Zuraidi Saad
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Zuraidi Saad has a total of 76 co-authors. Here are the top co-authors by number of shared publications:
M. Y. Mashor
Electronic & Biomedical Intelligent Systems Research Group, School of Mechatronic Engineering, Universiti Malaysia Perlis, Pauh Putra Campus, 02600 Arau, Perlis, Malaysia
5 shared publicationsM.K. Osman
5 shared publicationsS. Omar
Faculty of Electrical Engineering, Universiti Teknologi, Mara (UiTM), Penang Campus, Jln. Pmtg Pauh, 13500, Pulau Pinang, MALAYSIA
5 shared publicationsM. K. Osman
Faculty of Electrical Engineering, Universiti Teknologi, Mara (UiTM), Penang Campus, Jln. Pmtg Pauh, 13500, Pulau Pinang, MALAYSIA
4 shared publicationsM.Y. Mashor
Electronic & Biomedical Intelligent Systems (EBItS), Research Group, School of Mechatronic Engineering, Universiti Malaysia Perlis, Ulu Pauh Main Campus, 02600 Arau, Malaysia
4 shared publicationsMuhammad Khusairi Osman
4 shared publications
Latest Publications
- Open Access
- 1 July 2020
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 19 (1) , 222-228
Classification of healthy and white root disease infected rubber trees based on relative permittivity and capacitance input properties using LM and SCG artificial neural network
White root disease is one of the most serious diseases in rubber plantation in Malaysia that originally infects on the root surface of the rubber tree. So, prevention is important compared to treatment. The classification system proposed in the research had the ability of detecting the disease by classifying between healthy rubber trees and white root disease infected rubber trees. 600 samples of latex from healthy rubber trees and white root disease infected rubber trees were taken from the RRIM station in Kota Tinggi, Johor. These samples were measured based on its relative permittivity and capacitance. All of the measurement inputs from the experiment were tested using statistical analysis. These measurement input were then went through the process of classification in ANN to generate the optimized models by using LM and SCG algorithm. There were four optimized models selected from the classification process. The accuracy from the selected most optimized models were greater than 70%. The selected most optimized models were then used to classify between healthy trees and white root infected trees based on single input categories.
- 1 July 2020
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 19 (1) , 222-228
Open AccessClassification of healthy and white root disease infected rubber trees based on relative permittivity and capacitance input properties using LM and SCG artificial neural network
White root disease is one of the most serious diseases in rubber plantation in Malaysia that originally infects on the root surface of the rubber tree. So, prevention is important compared to treatment. The classification system proposed in the research had the ability of detecting the disease by classifying between healthy rubber trees and white root disease infected rubber trees. 600 samples of latex from healthy rubber trees and white root disease infected rubber trees were taken from the RRIM station in Kota Tinggi, Johor. These samples were measured based on its relative permittivity and capacitance. All of the measurement inputs from the experiment were tested using statistical analysis. These measurement input were then went through the process of classification in ANN to generate the optimized models by using LM and SCG algorithm. There were four optimized models selected from the classification process. The accuracy from the selected most optimized models were greater than 70%. The selected most optimized models were then used to classify between healthy trees and white root infected trees based on single input categories.
- Open Access
- 14 February 2020
- Published by Association for Computing Machinery (ACM)
- p. 205-209
Abandoned Baggage Detection & Alert System Via AI and IoT
Nowadays, piece of unattended luggage there are like tens thousands of lost or left behind in the airports in every year. Due to this issue, this research work aims to develop a camera system that can detect the baggage or luggage and human and notify the authority via media social WhatsApp application when abandoned baggage detected. On top of that, Raspberry Pi 3 model B is used as a hardware where Pi camera is installed in the hardware. This research work is using deep learning method to perform the detection of the baggage and human. The Single Shot Multibox Detector (SSD) is used as the deep learning object detection algorithms for this research work to train the object detection model. The OpenCV and Tensorflow Library is a deep learning library is installed in the Raspberry Pi 3 model B minicomputer to perform the process of detection of the human and baggage. As a result, the Abandoned Baggage Detection and Alert System (ABDAS) able to detected human and baggage using computer vision and the system will notify the authority when the system detected an abandoned baggage through WhatsApp using Twilio Internet of Things (IoT) application software.
