... Digital-signal-processor-based waveform generator LH Sia, SS Jamuar, Roslina Mohd Sidek and M... more ... Digital-signal-processor-based waveform generator LH Sia, SS Jamuar, Roslina Mohd Sidek and Mohd Hamiruce Marhaban Department of Electrical and Electronic Engineering, University Putra Malaysia, 43400 Serdang, Selangor, Malaysia E-mail: [email protected] ...
A nonlinear model-predictive control (NMPC) is demonstrated for nonlinear systems using an improv... more A nonlinear model-predictive control (NMPC) is demonstrated for nonlinear systems using an improved fuzzy switching law. The proposed moving average filter fuzzy switching law (MAFFSL) is composed of a quasi-ARX radial basis function neural network (RBFNN) prediction model and a fuzzy switching law. An adaptive controller is designed based on a NMPC. a MAFFSL is constructed based on the system switching criterion function which is better than the (ON/OFF) switching law and a RBFNN is used to replace the neural network (NN) in the quasi-ARX black box model which is understood in terms of parameters and is not an absolute black box model, in comparison with NN. The proposed controller performance is verified through numerical simulations to demonstrate the effectiveness of the proposed method.
Abstract—A quasi-ARX (quasi-linear ARX) neural network
(QARXNN) model is able to demonstrate its ... more Abstract—A quasi-ARX (quasi-linear ARX) neural network (QARXNN) model is able to demonstrate its ability for identification and prediction highly nonlinear system. The model is simplified by a linear correlation between the input vector and its nonlinear coefficients. The coefficients are used to parameterize the input vector performed by an embedded system called as state dependent parameter estimation (SDPE), which is executed by multi layer parceptron neural network (MLPNN). SDPE consists of the linear and nonlinear parts. The controller law is derived via SDPE of the linear and nonlinear parts through switching mechanism. The dynamic tracking controller error is derived then the stability analysis of the closed-loop controller is performed based Lyapunov theorem. Linear based adaptive robust control and nonlinear based adaptive robust control is performed with the switching of the linear and nonlinear parts parameters based Lyapunov theorem to guarantee bounded and convergence error
ABSTRACT Cited By (since 1996): 1, Export Date: 4 February 2013, Source: Scopus, Language of Orig... more ABSTRACT Cited By (since 1996): 1, Export Date: 4 February 2013, Source: Scopus, Language of Original Document: English, Correspondence Address: Lutfy, O. F.; Electrical and Electronic Engineering Dept., Faculty of Engineering, University Putra Malaysia, 43400 Selangor, Malaysia; email: [email protected], References: Kiranoudis, C.T., Maroulis, Z.B., Marinos-Kouris, D., Dynamic simulation and control of conveyor-belt dryers (1994) Drying Technology, 12 (7), pp. 1575-1603;
2009 International Conference on Advanced Computer Control, 2009
Abstract In this paper, adaptive neural fuzzy inference system (ANFIS) and fuzzy subtractive clus... more Abstract In this paper, adaptive neural fuzzy inference system (ANFIS) and fuzzy subtractive clustering method (FSCM) were used to solve non-linearity, trajectory, and interaction problems of twin rotor MIMO system (TRMS). Basically, four fuzzy logic controllers (FLC) ...
2009 IEEE PES/IAS Conference on Sustainable Alternative Energy (SAE), 2009
... Multilevel Inverter for Photovoltaic Application Ehsan Esfandiari, Norman Bin Mariun, Mohd Ha... more ... Multilevel Inverter for Photovoltaic Application Ehsan Esfandiari, Norman Bin Mariun, Mohd Hamiruce Marhaban, Azmi Zakaria [email protected], [email protected], [email protected], [email protected] University Putra Malaysia Malaysia ...
This article aims to review and analyze the aspects and characteristics related to infrared food ... more This article aims to review and analyze the aspects and characteristics related to infrared food drying. Indeed, with a review of 100 relevant publications all dealing with infrared food drying, this article notes that infrared drying has several advantages over other common food drying methods. Shorter drying time, a better final dried product quality, and more energy savings in the process are revealed as the most important advantages of infrared drying over convective heat drying. Infrared dryers can also be easily combined with other drying methods such as hot air, microwave, vibration, and vacuum. This article clearly shows that using infrared heating for food drying purposes has become more popular in the last decade and its application in the industrial drying of different foodstuffs has been employed widely.
