Support Vector Machines (SVMs) have gained prominence because of their high generalization ability for a wide range of applications. However, the size of the training data that it requires to achieve a commendable performance becomes extremely large with increasing dimensionality using RBF and polynomial kernels. Synthesizing new training patterns curbs this effect. In this paper, we propose a novel multiple kernel learning approach to generate a synthetic training set which is larger than the original training set. This method is evaluated on seven of the benchmark datasets and experimental studies showed that SVM classifier trained with synthetic patterns has demonstrated superior performance over the traditional SVM classifier.
Reactive power compensation is an important issue in the control of electric power system. Reactive power from the source increases the transmission losses and reduces the power transmission capability of the transmission lines. Moreover, reactive power should not be transmitted through the transmission line to a longer distance. Hence Flexible AC Transmission Systems (FACTS) devices such as static compensator (STATCOM) unified power flow controller (UPFC) and static volt-ampere compensator (SVC) are used to alleviate these problems. In this paper, a voltage source converter (VSC) based STATCOM is developed with Artificial Neural Network Controller (ANNC) and Adaptive Neuro Fuzzy Inference System(ANFIS) controllers. The conventional PI controller has more tuning difficulties while the system parameter changes, whereas a trained neural network and ANFIS controllers requires less computation time. They have the ability to generalize and can interpolate in between the training data. The ANNC and ANFIS controllers designed were tested on a 75 V, 100 VA STATCOM in real time environment via state-of-the-art of digital signal processor advanced control engineering (dSPACE) DS1104 board and it was found that ANFIS controller was producing better results than the ANNC.
Bioremediation is based on microorganisms able to use pollutants either as a source of carbon or in co-metabolism, and is a promising strategy in cleaning the environment. Using soil contaminated with petroleum products from an industrial area in Saudi Arabia (Jubail), and after enrichment with the polycyclic aromatic hydrocarbon (PAH) naphthalene, a Methylobacterium radiotolerans strain (N7A0) was isolated that can grow in the presence of naphthalene as the sole source of carbon. M. radiotolerans is known to be resistant to gamma radiation, and this is the first documented report of a strain of this bacterium using a PAH as the sole source of carbon. The commonly reported Pseudomonas aeruginosa (strain N7B1) that biodegrades naphthalene was also identified, and gas chromatography analyses have shown that the biodegradation of naphthalene by M. radiotolerans and P. aeruginosa did follow both the salicylate and phthalate pathways.
This work aims at studying the corrosion behavior of electroless Ni-P coating on mild steel substrate heat-treated by microwave and conventional furnace (muffle) annealing. The corrosion behavior of the deposits has been evaluated by potentiodynamic polarization studies in 3.5 wt% sodium chloride solution. The heat treatment temperatures of both the microwave and the conventional furnace annealing were kept at regular intervals to study the corrosion performance of the coatings. Microwave heat treatment significantly improved the corrosion performance of the coatings. Further, corrosion mechanisms of as-coated, various microwave- and conventional heat-treated coatings were discussed for the consideration of phase constituents, grain sizes and microstrain.
Silver nanoparticles (AgNPs) have been synthesized in the presence of Strawberry fruit extract (SBFE) at room temperature. The synthesized AgNPs was characterized by UV-vis spectroscopy, SEM, EDS, XRD, TEM and FTIR. The UV-vis spectra of the AgNPs show SPR band at 450 nm. TEM results indicate that AgNPs are spherical in shape and size range between 7–65 nm. Antibacterial activity of the synthesized AgNPs has been assessed against Pseudomonas aeruginosa and Bacillus licheniformis. The results show that AgNPs exhibit inhibitory effect and effect is a function of AgNPs concentration. The antibacterial activity of the prepared AgNPs has been compared with two antibiotics, amoxicillin and ciprofloxacin. It is found that the antibiotics perform better than AgNPs.
Aluminium oxide (Al2O3) nano particles were synthesized by using both the sol gel technique and solid state reaction (SSR) method. Different proportion of nano carbon cones from 0.5% to 3.5% is doped with aluminium nitrate nano hydrate and annealed subsequently at 1000°C for 3 h to synthesize the nano composite of carbon–alumina. The synthesized samples were characterized by X-ray diffraction to identify the presence of different phases and transitions during this process. The average crystallite size of the nano alumina is found to be 45 nm by sol gel and 43 nm by SSR method respectively by Debye–Scherrer method. Average crystallite size and lattice strain of nano alumina are also estimated from Williamson Hall (WH) plot analysis. It is found to be 69 nm with the strain of 3.3×10−3 in sol gel, and in SSR method, it is 72 nm with the strain is 3.9×10−3. The interplanar distance of various planes of alumina are estimated and compared with JCPDS values. Similar analysis has also been extended for the nano composite of carbon–alumina. The surface morphology of the samples are analyzed using scanning electron microscopy and rough estimate of the crystallites is also given. From the Raman analysis, the presence of alpha phase of alumina has been confirmed. The presence of carbon in the composite has been established through diffuse reflectance spectroscopy. The FTIR spectra of the composite samples ensured the presence of Al–O–Al, O–H and C=O bonds.