Mohd Zakimi Zakaria
Thumbnail Image
Preferred name
Mohd Zakimi Zakaria
Official Name
Mohd Zakimi , Zakaria
Alternative Name
Zakaria, Mohd Z.
Zakaria, M. Z.
Main Affiliation
Scopus Author ID
36999011800
Researcher ID
D-7223-2015

Research Output

Filters

9 3
21
Reset filters

Settings
Now showing 1 - 10 of 21
No Thumbnail Available
Publication

A comparative study of numerical modelling and analysis for large articulated pendulums

2025-05 , Siti Fatimah Azzahra Ahmad Noh , Mohamad Ezral Baharudin , Mohd Zakimi Zakaria , Mohd Sazli Saad , Azuwir Mohd Nor

In this article, we present a large system of multiple pendulums, also articulated pendulums, with twenty pendulums as a multibody model. The main objective of the study is to compare the computational time efficiency of two multibody formulations: the augmented Lagrangian and the recursive method for each articulated system. The equations of motion were derived for each formulation and the fourth- and fifth-order Runge-Kutta methods were utilised to solve for the equations by representing the kinematics and dynamics of the systems numerically. The computational times that corresponded to the manipulated step size and tolerance were compared for both formulations. The results showed that the augmented Lagrangian formulation had a significant divergence towards the negative y-axis at tolerance 0.1s for all modified step sizes. The animations also demonstrated elongation for specific pendulums based on the step size selection at a tolerance 0.1s. The recursive method, on the other hand, produced the best-fit plots and stable results for all xy-position and velocity-time plots for each adjusted step size and tolerance. Therefore, the recursive method is concluded to be more efficient than the augmented Lagrangian formulation in solving large open-loop multibody systems.

View
Thumbnail Image
Publication

ANN-Based Predictive Modelling for Fused Deposition Modelling: Material Consumption, Tensile Strength & Dimensional Accuracy

2023-01-01 , Irazman H.N.H. , Mohd Sazli Saad , Mohamad Ezral Baharudin , Mohd Zakimi Zakaria , Nor A.M. , Rahim Y.A.

Conventional modelling approaches fall short of accurately capturing the complexities of Fused Deposition Modelling (FDM). This research proposes an Artificial Neural Network (ANN) model to predict the FDM process's material consumption, tensile strength, and dimensional accuracy. Inputs such as layer height, infill density, printing temperature, and printing speed are considered. A Face-Centered Central Composite Design (FCCCD) with 78 specimens is employed to design experiments (DOE). Material consumption is measured using a densimeter, while tensile strength is determined using a Universal Testing Machine (UTM). The performance of the ANN models is evaluated based on metrics like Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and coefficient of determination (R2). The optimal ANN structure for material consumption prediction is found to be 4-19-14-1, achieving a low MSE of 0.00096. For tensile strength prediction, the best ANN structure is determined as 4-16-15-12-1 with an MSE of 0.005274145. Furthermore, dimensional accuracy is successfully captured using a 4-12-12-11-1 network configuration, which attains the lowest overall MSE of 0.002898. The proposed ANN model provides accurate predictions for material consumption, tensile strength, and dimensional accuracy in the FDM process. This study contributes to the optimization and understanding of FDM manufacturing processes through the utilization of optimized network architectures. The findings demonstrate the efficacy of the ANN model in improving FDM process control and performance.

View
Thumbnail Image
Publication

Prediction of the material consumption of PLA plus fused deposition models using artificial neural network technique

2024-04-22 , Nasuha H. , Mohd Sazli Saad , Mohamad Ezral Baharudin , Azuwir Mohd Nor , Mohd Zakimi Zakaria

Fused Deposition Modelling (FDM) is a complex additive manufacturing (AM) process involving multiple process parameters incapable of being modelled with conventional methods such as regression and mathematical modelling. The goal of the study is to develop an Artificial Neural Network (ANN) model that can accurately predict the material consumption of FDM printed parts considering the effect of process parameters such as layer height, infill density, printing temperature, and printing speed to create an ideal model that can optimize the use of resources and reduce material. The experiment was designed using face centered central composite design (FCCCD) yielding 78 specimens that were weighed using a densimeter to identify material consumption. Then, three networks with a different number of hidden layers and neurons were trained to identify the best-performing ANN structure with the lowest mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE) and highest coefficient of determination (R2). The fittest models were modelled and compared to identify the best-performing structure. Results indicated that the ANN model with double hidden layers with 19 and 14 neurons each showed the most precise prediction in modelling material consumption with the lowest MSE of 0.00096.

