approaches, in order to address the issues due to huge spectral dimension and high inter-band correlation, the proposed approaches implement sparse coding in a convolutional encoder–decoder framework. Two architectures, one based on shrinkage thresholding algorithm and other based on structured sparse regularization, are proposed. These architectures adopt 3D convolutions in such a way as to model both spectral and spatial features. Although Architecture-1 gives better results than other prominent approaches, its accuracy deteriorates with increase in spectral dimension of input. This can be attributed to the bottleneck due to shrinkage thresholding operation. To address the issue, Architecture-2 adopts structured sparse regularizations over filter weights and feature tensors. Results indicate that the proposed approaches achieve significant improvement in reconstruction accuracy when compared to the existing approaches. The frameworks are also evaluated with respect to super-resolution and dimensionality reduction of hyperspectral datasets. From different experiments, it is observed that the simultaneous usage of 3D and 2D filters along with sparse regularizations significantly improve learning capability of the networks.

In this paper, we investigate the use of coarse image features for predicting class label distributions at a finer scale. The major contributions of this work are 1) use of coarse image features to improve the optimization formulation of conventional rank based approaches 2) use of inter class compatibility information from coarse images to refine the predicted target distribution 3) an enhanced unsupervised variogram based sub-pixel mapping approach 4) inclusion of abundance estimation uncertainty in the unmixing process. The proposed modifications on rank based and variogram based approaches have produced an accuracy improvement of 10–15%. The sensitivities of these approaches towards tunable parameters are also analyzed.

Super-resolution techniques use subpixel information to predict high resolution classification maps from coarse images. This study investigates for an unsupervised super-resolution approach which considers the image features to predict target spatial dependencies. Novelty of the approach is that the convolution neural networks and deep autoencoders are explored in this context. Evaluation over standard datasets revealed that the proposed method is more effective than the state of art unsupervised approaches. The method is also found to be preferable over variogram based approaches for complex scenes. This study also compares the effectiveness of shallow and deep networks and investigates the possible assessment of the optimal depth for the learning network. This technique can be further extended to a supervised framework.

Spectral unmixing techniques decompose the pixels into constituent fractions in order to extract the subpixel information. This study reviews spectral unmixing techniques from a perspective different from earlier approaches in that the problem is studied from a classification as well as clustering perspective. In this research, we focus on addressing some core issues of spectral unmixing such as endmember variability, requirement of pure endmember values, and initialization sensitivity modelling. We propose a Support Vector Machine (SVM) based unmixing technique that incorporates endmember spectral variability. The method uses endmember extraction techniques to give optimal performance even in the absence of training samples. Further, our study presents an alternation of FCM based method for incorporating spectral variability, and the approach is found to be resilient to the brightness variation. An automatic approach for fuzziness parameter selection is also introduced. The sensitivity of FCM towards endmember initialization has been considerably reduced by optimizing the initial seed selection. The proposed approaches have been analyzed over various standard datasets.

Organic bistable memory devices were fabricated by embedding a thin layer of molybdenum trioxide (MoO3) between two tris-(8-hydroxyquinoline)aluminum (Alq3) layers. The device exhibited excellent switching characteristics with an ON/OFF current ratio of 1.15 × 103 at a read voltage of 1 V. The device showed repeatable write–erase capability and good stability in both the conductance states. These conductance states are non-volatile in nature and can be obtained by applying appropriate voltage pulses. The effect of MoO3 layer thickness and its location in the Alq3 matrix on characteristics of the memory device was investigated. The field emission scanning electron microscopy (FE-SEM) images of the MoO3 layer revealed the presence of isolated nanoparticles. Based on the experimental results, a mechanism has been proposed for explaining the conductance switching of fabricated devices.

In this paper, details related to preliminary efforts made towards developing a novel semi-active journal bearing have been presented. The study which is predominantly experimental in nature uses a test rig designed and fabricated for the purpose of simulating a multi-lobed journal bearing and also to study the possibility of utilising smart materials for conversion of a circular bearing profile to a multi lobed bearing profile during operation. Preliminary experimental studies have indicated that while shape memory alloy (SMA) can be used as a medium to apply necessary forces at identified locations to obtain the desired profile shapes for a multi-lobe bearing configuration facilitating the development of semi active journal bearings, ovality ratios of the bearing profile under investigation ranging from 1.5 to 2 have been realised by suitable application of direct forces in opposing directions. The performance enhancement defined in terms of stiffness and damping of the newly configured bearing equivalent to two lobed bearing over the conventional bearing has been studied. The test rig is versatile in nature, with a facility to investigate varieties of semi active journal bearings. The study indicates an increase in dynamic stiffness and damping with increase in the ovality ratio of the semi-active journal bearings.

Vehicle suspension plays a vital role in maintaining the center of gravity to achieve perfect balancing of the vehicle to provide the comfortable ride. While designing the suspension system of automobile, vibration is the main aspect to be considered. This paper aims to analyze the automobile front and rear suspension for a four wheeler using analytical and numerical approach. Existing details of the suspension is collected using the concept of reverse engineering. Natural and forced frequency of the front and rear suspension system is calculated theoretically based on the collected data's. The natural frequency and forced frequency is numerically computed for front and rear suspension. The amplitude of vibration is reduced by replacing the spring material and its forced frequency is reduced by 1.18 % and 1.56 % for front and rear suspension system respectively. This result reveals that low carbon steel has ability to reduce the forcing frequency and can produce comfort ride.