Characterization of surface electromyography using time domain features for determining hand motion and stages of contraction
Australasian Physical & Engineering Sciences in Medicine, 2014
Surface electromyography (SEMG) signals can provide important information for prosthetic hand con... more Surface electromyography (SEMG) signals can provide important information for prosthetic hand control application. In this study, time domain (TD) features were used in extracting information from the SEMG signal in determining hand motions and stages of contraction (start, middle and end). Data were collected from ten healthy subjects. Two muscles, which are flexor carpi ulnaris (FCU) and extensor carpi radialis (ECR) were assessed during three hand motions of wrist flexion (WF), wrist extension (WE) and co-contraction (CC). The SEMG signals were first segmented into 132.5Â ms windows, full wave rectified and filtered with a 6Â Hz low pass Butterworth filter. Five TD features of mean absolute value, variance, root mean square, integrated absolute value and waveform length were used for feature extraction and subsequently patterns were determined. It is concluded that the TD features that were used are able to differentiate hand motions. However, for the stages of contraction determination, although there were patterns observed, it is determined that the stages could not be properly be differentiated due to the variability of signal strengths between subjects.
This research presents a new approach to optimise the performance of a multiple-beam optical scan... more This research presents a new approach to optimise the performance of a multiple-beam optical scanning system in terms of its marking combinations and speed, using Genetic Algorithm (GA). The problem has been decomposed into two sub problems; task segregation, where the marking tasks need to be segregated and assigned for each scanner head and path planning where the best combinatorial
IEEJ Transactions on Electrical and Electronic Engineering, 2010
This work presents a self-tuning fuzzy PD controller for solving the control challenges of twin r... more This work presents a self-tuning fuzzy PD controller for solving the control challenges of twin rotor MIMO system. The controller is made adaptive through output scaling factor adjustment of the updating factor, α. The value of α is calculated directly from a fuzzy rule base defined as error and change of error of the controlled variable. A combination of adaptive neural fuzzy inference system and fuzzy subtractive clustering method was used, where the objective was to improve its time response, while reducing its computational complexity. Simulation results show performance improvement in comparison with that of the previous method. 
This paper presents a Mobile Hall Sensor Array system for the shape detection of ferromagnetic ma... more This paper presents a Mobile Hall Sensor Array system for the shape detection of ferromagnetic materials that are embedded in walls or floors. The operation of the Mobile Hall Sensor Array system is based on the principle of magnetic flux leakage to describe the shape of the ferromagnetic material. Two permanent magnets are used to generate the magnetic flux flow. The distribution of magnetic flux is perturbed as the ferromagnetic material is brought near the permanent magnets and the changes in magnetic flux distribution are detected by the 1-D array of the Hall sensor array setup. The process for magnetic imaging of the magnetic flux distribution is done by a signal processing unit before it displays the real time images using a netbook. A signal processing application software is developed for the 1-D Hall sensor array signal acquisition and processing to construct a 2-D array matrix. The processed 1-D Hall sensor array signals are later used to construct the magnetic image of fe...
EURASIP Journal on Advances in Signal Processing, 2012
In recent years, facial expression recognition (FER) has become an attractive research area, whic... more In recent years, facial expression recognition (FER) has become an attractive research area, which besides the fundamental challenges, it poses, finds application in areas, such as human-computer interaction, clinical psychology, lie detection, pain assessment, and neurology. Generally the approaches to FER consist of three main steps: face detection, feature extraction and expression recognition. The recognition accuracy of FER hinges immensely on the relevance of the selected features in representing the target expressions. In this article, we present a person and gender independent 3D facial expression recognition method, using maximum relevance minimum redundancy geometrical features. The aim is to detect a compact set of features that sufficiently represents the most discriminative features between the target classes. Multi-class one-against-one SVM classifier was employed to recognize the seven facial expressions; neutral, happy, sad, angry, fear, disgust, and surprise. The average recognition accuracy of 92.2% was recorded. Furthermore, inter database homogeneity was investigated between two independent databases the BU-3DFE and UPM-3DFE the results showed a strong homogeneity between the two databases.
