Ali Ghafary - Profile on Academia.edu (original) (raw)

Papers by Ali Ghafary

A combination of linear and nonlinear activation functions in neural networks for modeling a de-superheater

Simulation Modelling Practice and Theory, 2009

This paper deals with modeling a power plant component with mild nonlinear characteristics using ... more This paper deals with modeling a power plant component with mild nonlinear characteristics using a modified neural network structure. The hidden layer of the proposed neural network has a combination of neurons with linear and nonlinear activation functions. This approach is ...

the Proceedings of the …

In this paper the superheating system of a 325MW steam power generating plant is modeled by usage... more In this paper the superheating system of a 325MW steam power generating plant is modeled by usage of recurrent neuro-fuzzy networks and subtractive clustering. The experimental data are obtained from a complete set of field experiments under various operating conditions. Neuro-fuzzy models are constructed for each subsystem of the superheating unit. The nine fuzzy models are then constructed in a combination of series and parallel units in accordance with real power plant subsystems. Comparing the response of nonlinear neuro-fuzzy model of a subsystem with the response of its linear model obtained based on LSE method; shows that the nonlinear neuro-fuzzy model is more accurate than linear model in the sense that its response is closer to the response of the actual system. Since LSE is optimum modeling method for linear systems, it can be concluded that some of power plant subsystems are of nonlinear processes.

Porous silicon micro-particles (micro-pSi) with size in the range of 1-10 µm are obtained by etch... more Porous silicon micro-particles (micro-pSi) with size in the range of 1-10 µm are obtained by etching of silicon wafers followed by sonication. The derivatization of the micro-pSi surface by wet chemistry (silylation and coupling with a diamine) yields an interface, which exposes negative (carboxylic) or positive (amine) groups at pH 7.4. The surface modification, beyond to introduce groups for the drug loading by covalent or electrostatic interaction, stabilizes the intense orange luminescence characteristic of the silicon nano-crystallites. Derivatization by amines introduces also a second emission in the blue, which follows a different excitation pathway and can be attributed to the interface defects. The micro-pSi are efficiently internalized by human dendritic cells and do not show any toxic effect even at concentrations of 1 mg mL-1. The intrinsic luminescence of the differently functionalized micro-pSi is ...

Fuel and Energy Abstracts, 2011

Robust artificial neural network (ANN) was developed based on experimental data to predict dehydr... more Robust artificial neural network (ANN) was developed based on experimental data to predict dehydration of isopropanol by means of novel PVA-APTEOS/TEOS nanocomposite membranes in pervaporation (PV) process. The input properties were water concentration, feed temperature and nanoparticles content, while pervaporation separation index (PSI) was output. The Bayesian Regularization (BR) training method with full sampling was employed to train the network. Then, optimal ANN architecture was determined as 3:3:3:1 with log-sigmoid transfer function for hidden and output layers. The model finding revealed that nanoparticles content has significant effect on membrane performance (about 70%). The results demonstrated that the ANN model prediction and experimental data are quite match and the model can be employed with confidence for prediction of each nanocomposite membrane performance. Simulated annealing (SA) was also employed to determine controllable conditions to find the biggest PSI.

Fuel and Energy Abstracts, 2011

Robust artificial neural network (ANN) was developed based on experimental data to predict dehydr... more Robust artificial neural network (ANN) was developed based on experimental data to predict dehydration of isopropanol by means of novel PVA-APTEOS/TEOS nanocomposite membranes in pervaporation (PV) process. The input properties were water concentration, feed temperature and nanoparticles content, while pervaporation separation index (PSI) was output. The Bayesian Regularization (BR) training method with full sampling was employed to train the network. Then, optimal ANN architecture was determined as 3:3:3:1 with log-sigmoid transfer function for hidden and output layers. The model finding revealed that nanoparticles content has significant effect on membrane performance (about 70%). The results demonstrated that the ANN model prediction and experimental data are quite match and the model can be employed with confidence for prediction of each nanocomposite membrane performance. Simulated annealing (SA) was also employed to determine controllable conditions to find the biggest PSI.