View
Thumbnail Image
Publication

Medium sized industrial motor solutions to mitigate the issue of high inrush starting current

2021-05-03 , Azizan N.S. , Azizan M.M. , Muhammad Izuan Fahmi Romli , Muhammad Zaid Aihsan , Muzammil Jusoh , Nashrul Fazli Mohd Nasir , Mohd Zakimi Zakaria

Starting of a medium or large induction motors generate such a large current during direct-on-line (DOL) starting process to the point that it can drop the voltage of power supply. The induction motor can be broken, its characteristics can be changed, and performances of the motor can be worsened. A significant higher starting current than the rated current can generate mechanical and thermal stress on the motor and the loads. High-voltage fluctuations, dips and sags can arise in electrical power systems associated with the motor. To overcome this problem, various starters were designed. There are several types of starters in which can be divided into conventional starters and power electronic drives. Conventional starters are such as direct-online, star-delta and autotransformer whereas the example for power electronics are matrix converter, frequency inverter and soft starter. In this paper, the design of autotransformer and soft starter are focused in order to compare inrush current during the start-up three phase medium sized industrial induction motor by using MATLAB/Simulink software. Both starters were targeted to resolve the problems inherent in the dynamic operation of induction motors, which included current and torque surges during the motor start up. The autotransformer gives the choice to consumer in selecting different tap values, in which is the advantage to consumer to vary their starting current and starting torque according to its application. On the other hand, the three-phase soft starter employs two anti-parallel connected switches in each phase. Thyristors act as the switches due to their higher power rating and high efficiency. Then, both methods will be compared to prove the best performance in mitigate high inrush current. The best of the two is likely to be the answer for mitigating the issue of high inrush starting current for a medium size motor at industries.

View
No Thumbnail Available
Publication

Perturbation parameters tuning of multi-objective optimization differential evolution and its application to dynamic system modeling

2015 , Mohd Zakimi Zakaria , Hishamuddin Jamaluddin , Robiah Ahmad , Azmi Harun , Radhwan Hussin , Ahmad Nabil Mohd Khalil , Muhammad Khairy Md Naim , Ahmad Faizal Annuar

This paper presents perturbation parameters for tuning of multi-objective optimization differential evolution and its application to dynamic system modeling. The perturbation of the proposed algorithm was composed of crossover and mutation operators.  Initially, a set of parameter values was tuned vigorously by executing multiple runs of algorithm for each proposed parameter variation. A set of values for crossover and mutation rates were proposed in executing the algorithm for model structure selection in dynamic system modeling. The model structure selection was one of the procedures in the system identification technique. Most researchers focused on the problem in selecting the parsimony model as the best represented the dynamic systems. Therefore, this problem needed two objective functions to overcome it, i.e. minimum predictive error and model complexity.  One of the main problems in identification of dynamic systems is to select the minimal model from the huge possible models that need to be considered. Hence, the important concepts in selecting good and adequate model used in the proposed algorithm were elaborated, including the implementation of the algorithm for modeling dynamic systems. Besides, the results showed that multi-objective optimization differential evolution performed better with tuned perturbation parameters.

View
Thumbnail Image
Publication

Comparison of Algebraic Reconstruction Technique Methods and Generative Adversarial Network in Image Reconstruction of Magnetic Induction Tomography (MIT)

2021-11-25 , Lubis A.J. , Nashrul Fazli Mohd Nasir , Zulkarnay Zakaria , Muzammil Jusoh , Azizan M.M. , Mohd Zakimi Zakaria , Mohd Nor Fakhzan Mohd Kazim , Rahman S.