ABSTRACT With their highly robust nature and simple design, switched reluctance machines are find... more ABSTRACT With their highly robust nature and simple design, switched reluctance machines are finding their way into numerous modern day applications. However, they produce oscillatory torque that generates torque ripple and mechanical vibrations. A double rotor structure to maximize the flux linkage and thereby increase the torque generating capability is proposed. As the machine operates close to saturation, the torque computation depends heavily on the energy conversion as the rotor rolls over the stator for a fixed pole pitch. The flux linkage characteristics are highly non-linear, hence estimation of the magnetic and mechanical parameters is extremely cumbersome. Magnetic circuit analysis by interpretation of the number of flux tubes using integration techniques at different positions of the machine to develop the flux linkage characteristics of the double rotor structure is presented. Computation of the inductances during the movement of rotor from unaligned to aligned is crucial in determining the generated torque. Relevant equations of calculations for inductance and flux linkages in the aligned, partially aligned and unaligned positions are computed. The partially aligned computation is based on the average on two intermediate positions, namely the 1/4th aligned and 3/4th aligned conditions. The static torque characteristics based on the energy conversion principles are used to compute the torque value. Results from simulation and experiments used for performance evaluation of the proposed flux tube analysis for computation of the electro-magnetic torque are presented.
Singularities and uncertainties in arm configurations are the main problems in kinematics robot c... more Singularities and uncertainties in arm configurations are the main problems in kinematics robot control resulting from applying robot model, a solution based on using Artificial Neural Network (ANN) is proposed here. The main idea of this approach is the use of an ANN to learn the robot system characteristics rather than having to specify an explicit robot system model.
Oil Palm Fruit Ripeness Grading System Based on Basic Gray Level Aura Matrix Feature Extraction and Support Vector Machine, <I>K</I> Nearest Neighbourhood and Artificial Neural Network Supervised Machine Learning
... Digital-signal-processor-based waveform generator LH Sia, SS Jamuar, Roslina Mohd Sidek and M... more ... Digital-signal-processor-based waveform generator LH Sia, SS Jamuar, Roslina Mohd Sidek and Mohd Hamiruce Marhaban Department of Electrical and Electronic Engineering, University Putra Malaysia, 43400 Serdang, Selangor, Malaysia E-mail: [email protected] ...
A nonlinear model-predictive control (NMPC) is demonstrated for nonlinear systems using an improv... more A nonlinear model-predictive control (NMPC) is demonstrated for nonlinear systems using an improved fuzzy switching law. The proposed moving average filter fuzzy switching law (MAFFSL) is composed of a quasi-ARX radial basis function neural network (RBFNN) prediction model and a fuzzy switching law. An adaptive controller is designed based on a NMPC. a MAFFSL is constructed based on the system switching criterion function which is better than the (ON/OFF) switching law and a RBFNN is used to replace the neural network (NN) in the quasi-ARX black box model which is understood in terms of parameters and is not an absolute black box model, in comparison with NN. The proposed controller performance is verified through numerical simulations to demonstrate the effectiveness of the proposed method.
Abstract—A quasi-ARX (quasi-linear ARX) neural network
(QARXNN) model is able to demonstrate its ... more Abstract—A quasi-ARX (quasi-linear ARX) neural network (QARXNN) model is able to demonstrate its ability for identification and prediction highly nonlinear system. The model is simplified by a linear correlation between the input vector and its nonlinear coefficients. The coefficients are used to parameterize the input vector performed by an embedded system called as state dependent parameter estimation (SDPE), which is executed by multi layer parceptron neural network (MLPNN). SDPE consists of the linear and nonlinear parts. The controller law is derived via SDPE of the linear and nonlinear parts through switching mechanism. The dynamic tracking controller error is derived then the stability analysis of the closed-loop controller is performed based Lyapunov theorem. Linear based adaptive robust control and nonlinear based adaptive robust control is performed with the switching of the linear and nonlinear parts parameters based Lyapunov theorem to guarantee bounded and convergence error
ABSTRACT Cited By (since 1996): 1, Export Date: 4 February 2013, Source: Scopus, Language of Orig... more ABSTRACT Cited By (since 1996): 1, Export Date: 4 February 2013, Source: Scopus, Language of Original Document: English, Correspondence Address: Lutfy, O. F.; Electrical and Electronic Engineering Dept., Faculty of Engineering, University Putra Malaysia, 43400 Selangor, Malaysia; email: [email protected], References: Kiranoudis, C.T., Maroulis, Z.B., Marinos-Kouris, D., Dynamic simulation and control of conveyor-belt dryers (1994) Drying Technology, 12 (7), pp. 1575-1603;
2009 International Conference on Advanced Computer Control, 2009
Abstract In this paper, adaptive neural fuzzy inference system (ANFIS) and fuzzy subtractive clus... more Abstract In this paper, adaptive neural fuzzy inference system (ANFIS) and fuzzy subtractive clustering method (FSCM) were used to solve non-linearity, trajectory, and interaction problems of twin rotor MIMO system (TRMS). Basically, four fuzzy logic controllers (FLC) ...
