http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Mohammad Pazouki,Mohammad Zabihi,Jalal Shayegan,Mohammad Hossein Fatehi 한국화학공학회 2018 Korean Journal of Chemical Engineering Vol.35 No.3
An efficient, novel functionalized supported magnetic nanoparticle (AC@Fe3O4-NH2-COOH) has been synthesized by co-precipitation method for removal of mercury ions from saline solutions. High dispersed supported magnetic nanoparticles with particle sizes less than 30 nm were formed over activated carbon derived from local walnut shell. Surface characterizations of supported magnetic nanoparticles were evaluated by Boehm test, Brunauer- Emmett-Teller (BET) surface area, X-ray diffraction (XRD), transmission electron microscopy (TEM), Fourier transform infrared spectroscopy (FT-IR), thermogravimetric analysis (TGA) and X-ray fluorescence (XRF). A three-layer artificial neural network (ANN) code was developed to predict the Hg (II) ions removal from aqueous solution by AC@Fe3O4-NH2-COOH. The three-layer back-propagation (BP) is configured of tangent sigmoid transfer function (tansig) at hidden layer with eight neurons for AC@Fe3O4-NH2-COOH, and linear transfer function (purelin) at output layer. According to the calculated MSEs, Levenberg-Marquardt algorithm (LMA) was the best training algorithm among others. The linear regressions between the predicted and experimental outputs were proven to be a good agreement with a correlation coefficient of about 0.9984 for five model variables. Maximum adsorption capacity was achieved 80mg/g by Langmuir isotherm at pH of 7 and salinity of 25,000 ppm. Kinetic studies illustrated that mercury adsorption follows pseudo-second-order.
Seungdeog Choi,Pazouki, Elham,Baek, Jeihoon,Bahrami, Hamid Reza Institute of Electrical and Electronics Engineers 2015 IEEE transactions on industrial electronics Vol. No.
<P>This paper presents a robust diagnosis technique by iteratively analyzing the pattern of multiple fault signatures in a motor current signal. It is mathematically and experimentally proved that the proposed diagnosis algorithm provides highly accurate monitoring performance while minimizing both false detection and miss detection rate under high noise and nonlinear machine operating condition. These results are verified on a digital-signal-processor-based motor drive system where motor control and fault diagnosis are performed in real time.</P>
Emadaldin M. Golafshani,Gholamreza Pazouki 사단법인 한국계산역학회 2018 Computers and Concrete, An International Journal Vol.22 No.4
The compressive strength of self-compacting concrete (SCC) containing fly ash (FA) is highly related to its constituents. The principal purpose of this paper is to investigate the efficiency of hybrid fuzzy radial basis function neural network with biogeography-based optimization (FRBFNN-BBO) for predicting the compressive strength of SCC containing FA based on its mix design i.e., cement, fly ash, water, fine aggregate, coarse aggregate, superplasticizer, and age. In this regard, biogeography-based optimization (BBO) is applied for the optimal design of fuzzy radial basis function neural network (FRBFNN) and the proposed model, implemented in a MATLAB environment, is constructed, trained and tested using 338 available sets of data obtained from 24 different published literature sources. Moreover, the artificial neural network and three types of radial basis function neural network models are applied to compare the efficiency of the proposed model. The statistical analysis results strongly showed that the proposed FRBFNN-BBO model has good performance in desirable accuracy for predicting the compressive strength of SCC with fly ash.