Journal of Hazardous Materials, 2011

Using the Taguchi method, this study presents a systematic optimization approach for removal of l... more Using the Taguchi method, this study presents a systematic optimization approach for removal of lead (Pb) and mercury (Hg) by a nanostructure, zinc oxide-modified mesoporous carbon CMK-3 denoted as Zn-OCMK-3. CMK-3 was synthesized by using SBA-15 and then oxidized by nitric acid. The zinc oxide was loaded to the modified CMK-3 by the equilibrium adsorption of Zn(II) ions from aqueous solution followed by calcination to convert zinc nitrate to zinc oxide. The CMK-3 had porous structure and high specific surface area which can accommodate zinc oxide in a spreading manner, the zinc oxide connects to the carbon surface via oxygen atoms. The controllable factors such as agitation time, initial concentration, temperature, dose and pH of solution have been optimized. Under optimum conditions, the pollutant removal efficiency (PRE) was 97.25% for Pb(II) and 99% for Hg(II). The percentage contribution of each controllable factor was also determined. The initial concentration of pollutant is the most influential factor, and its value of percentage contribution is up to 31% and 43% for Pb and Hg, respectively. Our results show that the Zn-OCMK-3 is an effective nanoadsorbent for lead and mercury pollution remediation. Langmuir and Freundlich adsorption isotherms were used to model the equilibrium adsorption data for Pb(II) and Hg(II).

Taguchi based fuzzy logic optimization of multiple quality characteristics of cobalt disulfide nanostructures

Simulation of Structural Features on Mechanochemical Synthesis of Al2O3–TiB2 Nanocomposite by Optimized Artificial Neural Network

Antibacterial activity of silver photodeposited nepheline thin film coatings

Taguchi optimization approach for Pb(II) and Hg(II) removal from aqueous solutions using modified mesoporous carbon

A combination of linear and nonlinear activation functions in neural networks for modeling a de-superheater

Simulation Modelling Practice and Theory, 2009

This paper deals with modeling a power plant component with mild nonlinear characteristics using ... more This paper deals with modeling a power plant component with mild nonlinear characteristics using a modified neural network structure. The hidden layer of the proposed neural network has a combination of neurons with linear and nonlinear activation functions. This approach is ...

the Proceedings of the …

In this paper the superheating system of a 325MW steam power generating plant is modeled by usage... more In this paper the superheating system of a 325MW steam power generating plant is modeled by usage of recurrent neuro-fuzzy networks and subtractive clustering. The experimental data are obtained from a complete set of field experiments under various operating conditions. Neuro-fuzzy models are constructed for each subsystem of the superheating unit. The nine fuzzy models are then constructed in a combination of series and parallel units in accordance with real power plant subsystems. Comparing the response of nonlinear neuro-fuzzy model of a subsystem with the response of its linear model obtained based on LSE method; shows that the nonlinear neuro-fuzzy model is more accurate than linear model in the sense that its response is closer to the response of the actual system. Since LSE is optimum modeling method for linear systems, it can be concluded that some of power plant subsystems are of nonlinear processes.

Porous silicon micro-particles (micro-pSi) with size in the range of 1-10 µm are obtained by etch... more Porous silicon micro-particles (micro-pSi) with size in the range of 1-10 µm are obtained by etching of silicon wafers followed by sonication. The derivatization of the micro-pSi surface by wet chemistry (silylation and coupling with a diamine) yields an interface, which exposes negative (carboxylic) or positive (amine) groups at pH 7.4. The surface modification, beyond to introduce groups for the drug loading by covalent or electrostatic interaction, stabilizes the intense orange luminescence characteristic of the silicon nano-crystallites. Derivatization by amines introduces also a second emission in the blue, which follows a different excitation pathway and can be attributed to the interface defects. The micro-pSi are efficiently internalized by human dendritic cells and do not show any toxic effect even at concentrations of 1 mg mL-1. The intrinsic luminescence of the differently functionalized micro-pSi is ...