2009 IEEE PES/IAS Conference on Sustainable Alternative Energy (SAE), 2009
... Multilevel Inverter for Photovoltaic Application Ehsan Esfandiari, Norman Bin Mariun, Mohd Ha... more ... Multilevel Inverter for Photovoltaic Application Ehsan Esfandiari, Norman Bin Mariun, Mohd Hamiruce Marhaban, Azmi Zakaria [email protected], [email protected], [email protected], [email protected] University Putra Malaysia Malaysia ...
This article aims to review and analyze the aspects and characteristics related to infrared food ... more This article aims to review and analyze the aspects and characteristics related to infrared food drying. Indeed, with a review of 100 relevant publications all dealing with infrared food drying, this article notes that infrared drying has several advantages over other common food drying methods. Shorter drying time, a better final dried product quality, and more energy savings in the process are revealed as the most important advantages of infrared drying over convective heat drying. Infrared dryers can also be easily combined with other drying methods such as hot air, microwave, vibration, and vacuum. This article clearly shows that using infrared heating for food drying purposes has become more popular in the last decade and its application in the industrial drying of different foodstuffs has been employed widely.
Characterization of surface electromyography using time domain features for determining hand motion and stages of contraction
Australasian Physical & Engineering Sciences in Medicine, 2014
Surface electromyography (SEMG) signals can provide important information for prosthetic hand con... more Surface electromyography (SEMG) signals can provide important information for prosthetic hand control application. In this study, time domain (TD) features were used in extracting information from the SEMG signal in determining hand motions and stages of contraction (start, middle and end). Data were collected from ten healthy subjects. Two muscles, which are flexor carpi ulnaris (FCU) and extensor carpi radialis (ECR) were assessed during three hand motions of wrist flexion (WF), wrist extension (WE) and co-contraction (CC). The SEMG signals were first segmented into 132.5Â ms windows, full wave rectified and filtered with a 6Â Hz low pass Butterworth filter. Five TD features of mean absolute value, variance, root mean square, integrated absolute value and waveform length were used for feature extraction and subsequently patterns were determined. It is concluded that the TD features that were used are able to differentiate hand motions. However, for the stages of contraction determination, although there were patterns observed, it is determined that the stages could not be properly be differentiated due to the variability of signal strengths between subjects.
This research presents a new approach to optimise the performance of a multiple-beam optical scan... more This research presents a new approach to optimise the performance of a multiple-beam optical scanning system in terms of its marking combinations and speed, using Genetic Algorithm (GA). The problem has been decomposed into two sub problems; task segregation, where the marking tasks need to be segregated and assigned for each scanner head and path planning where the best combinatorial
IEEJ Transactions on Electrical and Electronic Engineering, 2010
This work presents a self-tuning fuzzy PD controller for solving the control challenges of twin r... more This work presents a self-tuning fuzzy PD controller for solving the control challenges of twin rotor MIMO system. The controller is made adaptive through output scaling factor adjustment of the updating factor, α. The value of α is calculated directly from a fuzzy rule base defined as error and change of error of the controlled variable. A combination of adaptive neural fuzzy inference system and fuzzy subtractive clustering method was used, where the objective was to improve its time response, while reducing its computational complexity. Simulation results show performance improvement in comparison with that of the previous method. 
This paper presents a Mobile Hall Sensor Array system for the shape detection of ferromagnetic ma... more This paper presents a Mobile Hall Sensor Array system for the shape detection of ferromagnetic materials that are embedded in walls or floors. The operation of the Mobile Hall Sensor Array system is based on the principle of magnetic flux leakage to describe the shape of the ferromagnetic material. Two permanent magnets are used to generate the magnetic flux flow. The distribution of magnetic flux is perturbed as the ferromagnetic material is brought near the permanent magnets and the changes in magnetic flux distribution are detected by the 1-D array of the Hall sensor array setup. The process for magnetic imaging of the magnetic flux distribution is done by a signal processing unit before it displays the real time images using a netbook. A signal processing application software is developed for the 1-D Hall sensor array signal acquisition and processing to construct a 2-D array matrix. The processed 1-D Hall sensor array signals are later used to construct the magnetic image of fe...
