M.A. Rafiee , J. Rafiee (original) (raw)
Papers by M.A. Rafiee , J. Rafiee
Bookmarks Related papers MentionsView impact
A Rigaku D/Max 2500 XRD with Cu K α radiation (λ = 1.54 Å) at a generator voltage of 40 kV and a ... more A Rigaku D/Max 2500 XRD with Cu K α radiation (λ = 1.54 Å) at a generator voltage of 40 kV and a generator current of 50 mA was used to measure the diffraction behavior of natural graphite and graphite oxide. All experiments were carried out in the reflection mode at ambient temperature with 2θ varying between 1 and 30°. The scanning speed was 2.4°/min, and the step size was 0.002°. The figure below shows XRD patterns of natural graphite (a) and graphite oxide (b).
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
A B S T R A C T Polytetrafluoroethylene (PTFE) is one of the most widely used solid lubricants bu... more A B S T R A C T Polytetrafluoroethylene (PTFE) is one of the most widely used solid lubricants but suffers from a high wear rate which limits its applications. Here we report four orders of magnitude reduction in the steady state wear rate of PTFE due to graphene additives. The wear rate of unfilled PTFE was measured to be 0.4⋅10Aˋ3mm3/Nmwhichisreducedto0.4 · 10 À3 mm 3 /N m which is reduced to 0.4⋅10Aˋ3mm3/Nmwhichisreducedto10 À7 mm 3 /N m by the incorporation of 10 wt% of graphene platelets. We also performed a head-to-head comparison of wear rate with graphene and micro-graphite fillers at the same weight fractions. In general, we find that graphene fillers gave 10–30 times lower wear rates than micro-graphite at the same loading fraction. Scanning electron microscopy analysis indicated noticeably smaller wear debris size in the case of graphene/PTFE composites indicating that graphene additives are highly effective in regulating debris formation in PTFE leading to reduced wear.
Bookmarks Related papers MentionsView impact
Polymer nanocomposites reinforced by carbon nanotubes, fullerene and nanoparticles have been broa... more Polymer nanocomposites reinforced by carbon nanotubes, fullerene and nanoparticles have been broadly studied within the last two decades. However, it was recently observed that polymer nanocomposites filled with graphene sheets showed exceptional mechanical and electrical properties. The advantage of graphene sheets over carbon nanotubes in nanocomposites may be related to their high specific surface area, enhanced nanofiller-matrix adhesion/interlocking arising from their wrinkled (rough) surface as well as the two-dimensional geometry of graphene sheets. We have compared the vibration damping properties of epoxy nanocomposite filled with single-walled carbon nanotubes (SWNT), multi-walled nanotubes (MWNT), and graphene platelet (GPL) fillers. Our results show the evidence of Coulomb friction damping in nanocomposites comparing with the pure epoxy.
Bookmarks Related papers MentionsView impact
It is well established that in-plane tensile properties of unidirectional microfiber-reinforced c... more It is well established that in-plane tensile properties of unidirectional microfiber-reinforced composites are not significantly influenced by addition of carbon nanotubes to the matrix. This is because the principal effect of the nanotubes is to enhance the matrix dominated (out-of-plane) properties. Here we report that the above situation changes when nanotubes are incorporated into woven-fabric (textile) composites. We report up to 200% increase in strain-to-break and 180% increase in toughness under in-plane tensile load with ∼0.05% weight of nanotube additives. We attribute this effect to the geometrical arrangement of the micro-fibers and the critical role of the pure-matrix-block in textile composites.
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Dependence of the yield and fatigue strength of steel bolts with composition in accordance to AIS... more Dependence of the yield and fatigue strength of steel bolts with composition in accordance to AISI 1035 manufactured by thread rolling and machining process on dislocation density were investigated. The results indicate that the fatigue strength of the rolled bolts are 55% higher than the machined bolts and by full anneal-ing at 850° C, it reduced to the extent of machined specimen. Partial annealing of the thread rolled bolts at 680° C caused a reduction of fatigue strength by approximately 61% due to reduction in the dislocation density. Fatigue strength was improved by deformation rate (i.e., rolling speed), which is also due to the increasing dislocation density. Yield stress of the studied specimens followed the same pattern as fatigue strength. Considering the obtained results from the low and high speed, partial and full annealed thread rolled specimens, yield stress of the thread rolled bolts has been modeled based on the dislocation density. The obtained results from the model are in good agreement with the experimental results. The contribution to fatigue strength by thread rolling stems from the strain hardening effect which would facilitate the formation of compressive residual stress near the surface layer. The strengthening may be attributed to increasing dislocation density in the ferrite phase (i.e., substructure formation), in addition to the formation of a fine layered structure consisting of elongated pearlite colonies and ferrite grains.