- 14 February 2020
- Published by Association for Computing Machinery (ACM)
- p. 205-209
Open AccessAbandoned Baggage Detection & Alert System Via AI and IoT
Nowadays, piece of unattended luggage there are like tens thousands of lost or left behind in the airports in every year. Due to this issue, this research work aims to develop a camera system that can detect the baggage or luggage and human and notify the authority via media social WhatsApp application when abandoned baggage detected. On top of that, Raspberry Pi 3 model B is used as a hardware where Pi camera is installed in the hardware. This research work is using deep learning method to perform the detection of the baggage and human. The Single Shot Multibox Detector (SSD) is used as the deep learning object detection algorithms for this research work to train the object detection model. The OpenCV and Tensorflow Library is a deep learning library is installed in the Raspberry Pi 3 model B minicomputer to perform the process of detection of the human and baggage. As a result, the Abandoned Baggage Detection and Alert System (ABDAS) able to detected human and baggage using computer vision and the system will notify the authority when the system detected an abandoned baggage through WhatsApp using Twilio Internet of Things (IoT) application software.
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
A Short-Term Load Forecasting of 33kV, 11kV and 415V Electrical Systems using HMLP Network
This research is utilizing three different voltages for load flow forecasting to establish a short-term online forecasting. Upon completion of this research, several neural network learning algorithms will be compared that is an Adaptive Learning Recursive Prediction Error, Modified Recursive Prediction Error, Recursive Prediction Error and Back Propagation. A network entitled Hybrid Multilayered Perceptron Network is coupled to these training algorithms. By using an on-line model, it is applied to estimate the future trend. The future trend network model is train using nonlinear autoregressive moving average with an exogenous input. The projected data is collected from the utility power supplies of 33kV, 11kV and 415V at three different locations in MARA University of Teknologi, Penang, Malaysia. Three different sets of data are applied to evaluate the performance of these learning algorithms. From the investigational results gathered, it showed that Adaptive Learning Recursive Prediction Error learning algorithm can be more enhanced the performance of other learning algorithm as an online model in the series of 0.45 dB to 9.481 dB of Mean Square Error during validation.
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
A Short-Term Load Forecasting of 33kV, 11kV and 415V Electrical Systems using HMLP Network
This research is utilizing three different voltages for load flow forecasting to establish a short-term online forecasting. Upon completion of this research, several neural network learning algorithms will be compared that is an Adaptive Learning Recursive Prediction Error, Modified Recursive Prediction Error, Recursive Prediction Error and Back Propagation. A network entitled Hybrid Multilayered Perceptron Network is coupled to these training algorithms. By using an on-line model, it is applied to estimate the future trend. The future trend network model is train using nonlinear autoregressive moving average with an exogenous input. The projected data is collected from the utility power supplies of 33kV, 11kV and 415V at three different locations in MARA University of Teknologi, Penang, Malaysia. Three different sets of data are applied to evaluate the performance of these learning algorithms. From the investigational results gathered, it showed that Adaptive Learning Recursive Prediction Error learning algorithm can be more enhanced the performance of other learning algorithm as an online model in the series of 0.45 dB to 9.481 dB of Mean Square Error during validation.
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Tajweed Classification Using Artificial Neural Network
Consistent in recitation of the Quran is very important to become a perfect Muslim. The recitation has to be performed using correct rule of Tajweed in order to avoid recitation error that may leads to the mistranslation of the recited word or sentence. As the technology grown faster and so the number of Muslims, a variety of digital forms of Quran have been produced either in the form of mobile application or computer software. It is a good phenomenon as it will encourage more Muslims to recite the Quran. However, without the Tajweed knowledge, they tend to make mistakes during recitation. The traditional method of Quran Tajweed learning is based on Talaqqi and Musyafahah. This is a manual learning technique that involves face-to-face learning process between students and teacher. In order to address this issue, a Tajweed classification model based on digital speech processing technique and artificial neural network is developed as a fundamental research in this area. This study focuses on the rule of the Noon Sakinah and Tanween for the classification of Idgham with and without Ghunnah. The dataset is developed based on the Quran recitation from well-known reciters. Mel-Frequency Cepstral Coefficient is used for the feature extraction of the recitation sample. Meanwhile, the neural network is used for the Tajweed classifier. The training process of the neural network has been evaluated using three different training algorithms - Gradient Descent with Momentum, Resilient Backpropagation and Levenberg-Marquardt optimization. From the results, it can be concluded that the highest test accuracy is obtained by the Levernberg Marquardt training algorithm (77.7%) followed by the Gradient Descent with Momentum (76.7%) and Resilient Backpropagation (73.3%).
