Identifying Potato Varieties Using Machine Vision and Artificial Neural Networks

The objective of this study is to develop a method for identifying and discriminating 10 potato varieties by combining machine vision and artificial neural network methods. The potato varieties include Agria, Savalan, Florida, Fontaneh, Natasha, Verona, Karso, Elody, Satina, and Emrad. A total number of 72 characteristic parameters specifying color, textural, and morphological features are found among these varieties. By using principal component analysis, 16 principal features are selected for identifying and discriminating potato varieties. The data obtained from image processing were classified using linear discriminant analysis and non-linear artificial neural network method. The accuracy of discriminant analysis were 73.3, 93.3, 73.3, 40, 73.3, 73.3, 66.7, 80, 40, and 53.3%, respectively, for the varieties used in this study. The classification accuracy was improved by 100% for all the varieties using neural network analysis and the correct classification ratio was 100% using this method. It is revealed from the results that machine vision technique and neural network analysis could identify potato varieties with acceptable accuracy.     

Development of a novel graphene oxide-blended polysulfone mixed matrix membrane with improved hydrophilicity and evaluation of nitrate removal from aqueous solutions

In this study, four types of mixed matrix membranes were fabricated using polysulfone (as the base polymer) and different contents of graphene oxide (GO) nanosheets (as modifier) through wet phase inversion method. Based on the amounts of GO (0, 0.5, 1, and 2?wt%), the synthesized membranes named as M1, M2, M3, and M4, respectively. The membranes characteristics were evaluated using FE-SEM, FT-IR, and water contact angle measurements. In addition, the performance of the prepared membranes was investigated in terms of basic parameters: filtrate water flux, nitrate removal efficiency, and antifouling properties. Results showed significant improvements of the characteristics of modified membranes with GO. Accordingly, the permeability and hydrophilicity were enhanced and water flux was considerably improved. At operating pressure of 4?bar and nitrate concentration of 110?mg/L, the removal efficiency for unmodified membrane (M1) was 15.5% and for modified M2, M3, and M4 membranes were 22.78%, 39.12%, and 41.37%, respectively. In addition, the results of flux recovery ratio (FRR) showed that the anti-fouling properties of the GO modified membranes were improved due to the increase in membrane surface hydrophilicity.     

Methanol steam reforming in a microchannel reactor by Zn-, Ce- and Zr- modified mesoporous Cu/SBA-15 nanocatalyst

Hydrogen production by steam reforming of methanol was studied over several Cu/SAB-15-based nanocatalysts in a parallel-type microchannel reactor. The catalysts were prepared through impregnation method and XRD, BET, FT-IR, FE-SEM, TEM, H₂-TPR and TGA techniques were used to characterize surface and structural properties of the synthesized catalysts. The effects of reaction temperature, WHSV and S/C molar ratio on the methanol conversion and selectivities of the gaseous products were studied. Then, effects of the metallic promoters were investigated to improve performance of the catalysts. It was revealed that ZnO and CeO₂ promoters have positive effects on decreasing CO selectivity and ZrO₂ promotes methanol conversion. Furthermore, ZrO₂ and CeO₂ were declared to improve stability of the catalyst. Among the evaluated catalysts, Cu/ZnO/CeO₂/ZrO₂/SBA-15 can provide optimal methanol conversion with low CO concentration in the gaseous products. For this catalyst, the methanol conversion and hydrogen selectivity reached 95.2% and 94.6%, respectively.     

Studying effect of cell wall's carboxyl-carboxylate ratio change of Lemna minor to remove heavy metals from aqueous solution

The pre-treated Lemna minor can remove Hg(II), Cr(III), Cr(VI) and Cu(II) from the aqueous solution. The concentration determination of carboxyl and carboxylate groups and its rule to change the metal ions uptake was done by the curves of the potentiometric titration. It was shown that the removal percent of the heavy metal ions (Co=1.00 mM) increased 25.1, 26.0, 17.2 and 24.1% for these ions, respectively, with increasing the carboxylate from 0.92 to 2.42 mmol/g L. minor and then activating by the activator chloride salts. The removal percent of these ions was decreased 33.1, 27.5, 20.7 and 15.01%, respectively, with increasing the carboxyl from 1.50 to 2.41 mmol/g L. minor, inspite of activating by the chloride salts. The enthalpy change (DeltaH) was -27.43, -25.94, -28.12 and -22.27 kJ/mol and the entropy change (DeltaS) was 81.3, 79.9, 86.1 and 67.7 J/mol K, by activated biomass, respectively. L. minor removed these heavy metals corresponding to pseudo-second-order kinetic model that the activation energy (E(a)) was obtained 18.59, 15.93, 20.36 and 21.12 kJ/mol by the activated and 23.02, 22.27, 23.98 and 24.46 kJ/mol by the reference biomass to uptake Hg(II), Cr(III), Cr(VI) and Cu(II), respectively.     