Bookmarks Related papers MentionsView impact
Herein, we report a multifunctional n-type emitter with strong green luminescence of polybenzodio... more Herein, we report a multifunctional n-type emitter with strong green luminescence of polybenzodioxane polymer (PIM-1) and their suitability as an electron transport layer for OLEDs devices. The Brunauer–Emmett–Teller (BET) test and photo-electrical properties of as-synthesized PIM-1 confirm the presence of large microporosity and excellent electron mobility. The photoluminescence (PL) spectroscopy shows the intense green emission at 10 515 nm upon 332 nm excitation wavelength. Moreover, the Hall Effect study reveals the negative Hall resistivity that indicates the PIM-1 consisting n-type semiconductor characteristic. It enables the highly-efficient polymer-based green LEDs with configuration; ITO(120 nm)/PEDOT:PSS(30 nm)/PIM-1(100 nm)/LiF(1 nm)/Al(150 nm), which are fabricated by sequential solution-processing method. The OLED incorporating PIM-1 thin layer has achieved maximum current efficiency of 1.71 Cd/A and power efficiency of 0.49 lm/W. Additionally, the 15 electron mobility is found to be 4.4x10-6 cm 2 /Vs. Hence, these results demonstrate that PIM-1 could be an ultimate choice as an n-type emitter for awaited next generation advanced electronic devices.
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
⎯ In this research, an automatic optimized ANN-based feature recognition system was proposed whic... more ⎯ In this research, an automatic optimized ANN-based feature recognition system was proposed which due to its capability in identifying simple and intricate features is fully integratable with any feature-based application e.g. computer-aided process planning (CAPP), computer-aided manufacturing (CAM). Boundary representation (B-rep) of CAD data was used to obtain face-scores which were applied to a multilayer perceptron network to detect and recognize the features from the CAD database as they convey invaluable geometrical and topological information of the 3D model. A genetic algorithm (GA) was applied to the network to deliver an entirely optimized recognition system. The genetic algorithm was used to optimize the network architecture and weights to impart precision and promptness to the system. Back propagation and GA was integrated to reduce computation time by training the network using the optimized initial weights.
Bookmarks Related papers MentionsView impact
—This paper experimentally investigates the type of tooth defects in a gear of gearbox system usi... more —This paper experimentally investigates the type of tooth defects in a gear of gearbox system using neural networks with a astonishing high accuracy. Slight-worn, medium-worn, and broken-teeth of gears were considered as gears faults. In almost most of fault-diagnosis methods, vibration signals of a machine are predominantly used to detect mechanical faults. A new feature vector for training purposes of neural network is also presented using energy coefficients of wavelet packet from these signatures. Ultimately a MLP network was presented that offers not only an undersized configuration but also high performance giving it the capability to identify faults perfectly in 99% of the circumstances.
Bookmarks Related papers MentionsView impact
— This paper presents the development of a novel fault detection procedure for ball bearings of g... more — This paper presents the development of a novel fault detection procedure for ball bearings of gearbox systems which involves a combination of statistical characteristics, wavelet analysis and artificial neural network methods with 100% prefect performance. In order to test the proposed procedure, experimentally obtained vibration data from a bearing of gearbox was used to exploit a new-brand diagnostic system. Standard deviation of wavelet packet coefficients was applied as a new feature vector to train the neural network.