EURASIP Journal on Advances in Signal Processing, 2012
In recent years, facial expression recognition (FER) has become an attractive research area, whic... more In recent years, facial expression recognition (FER) has become an attractive research area, which besides the fundamental challenges, it poses, finds application in areas, such as human-computer interaction, clinical psychology, lie detection, pain assessment, and neurology. Generally the approaches to FER consist of three main steps: face detection, feature extraction and expression recognition. The recognition accuracy of FER hinges immensely on the relevance of the selected features in representing the target expressions. In this article, we present a person and gender independent 3D facial expression recognition method, using maximum relevance minimum redundancy geometrical features. The aim is to detect a compact set of features that sufficiently represents the most discriminative features between the target classes. Multi-class one-against-one SVM classifier was employed to recognize the seven facial expressions; neutral, happy, sad, angry, fear, disgust, and surprise. The average recognition accuracy of 92.2% was recorded. Furthermore, inter database homogeneity was investigated between two independent databases the BU-3DFE and UPM-3DFE the results showed a strong homogeneity between the two databases.
ABSTRACT With their highly robust nature and simple design, switched reluctance machines are find... more ABSTRACT With their highly robust nature and simple design, switched reluctance machines are finding their way into numerous modern day applications. However, they produce oscillatory torque that generates torque ripple and mechanical vibrations. A double rotor structure to maximize the flux linkage and thereby increase the torque generating capability is proposed. As the machine operates close to saturation, the torque computation depends heavily on the energy conversion as the rotor rolls over the stator for a fixed pole pitch. The flux linkage characteristics are highly non-linear, hence estimation of the magnetic and mechanical parameters is extremely cumbersome. Magnetic circuit analysis by interpretation of the number of flux tubes using integration techniques at different positions of the machine to develop the flux linkage characteristics of the double rotor structure is presented. Computation of the inductances during the movement of rotor from unaligned to aligned is crucial in determining the generated torque. Relevant equations of calculations for inductance and flux linkages in the aligned, partially aligned and unaligned positions are computed. The partially aligned computation is based on the average on two intermediate positions, namely the 1/4th aligned and 3/4th aligned conditions. The static torque characteristics based on the energy conversion principles are used to compute the torque value. Results from simulation and experiments used for performance evaluation of the proposed flux tube analysis for computation of the electro-magnetic torque are presented.
Singularities and uncertainties in arm configurations are the main problems in kinematics robot c... more Singularities and uncertainties in arm configurations are the main problems in kinematics robot control resulting from applying robot model, a solution based on using Artificial Neural Network (ANN) is proposed here. The main idea of this approach is the use of an ANN to learn the robot system characteristics rather than having to specify an explicit robot system model.
Oil Palm Fruit Ripeness Grading System Based on Basic Gray Level Aura Matrix Feature Extraction and Support Vector Machine, <I>K</I> Nearest Neighbourhood and Artificial Neural Network Supervised Machine Learning
Uploads
Papers by Mohd Marhaban
(QARXNN) model is able to demonstrate its ability for identification
and prediction highly nonlinear system. The model is
simplified by a linear correlation between the input vector and its
nonlinear coefficients. The coefficients are used to parameterize
the input vector performed by an embedded system called as state
dependent parameter estimation (SDPE), which is executed by
multi layer parceptron neural network (MLPNN). SDPE consists
of the linear and nonlinear parts. The controller law is derived
via SDPE of the linear and nonlinear parts through switching
mechanism. The dynamic tracking controller error is derived then
the stability analysis of the closed-loop controller is performed
based Lyapunov theorem. Linear based adaptive robust control
and nonlinear based adaptive robust control is performed with
the switching of the linear and nonlinear parts parameters based
Lyapunov theorem to guarantee bounded and convergence error
(QARXNN) model is able to demonstrate its ability for identification
and prediction highly nonlinear system. The model is
simplified by a linear correlation between the input vector and its
nonlinear coefficients. The coefficients are used to parameterize
the input vector performed by an embedded system called as state
dependent parameter estimation (SDPE), which is executed by
multi layer parceptron neural network (MLPNN). SDPE consists
of the linear and nonlinear parts. The controller law is derived
via SDPE of the linear and nonlinear parts through switching
mechanism. The dynamic tracking controller error is derived then
the stability analysis of the closed-loop controller is performed
based Lyapunov theorem. Linear based adaptive robust control
and nonlinear based adaptive robust control is performed with
the switching of the linear and nonlinear parts parameters based
Lyapunov theorem to guarantee bounded and convergence error