Asymmetry in retail gasoline and crude oil price movements in the United States: An application of hidden cointegration technique

There is a common belief that gasoline prices respond more quickly to crude oil price increases than decreases. Some economists and politicians believe that asymmetry in oil and gasoline price movements is the outcome of a non-competitive gasoline market requiring that governments take policy action to address “unfair pricing”. There is no consensus as to the existence, or nature, of the asymmetric relationship between prices of gasoline and crude oil. Much of this literature specifies asymmetry in the speed of adjustment and short-run adjustment coefficients. In contrast, Granger and Yoon's [Granger, C.W. and Yoon, G. “Hidden Cointegration”, University of California, San Diego, Department of Economics Working Paper, (2002).] Crouching Error Correction Model (CECM) identifies asymmetry of the cointegrating vectors between components (cumulative positive and negative changes) of the series. Applying the CECM to retail gasoline and crude oil prices for the U.S., we find that there is only evidence of cointegration between positive components of crude oil prices and negative components of gasoline prices. In contrast to the literature which attributes asymmetric price movements to market power of refiners, these findings suggest that gasoline prices –in the long run– are more influenced by the technological changes on the demand side than crude oil price movements on the supply side.     

CFD design and simulation of ethylene dichloride (EDC) thermal cracking reactor

At high temperature, ethylene dichloride decomposes into vinyl chloride monomer and HCl by a complex reaction mechanism. Due to chemical reaction type and heat transfer mechanism, design and simulation of cracking reactors are complicated. This process is very important from either chemical reaction or transport phenomena’s point of view. In this study, 2D simulation of cracking reactor is performed with FLUENT software. Simulation results were compared with the experimental data. Then, we investigate the simulation results and effect of various parameters on the process. Results show that with process progress and temperature increase during reactor length, conversion and coke formation amount are increased. At the end of the process, conversion is 52.69 that has good agreement with industrial data. Addition of promoter to the process fluid and fluid behavior in different coil outlet temperatures was studied; finally, the effects of these factors are described in detail.     

Alcalase-generated proteolysates of stone fish (Actinopyga lecanora) flesh as a new source of antioxidant peptides

In the present study, the alcalase-generated proteolysates obtained after 8 h of proteolysis of stone fish flesh showed the most potent antioxidant activity in terms of DPPH^? radical scavenging activity (77.43%, IC₅₀ of 0.5 mg/mL), ABTS^? radical scavenging activity (92.73%, IC₅₀ of 0.33 mg/mL) and FRAP value (39.2 mmol/100 mL FeSO₄). These proteolysates profiled and characterized as antioxidative peptides. The proteolysates were initially subjected to ultrafiltration using MWCO Spin-X UF. Potent fractions were further characterized based on hydrophobicity using reversed-phase high-performance liquid chromatography (RP-HPLC) and isoelectric point using an OFFGEL isoelectric focusing fractionator. Results indicated that most of the antioxidative peptides found in fractions with a molecular weight (MW) of less than 2 kDa, hydrophilic (hydrophilicity >80%) and basic (pI = 9.7). The final purified fraction with the highest antioxidant activity was selected for peptide identification and sequencing using Q-TOF mass spectrometry. A total of four peptides were identified, from which Peptide 1 (GVSGLHID) showed the highest antioxidant activity and this has potential as a novel bioingredient of nutraceuticals and functional foods to promote human health.     

Estimation of research reactor core parameters using cascade feed forward artificial neural networks

The pattern of the core reload program is very important for an optimize use of research reactors. Reactor safety issues and economic efficiency should be considered during pattern studies. In order to find the best core pattern for a research reactor, its reloading program should be solved as a multi-objective and constrained optimization problem. If considered objective functions of the optimization problem can be estimated in very short time, the optimal fuel reloading pattern can be used effectively. In this research a very fast estimation system for suggested core parameters has been developed using cascade feed-forward type of artificial neural networks (ANNs). Four main core parameters are suggested to optimize reactor core adequately. And also to get larger thermal fluxes in the desired irradiation box, a new flexible method was selected. A Software package has been developed to prepare and reform required data for ANNs training. The gradient descent method with momentum weight/bias learning rule has been used to train ANNs. To get the best conditions for considered ANNs training a vast study has been performed. It includes the effects of variation of hidden neurons, hidden layers, activation functions, learning and momentum coefficients, and also the number of training data sets on the training and simulation results. Some experimental convergence criteria are used to study them. A comparison selection rule has been used to adjust desirable conditions. Final training and simulation results show that developed ANNs can be trained and estimate suggested core parameters of research reactors very quickly. It improves effectively pattern optimization process of core reload program.