Bookmarks Related papers MentionsView impact
— This paper presents Daubechies 44 (db44) as a multipurpose mother wavelet function for complex ... more — This paper presents Daubechies 44 (db44) as a multipurpose mother wavelet function for complex signals of human beings. During the last two decades, wavelet transform has been quickly developed as one of the most powerful processors in various areas of science and technology. Many papers have been presented with different types of mother wavelet functions in various fields based on different criteria such as similarity between signals and mother wavelet. To name a few, machine condition monitoring including vibration, acoustic and ultrasonic signals, image processing, Electromyographic (EMG) and Electroencephalographic (EEG) signals have been taken into consideration using wavelet transform. However, finding the optimal wavelet function represents as ongoing challenge in this area. Therefore, this paper focuses on finding the optimal mother wavelet function for complicated human biological signals. In this research, three measures were analyzed. These include surface and intramuscular EMG of the upper limb and EEG in response to visual stimuli. 324 mother wavelet functions from wavelet families including Haar, Daubechies (db), Symlet, Coiflet, Gaussian, Morlet, complex Morlet, Mexican hat, bio-orthogonal, reverse bio-orthogonal, Meyer, discrete approximation of Meyer, complex Gaussian, Shannon, and frequency B-spline were investigated. Higher correlations, and thus greater similarity, is indicated by more values of wavelet coefficients. Thus, the best mother wavelet was identified as the mother wavelet with more values in wavelet coefficients. The db44 wavelet was identified as the best mother wavelet across both surface and intramuscular EMG and EEG signals.
Bookmarks Related papers MentionsView impact
2005 International Conference on Neural Networks and Brain, 2005
This paper concentrates on a new procedure which experimentally recognizes gears and bearings fau... more This paper concentrates on a new procedure which experimentally recognizes gears and bearings faults of a typical gearbox system using a multi-layer perceptron neural network. Feature vector which is one of the most significant parameters to design an appropriate neural network was innovated by standard deviation of wavelet packet coefficients. The gear conditions were considered to be normal gearbox and
Bookmarks Related papers MentionsView impact
Volume 13: Sound, Vibration and Design, 2010
ABSTRACT
Bookmarks Related papers MentionsView impact
2009 IEEE 35th Annual Northeast Bioengineering Conference, 2009
ABSTRACT This paper presents a new technique to optimally classify forearm electromyographic (EMG... more ABSTRACT This paper presents a new technique to optimally classify forearm electromyographic (EMG) measures using proposed mother wavelet matrix (MWM) for biorobots. Among 324 mother wavelet candidates, a MWM including 15 potential mother wavelet functions were selected to optimally classify surface and intramuscular EMG signals collected from multiple locations on the upper forearm of six subjects for nine classes of hand motions plus a rest state.
Bookmarks Related papers MentionsView impact
2009 2nd International Conference on Computer, Control and Communication, 2009
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
A Rigaku D/Max 2500 XRD with Cu K α radiation (λ = 1.54 Å) at a generator voltage of 40 kV and a ... more A Rigaku D/Max 2500 XRD with Cu K α radiation (λ = 1.54 Å) at a generator voltage of 40 kV and a generator current of 50 mA was used to measure the diffraction behavior of natural graphite and graphite oxide. All experiments were carried out in the reflection mode at ambient temperature with 2θ varying between 1 and 30°. The scanning speed was 2.4°/min, and the step size was 0.002°. The figure below shows XRD patterns of natural graphite (a) and graphite oxide (b).
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
A B S T R A C T Polytetrafluoroethylene (PTFE) is one of the most widely used solid lubricants bu... more A B S T R A C T Polytetrafluoroethylene (PTFE) is one of the most widely used solid lubricants but suffers from a high wear rate which limits its applications. Here we report four orders of magnitude reduction in the steady state wear rate of PTFE due to graphene additives. The wear rate of unfilled PTFE was measured to be 0.4⋅10Aˋ3mm3/Nmwhichisreducedto0.4 · 10 À3 mm 3 /N m which is reduced to 0.4⋅10Aˋ3mm3/Nmwhichisreducedto10 À7 mm 3 /N m by the incorporation of 10 wt% of graphene platelets. We also performed a head-to-head comparison of wear rate with graphene and micro-graphite fillers at the same weight fractions. In general, we find that graphene fillers gave 10–30 times lower wear rates than micro-graphite at the same loading fraction. Scanning electron microscopy analysis indicated noticeably smaller wear debris size in the case of graphene/PTFE composites indicating that graphene additives are highly effective in regulating debris formation in PTFE leading to reduced wear.
