Neural Computing and Applications - Support vector machine (SVM) is a well-regarded machine learning algorithm widely applied to classification tasks and regression problems. SVM was founded based... 相似文献
Thymus species are well known as medicinal plants because of their biological and pharmacological properties. Thymus caramanicus is an endemic species grown in Iran. Variation in the quantity and quality of the essential oil of wild population of T. caramanicus at different phenological stages including vegetative, floral budding, flowering and seed set are reported. The oils of air-dried samples were obtained by hydrodistillation. The yields of oils (w/w%) at different stages were in the order of: flowering (2.5%), floral budding (2.1%), seed set (2.0%) and vegetative (1.9%). The oils were analyzed by GC and GC–MS. In total 37, 37, 29 and 35 components were identified and quantified in vegetative, floral budding, full flowering and seed set, representing 99.3, 98.6, 99.2 and 97.8% of the oil, respectively. Carvacrol was the major compound in all samples. The ranges of major constituents were as follow: carvacrol (58.9–68.9%), p-cymene (3.0–8.9%), γ-terpinene (4.3–8.0%), thymol (2.4–6.0%) and borneol (2.3–4.0%). Antibacterial activity of the oils and their main compounds were tested against seven Gram-positive and Gram-negative bacteria by disc diffusion method and determining their minimum inhibitory concentration (MIC) values. The inhibition zones (IZ) and MIC values for bacterial strains, which were sensitive to the essential oil of T. caramanicus, were in the range of 15–36 mm and 0.5–15.0 mg/ml, respectively. The oils of various phenological stages showed high activity against all tested bacteria, of which Bacillus subtilis and Pseudomonas aeruginosa were the most sensitive and resistant strains, respectively. Thus, they represent an inexpensive source of natural antibacterial substances that exhibited potential for use in pathogenic systems. 相似文献
Photonic Network Communications - This paper proposes an effective method for shaping the radiation pattern intensity of photonic crystal (PhC) light-emitting diode (LED). In this method, the... 相似文献
Training artificial neural networks is considered as one of the most challenging machine learning problems. This is mainly due to the presence of a large number of solutions and changes in the search space for different datasets. Conventional training techniques mostly suffer from local optima stagnation and degraded convergence, which make them impractical for datasets with many features. The literature shows that stochastic population-based optimization techniques suit this problem better and are reliably alternative because of high local optima avoidance and flexibility. For the first time, this work proposes a new learning mechanism for radial basis function networks based on biogeography-based optimizer as one of the most well-regarded optimizers in the literature. To prove the efficacy of the proposed methodology, it is employed to solve 12 well-known datasets and compared to 11 current training algorithms including gradient-based and stochastic approaches. The paper considers changing the number of neurons and investigating the performance of algorithms on radial basis function networks with different number of parameters as well. A statistical test is also conducted to judge about the significance of the results. The results show that the biogeography-based optimizer trainer is able to substantially outperform the current training algorithms on all datasets in terms of classification accuracy, speed of convergence, and entrapment in local optima. In addition, the comparison of trainers on radial basis function networks with different neurons size reveal that the biogeography-based optimizer trainer is able to train radial basis function networks with different number of structural parameters effectively.
In this paper, the dispersion stability of multiwall carbon nanotubes (MWNTs) mixed with an epoxy resin is studied. An instrumented optical microscope with a hot stage was used to study the evolution of the carbon nanotubes (CNTs) dispersion during the cure of the resin. A new image processing approach is then introduced to quantify dispersion and identify the source of dispersion degradation during the cure. The results showed that the reduction of the resin viscosity at temperatures greater than 100?°C caused an irreversible re-agglomeration of the CNTs in the matrix. It was shown that the fine-tuning of the ratio and type of curing agent as well as the curing temperature directly affect the dispersion stability of MWNTs in the epoxy polymer. The dispersion quality was then directly correlated to the fracture toughness of the modified resin and a maximum of 20% improvement was achieved. 相似文献
The aim of this work was the fabrication of cotton fabric with multifunctional properties such as photochromic, hydrophobicity, antibacterial, and ultraviolet (UV) blocking. In this regard, a mixture of silica nanoparticles with spirooxazine as a photochromic dye first applied on the cotton fabric, and then, the fabric surface was coated with an alkylsilane compound. The homogenous distribution of silica and silica/spirooxazine hybrid on the cotton fibers was established by field emission scanning electron microscopy, energy-dispersive X-ray spectroscopy, and X-ray mapping. X-ray diffraction patterns confirmed the amorphous nature of the hybrid on the treated cotton fabric. The photochromic effect of the fabrics was measured after 5-min sunlight irradiation. Other characteristics of treated fabrics, such as antibacterial activity, hydrophobic properties, and UV-blocking activity, were also assessed. The results indicated that adding silica nanoparticles to spirooxazine had a tangible effect on photochromic activity of treated cotton fabrics and its photochromic performance was higher than that of the photochromic dye. The fabric showed stable hydrophobicity with static water contact angle values of 141.2° ± 1°. Moreover, the coated cotton fabric demonstrated proper antibacterial properties and UV-blocking activity. 相似文献
This research aims to create the durable self-cleaning property on cotton fabric using nano titanium dioxide and corona treatment as a dry and environmentally friendly process. In this regard, cotton fabrics were investigated under corona treatment with intensities of 2 and 3 A for 1, 3, and 5?min of irradiation. The treated samples with 0.1, 0.5, 1, and 1.5 on weight of bath percent were cured by nano titanium dioxide to obtain the highest self-cleaning property. The treated cotton fabrics, which were stained with direct green 6, were exposed to 20 and 400?W of UV radiation and their self-cleaning property was examined, and the results were reported. In addition, using corona treatment helped to stabilize the nano titanium dioxide on the cotton fabric surface even after 50 washing cycles. Through SEM images, Thermo Gravimetric Analysis plots, and XRD patterns, the presence of nano titanium dioxide on the surface of the treated cotton samples after washing was verified. Also, the tensile strengths of TiO2-coated cotton fabrics were measured. 相似文献
This work presents a single-objective optimization framework for designing complex photonic crystal (PhC) filters. As a case study, a super defect PhC filter with five rods is considered. Due to the large number of structural parameters and complexity of designing process, the problem is formulated and optimized by using a recent optimization algorithm called multi-verse optimizer (MVO). Six optimal super defect filters are obtained by MVO with respect to the WDM standard, which is defined by ITU-T Recommendation G.694.2. The designed super defect filters are then placed side by side to implement WDM. The results of FDTD simulation of the designed WDM show that the magnitude of output spectral transmission is higher than that of the current works in the literature. In addition, the high-quality factor and low crosstalk value (−32.9 dB) are the other advantages of the designed WDM with optimal super defect filters.