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Tajweed Classification Using Artificial Neural Network
Consistent in recitation of the Quran is very important to become a perfect Muslim. The recitation has to be performed using correct rule of Tajweed in order to avoid recitation error that may leads to the mistranslation of the recited word or sentence. As the technology grown faster and so the number of Muslims, a variety of digital forms of Quran have been produced either in the form of mobile application or computer software. It is a good phenomenon as it will encourage more Muslims to recite the Quran. However, without the Tajweed knowledge, they tend to make mistakes during recitation. The traditional method of Quran Tajweed learning is based on Talaqqi and Musyafahah. This is a manual learning technique that involves face-to-face learning process between students and teacher. In order to address this issue, a Tajweed classification model based on digital speech processing technique and artificial neural network is developed as a fundamental research in this area. This study focuses on the rule of the Noon Sakinah and Tanween for the classification of Idgham with and without Ghunnah. The dataset is developed based on the Quran recitation from well-known reciters. Mel-Frequency Cepstral Coefficient is used for the feature extraction of the recitation sample. Meanwhile, the neural network is used for the Tajweed classifier. The training process of the neural network has been evaluated using three different training algorithms - Gradient Descent with Momentum, Resilient Backpropagation and Levenberg-Marquardt optimization. From the results, it can be concluded that the highest test accuracy is obtained by the Levernberg Marquardt training algorithm (77.7%) followed by the Gradient Descent with Momentum (76.7%) and Resilient Backpropagation (73.3%).
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Online Modelling and Forecasting of the Production of Isopropyl Myristate using TD- HMLP Neural Network
The objective of this study is to measure modelling performance using an online modelling and forecasting for the fabrication of Isopropyl Myristate in Semibatch Reactive Distillation. A network which called Trend Data Hybrid Multilayered Perceptron Network was applied to compare with conventional Hybrid Multilayered Perceptron Network. These two networks were coupled with an online learning algorithm as a nonlinear model. The input-output data for data training were determined from simulation of Isopropyl Myristate production using Aspen Plus. An online model was used to predict the percentage of Isopropyl Myristate fabrication for the determination and direction of future trends. The results of the both networks performance are based on the one step ahead forecasting, multi-step ahead forecasting and adjusted R square. The results of the multi step ahead forecasting indicated that Trend Data-Hybrid Multilayered Perceptron Network is preferable than the conventional Hybrid Multilayered Perceptron Network. Trend Data-Hybrid Multilayered Perceptron Network has improved the online forecasting performance in the generated of more promising steps in multi-step ahead forecasting.
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Online Modelling and Forecasting of the Production of Isopropyl Myristate using TD- HMLP Neural Network
The objective of this study is to measure modelling performance using an online modelling and forecasting for the fabrication of Isopropyl Myristate in Semibatch Reactive Distillation. A network which called Trend Data Hybrid Multilayered Perceptron Network was applied to compare with conventional Hybrid Multilayered Perceptron Network. These two networks were coupled with an online learning algorithm as a nonlinear model. The input-output data for data training were determined from simulation of Isopropyl Myristate production using Aspen Plus. An online model was used to predict the percentage of Isopropyl Myristate fabrication for the determination and direction of future trends. The results of the both networks performance are based on the one step ahead forecasting, multi-step ahead forecasting and adjusted R square. The results of the multi step ahead forecasting indicated that Trend Data-Hybrid Multilayered Perceptron Network is preferable than the conventional Hybrid Multilayered Perceptron Network. Trend Data-Hybrid Multilayered Perceptron Network has improved the online forecasting performance in the generated of more promising steps in multi-step ahead forecasting.