Bookmarks Related papers MentionsView impact
Polymer nanocomposites reinforced by carbon nanotubes, fullerene and nanoparticles have been broa... more Polymer nanocomposites reinforced by carbon nanotubes, fullerene and nanoparticles have been broadly studied within the last two decades. However, it was recently observed that polymer nanocomposites filled with graphene sheets showed exceptional mechanical and electrical properties. The advantage of graphene sheets over carbon nanotubes in nanocomposites may be related to their high specific surface area, enhanced nanofiller-matrix adhesion/interlocking arising from their wrinkled (rough) surface as well as the two-dimensional geometry of graphene sheets. We have compared the vibration damping properties of epoxy nanocomposite filled with single-walled carbon nanotubes (SWNT), multi-walled nanotubes (MWNT), and graphene platelet (GPL) fillers. Our results show the evidence of Coulomb friction damping in nanocomposites comparing with the pure epoxy.
Bookmarks Related papers MentionsView impact
It is well established that in-plane tensile properties of unidirectional microfiber-reinforced c... more It is well established that in-plane tensile properties of unidirectional microfiber-reinforced composites are not significantly influenced by addition of carbon nanotubes to the matrix. This is because the principal effect of the nanotubes is to enhance the matrix dominated (out-of-plane) properties. Here we report that the above situation changes when nanotubes are incorporated into woven-fabric (textile) composites. We report up to 200% increase in strain-to-break and 180% increase in toughness under in-plane tensile load with ∼0.05% weight of nanotube additives. We attribute this effect to the geometrical arrangement of the micro-fibers and the critical role of the pure-matrix-block in textile composites.
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Dependence of the yield and fatigue strength of steel bolts with composition in accordance to AIS... more Dependence of the yield and fatigue strength of steel bolts with composition in accordance to AISI 1035 manufactured by thread rolling and machining process on dislocation density were investigated. The results indicate that the fatigue strength of the rolled bolts are 55% higher than the machined bolts and by full anneal-ing at 850° C, it reduced to the extent of machined specimen. Partial annealing of the thread rolled bolts at 680° C caused a reduction of fatigue strength by approximately 61% due to reduction in the dislocation density. Fatigue strength was improved by deformation rate (i.e., rolling speed), which is also due to the increasing dislocation density. Yield stress of the studied specimens followed the same pattern as fatigue strength. Considering the obtained results from the low and high speed, partial and full annealed thread rolled specimens, yield stress of the thread rolled bolts has been modeled based on the dislocation density. The obtained results from the model are in good agreement with the experimental results. The contribution to fatigue strength by thread rolling stems from the strain hardening effect which would facilitate the formation of compressive residual stress near the surface layer. The strengthening may be attributed to increasing dislocation density in the ferrite phase (i.e., substructure formation), in addition to the formation of a fine layered structure consisting of elongated pearlite colonies and ferrite grains.
Bookmarks Related papers MentionsView impact
Herein, we report a multifunctional n-type emitter with strong green luminescence of polybenzodio... more Herein, we report a multifunctional n-type emitter with strong green luminescence of polybenzodioxane polymer (PIM-1) and their suitability as an electron transport layer for OLEDs devices. The Brunauer–Emmett–Teller (BET) test and photo-electrical properties of as-synthesized PIM-1 confirm the presence of large microporosity and excellent electron mobility. The photoluminescence (PL) spectroscopy shows the intense green emission at 10 515 nm upon 332 nm excitation wavelength. Moreover, the Hall Effect study reveals the negative Hall resistivity that indicates the PIM-1 consisting n-type semiconductor characteristic. It enables the highly-efficient polymer-based green LEDs with configuration; ITO(120 nm)/PEDOT:PSS(30 nm)/PIM-1(100 nm)/LiF(1 nm)/Al(150 nm), which are fabricated by sequential solution-processing method. The OLED incorporating PIM-1 thin layer has achieved maximum current efficiency of 1.71 Cd/A and power efficiency of 0.49 lm/W. Additionally, the 15 electron mobility is found to be 4.4x10-6 cm 2 /Vs. Hence, these results demonstrate that PIM-1 could be an ultimate choice as an n-type emitter for awaited next generation advanced electronic devices.
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
⎯ In this research, an automatic optimized ANN-based feature recognition system was proposed whic... more ⎯ In this research, an automatic optimized ANN-based feature recognition system was proposed which due to its capability in identifying simple and intricate features is fully integratable with any feature-based application e.g. computer-aided process planning (CAPP), computer-aided manufacturing (CAM). Boundary representation (B-rep) of CAD data was used to obtain face-scores which were applied to a multilayer perceptron network to detect and recognize the features from the CAD database as they convey invaluable geometrical and topological information of the 3D model. A genetic algorithm (GA) was applied to the network to deliver an entirely optimized recognition system. The genetic algorithm was used to optimize the network architecture and weights to impart precision and promptness to the system. Back propagation and GA was integrated to reduce computation time by training the network using the optimized initial weights.