Fuel and Energy Abstracts, 2011

Robust artificial neural network (ANN) was developed based on experimental data to predict dehydr... more Robust artificial neural network (ANN) was developed based on experimental data to predict dehydration of isopropanol by means of novel PVA-APTEOS/TEOS nanocomposite membranes in pervaporation (PV) process. The input properties were water concentration, feed temperature and nanoparticles content, while pervaporation separation index (PSI) was output. The Bayesian Regularization (BR) training method with full sampling was employed to train the network. Then, optimal ANN architecture was determined as 3:3:3:1 with log-sigmoid transfer function for hidden and output layers. The model finding revealed that nanoparticles content has significant effect on membrane performance (about 70%). The results demonstrated that the ANN model prediction and experimental data are quite match and the model can be employed with confidence for prediction of each nanocomposite membrane performance. Simulated annealing (SA) was also employed to determine controllable conditions to find the biggest PSI.

Fuel and Energy Abstracts, 2011

Robust artificial neural network (ANN) was developed based on experimental data to predict dehydr... more Robust artificial neural network (ANN) was developed based on experimental data to predict dehydration of isopropanol by means of novel PVA-APTEOS/TEOS nanocomposite membranes in pervaporation (PV) process. The input properties were water concentration, feed temperature and nanoparticles content, while pervaporation separation index (PSI) was output. The Bayesian Regularization (BR) training method with full sampling was employed to train the network. Then, optimal ANN architecture was determined as 3:3:3:1 with log-sigmoid transfer function for hidden and output layers. The model finding revealed that nanoparticles content has significant effect on membrane performance (about 70%). The results demonstrated that the ANN model prediction and experimental data are quite match and the model can be employed with confidence for prediction of each nanocomposite membrane performance. Simulated annealing (SA) was also employed to determine controllable conditions to find the biggest PSI.

Journal of Hazardous Materials, 2011

Using the Taguchi method, this study presents a systematic optimization approach for removal of l... more Using the Taguchi method, this study presents a systematic optimization approach for removal of lead (Pb) and mercury (Hg) by a nanostructure, zinc oxide-modified mesoporous carbon CMK-3 denoted as Zn-OCMK-3. CMK-3 was synthesized by using SBA-15 and then oxidized by nitric acid. The zinc oxide was loaded to the modified CMK-3 by the equilibrium adsorption of Zn(II) ions from aqueous solution followed by calcination to convert zinc nitrate to zinc oxide. The CMK-3 had porous structure and high specific surface area which can accommodate zinc oxide in a spreading manner, the zinc oxide connects to the carbon surface via oxygen atoms. The controllable factors such as agitation time, initial concentration, temperature, dose and pH of solution have been optimized. Under optimum conditions, the pollutant removal efficiency (PRE) was 97.25% for Pb(II) and 99% for Hg(II). The percentage contribution of each controllable factor was also determined. The initial concentration of pollutant is the most influential factor, and its value of percentage contribution is up to 31% and 43% for Pb and Hg, respectively. Our results show that the Zn-OCMK-3 is an effective nanoadsorbent for lead and mercury pollution remediation. Langmuir and Freundlich adsorption isotherms were used to model the equilibrium adsorption data for Pb(II) and Hg(II).

Taguchi based fuzzy logic optimization of multiple quality characteristics of cobalt disulfide nanostructures

Simulation of Structural Features on Mechanochemical Synthesis of Al2O3–TiB2 Nanocomposite by Optimized Artificial Neural Network

Antibacterial activity of silver photodeposited nepheline thin film coatings

Taguchi optimization approach for Pb(II) and Hg(II) removal from aqueous solutions using modified mesoporous carbon