- 1 November 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Short-term load forecasting of 415V, 11kV and 33kV electrical systems using MLP network
This research explores a short-term on-line forecasting for load-flow forecasting of three different voltage systems. Upon completing this study, four training algorithm of neural network is used namely Back Propagation (BP), Recursive Prediction Error (RPE), Modified Recursive Prediction Error (MRPE) and Adaptive Learning Recursive Prediction Error (ALRPE). All these training algorithms are performed by using a structured network called Multilayered Perceptron Network (MLP). An on-line MLP is used to predict the usage of electrical power demand. Non-linear autoregressive moving average with an exogenous input model (NARMAX) is selected to train the network. The analyzed data are collected from some power load usage of 415V, 11kV and 33kV systems at UiTM Pulau Pinang. These data sets are used to compare the performance of MLP with different types of learning algorithm. Experimental results showed that ALRPE training algorithm can further improved the performance of non-linear MLP model in the range of 1.02 dB to 3.224 dB of mean square error (MSE) in the model validation.
- 1 November 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Short-term load forecasting of 415V, 11kV and 33kV electrical systems using MLP network
This research explores a short-term on-line forecasting for load-flow forecasting of three different voltage systems. Upon completing this study, four training algorithm of neural network is used namely Back Propagation (BP), Recursive Prediction Error (RPE), Modified Recursive Prediction Error (MRPE) and Adaptive Learning Recursive Prediction Error (ALRPE). All these training algorithms are performed by using a structured network called Multilayered Perceptron Network (MLP). An on-line MLP is used to predict the usage of electrical power demand. Non-linear autoregressive moving average with an exogenous input model (NARMAX) is selected to train the network. The analyzed data are collected from some power load usage of 415V, 11kV and 33kV systems at UiTM Pulau Pinang. These data sets are used to compare the performance of MLP with different types of learning algorithm. Experimental results showed that ALRPE training algorithm can further improved the performance of non-linear MLP model in the range of 1.02 dB to 3.224 dB of mean square error (MSE) in the model validation.
- 10 November 2015
- Published by Springer Nature
- p. 268-275
Formation of Momentum and Learning Rate Profile for Online Training and Testing of HMLP with ALRPE
- 10 November 2015
- Published by Springer Nature
- p. 268-275
Formation of Momentum and Learning Rate Profile for Online Training and Testing of HMLP with ALRPE
- 10 November 2015
- Published by Springer Nature
- p. 198-205
Enhanced Genetic Algorithm Applied for Global Optimization
- 10 November 2015
- Published by Springer Nature
- p. 198-205
Enhanced Genetic Algorithm Applied for Global Optimization
- 13 October 2015
- Published by Penerbit UTM Press in Jurnal Teknologi (Sciences & Engineering)
- Vol. 76 (12)
ON-LINE MODELLING AND FORECASTING OF A MOVING CAR INFORMATION USING TD-HMLP NETWORK
The study proposed a model called trend data hybrid multilayered perceptron network (TD-HMLP) coupled with a modified recursive prediction error (MRPE) training algorithm as a nonlinear modeling. An on-line model was used to forecast speed, revolution and fuel balanced in a Proton Gen2 car tank. The car measured the injected fuel from fuel injection sensor and become an input for the TD-HMLP model to forecast the speed, revolution and fuel balanced in tank. These forecasted variables were also measured from the car sensors. The criterions for performances are based on the one step ahead forecasting (OSA), multi-step ahead forecasting (MSA) and adjusted R2. The forecasting result showed that TD-HMLP network is better than the conventional HMLP network to maintain higher value in adjusted R2 and produce better step in multi-step ahead forecasting. These preliminary results show that the proposed modeling approach is capable to be used as an on-line information forecaster of a moving car.
- 13 October 2015
- Published by Penerbit UTM Press in Jurnal Teknologi (Sciences & Engineering)
- Vol. 76 (12)
ON-LINE MODELLING AND FORECASTING OF A MOVING CAR INFORMATION USING TD-HMLP NETWORK
The study proposed a model called trend data hybrid multilayered perceptron network (TD-HMLP) coupled with a modified recursive prediction error (MRPE) training algorithm as a nonlinear modeling. An on-line model was used to forecast speed, revolution and fuel balanced in a Proton Gen2 car tank. The car measured the injected fuel from fuel injection sensor and become an input for the TD-HMLP model to forecast the speed, revolution and fuel balanced in tank. These forecasted variables were also measured from the car sensors. The criterions for performances are based on the one step ahead forecasting (OSA), multi-step ahead forecasting (MSA) and adjusted R2. The forecasting result showed that TD-HMLP network is better than the conventional HMLP network to maintain higher value in adjusted R2 and produce better step in multi-step ahead forecasting. These preliminary results show that the proposed modeling approach is capable to be used as an on-line information forecaster of a moving car.