Bookmarks Related papers MentionsView impact
—This paper experimentally investigates the type of tooth defects in a gear of gearbox system usi... more —This paper experimentally investigates the type of tooth defects in a gear of gearbox system using neural networks with a astonishing high accuracy. Slight-worn, medium-worn, and broken-teeth of gears were considered as gears faults. In almost most of fault-diagnosis methods, vibration signals of a machine are predominantly used to detect mechanical faults. A new feature vector for training purposes of neural network is also presented using energy coefficients of wavelet packet from these signatures. Ultimately a MLP network was presented that offers not only an undersized configuration but also high performance giving it the capability to identify faults perfectly in 99% of the circumstances.
Bookmarks Related papers MentionsView impact
— This paper presents the development of a novel fault detection procedure for ball bearings of g... more — This paper presents the development of a novel fault detection procedure for ball bearings of gearbox systems which involves a combination of statistical characteristics, wavelet analysis and artificial neural network methods with 100% prefect performance. In order to test the proposed procedure, experimentally obtained vibration data from a bearing of gearbox was used to exploit a new-brand diagnostic system. Standard deviation of wavelet packet coefficients was applied as a new feature vector to train the neural network.
Bookmarks Related papers MentionsView impact
— This paper presents Daubechies 44 (db44) as a multipurpose mother wavelet function for complex ... more — This paper presents Daubechies 44 (db44) as a multipurpose mother wavelet function for complex signals of human beings. During the last two decades, wavelet transform has been quickly developed as one of the most powerful processors in various areas of science and technology. Many papers have been presented with different types of mother wavelet functions in various fields based on different criteria such as similarity between signals and mother wavelet. To name a few, machine condition monitoring including vibration, acoustic and ultrasonic signals, image processing, Electromyographic (EMG) and Electroencephalographic (EEG) signals have been taken into consideration using wavelet transform. However, finding the optimal wavelet function represents as ongoing challenge in this area. Therefore, this paper focuses on finding the optimal mother wavelet function for complicated human biological signals. In this research, three measures were analyzed. These include surface and intramuscular EMG of the upper limb and EEG in response to visual stimuli. 324 mother wavelet functions from wavelet families including Haar, Daubechies (db), Symlet, Coiflet, Gaussian, Morlet, complex Morlet, Mexican hat, bio-orthogonal, reverse bio-orthogonal, Meyer, discrete approximation of Meyer, complex Gaussian, Shannon, and frequency B-spline were investigated. Higher correlations, and thus greater similarity, is indicated by more values of wavelet coefficients. Thus, the best mother wavelet was identified as the mother wavelet with more values in wavelet coefficients. The db44 wavelet was identified as the best mother wavelet across both surface and intramuscular EMG and EEG signals.
Bookmarks Related papers MentionsView impact
2005 International Conference on Neural Networks and Brain, 2005
This paper concentrates on a new procedure which experimentally recognizes gears and bearings fau... more This paper concentrates on a new procedure which experimentally recognizes gears and bearings faults of a typical gearbox system using a multi-layer perceptron neural network. Feature vector which is one of the most significant parameters to design an appropriate neural network was innovated by standard deviation of wavelet packet coefficients. The gear conditions were considered to be normal gearbox and
Bookmarks Related papers MentionsView impact
Volume 13: Sound, Vibration and Design, 2010
ABSTRACT
Bookmarks Related papers MentionsView impact
2009 IEEE 35th Annual Northeast Bioengineering Conference, 2009
ABSTRACT This paper presents a new technique to optimally classify forearm electromyographic (EMG... more ABSTRACT This paper presents a new technique to optimally classify forearm electromyographic (EMG) measures using proposed mother wavelet matrix (MWM) for biorobots. Among 324 mother wavelet candidates, a MWM including 15 potential mother wavelet functions were selected to optimally classify surface and intramuscular EMG signals collected from multiple locations on the upper forearm of six subjects for nine classes of hand motions plus a rest state.
Bookmarks Related papers MentionsView impact
2009 2nd International Conference on Computer, Control and Communication, 2009
Bookmarks Related papers MentionsView impact