Magnetic induction tomography (MIT) is a technique used for imaging electromagnetic properties of objects using eddy current effects. The non-linear characteristics had led to more difficulties with its solution especially in dealing with low conductivity imaging materials such as biological tissues. Two methods that could be applied for MIT image processing which is the Generative Adversarial Network (GAN) and the Algebraic Reconstruction Technique (ART). ART is widely used in the industry due to its ability to improve the quality of the reconstructed image at a high scanning speed. GAN is an intelligent method which would be able to carry out the training process. In the GAN method, the MIT principle is used to find the optimum global conductivity distribution and it is described as a training process and later, reconstructed by a generator. The output is an approximate reconstruction of the distribution's internal conductivity image. Then, the results were compared with the previous traditional algorithm, namely the regularization algorithm of BPNN and Tikhonov Regularization method. It turned out that GAN had able to adjust the non-linear relationship between input and output. GAN was also able to solve non-linear problems that cannot be solved in the previous traditional algorithms, namely Back Propagation Neural Network (BPNN) and Tikhonov Regularization method. There are several other intelligent algorithms such as CNN (Convolution Neural Network) and K-NN (K-Nearest Neighbor), but such algorithms have not been able to produce the expected image quality. Thus, further study is still needed for the improvement of the image quality. The expected result in this study is the comparison of these two techniques, namely ART and GAN to get the best results on the image reconstruction using MIT. Thus, it is shown that GAN is a better candidate for this purpose.

View
Thumbnail Image
Publication

System identification of flexible structure using firefly algorithm

2021-05-03 , Mohd Sazli Saad , Noor Fadhilah Mat Ros , Mohamad Ezral Baharudin , Mohd Zakimi Zakaria , Azuwir Mohd Nor

This paper originally concentrates on application of firefly algorithm on modeling of flexible beam. An attempt of obtaining a linear parametric model is accomplished by acquiring the input-output data from the vibration of beam and followed by the selection the auto regressive with exogenous inputs (ARX) model structure as a benchmark for mathematical model of beam. The parameters of the model structure are identified using firefly algorithm. A few sets of parameter settings are tested to find the most optimal value. The best parameter setting estimation is selected based on performance of model. The model is compared with the model from conventional estimation approach and validated using validity test. It is found that firefly algorithm produces the best ARX model with the lowest MSE compared to the least square algorithm.

View
Thumbnail Image
Publication

Synthesizing and Optimization the Hydroxyapatite Based on Corbiculacea Seashells

2021-01-01 , Mohd Riza Mohd Roslan , Nashrul Fazli Mohd Nasir , Nur Farahiyah Mohammad , Cheng Ee Meng , Nasrul Amri Mohd Amin , Abdul Khalid M.F. , Mohd Zakimi Zakaria , Muhammad Mokhzaini Azizan , Muzammil Jusoh

Hydroxyapatite (HA) is one of the main components in bone which functions to enhance its cell regeneration. Synthetically produced HA, based on seashell resources has higher biocompatibility, and in high demand especially in bone tissue engineering. However, the secondary phase of HA production are calcium oxide and carbonate, which impedes its performance. HA is synthesized via wet chemical precipitation and optimization were done to obtain nearly pure HA by manipulating the pH value and sintering temperature. It is expected that the combination of these parameters will optimize the amount of secondary phase hence attained nearly stoichiometric or pure HA. HA powders were analyzed through Fourier Transform Infrared Spectroscopy (FTIR) and Energy Dispersive X-ray Spectrometry (EDX).

View
Thumbnail Image
Publication

FDM Parameters Optimization for Improving Tensile Strength using Response Surface Methodology and Particle Swarm Optimization

2024-09-01 , Mohamad Ezral Baharudin , Mohd Sazli Saad , Mohd Zakimi Zakaria , Azuwir Mohd Nor , Ab Talib M.H.

Fused deposition modelling (FDM) is a popular 3D printing technique that uses a thermoplastic filament as the build material. In FDM 3D printing, tensile strength can be an issue because the layers of the object are built on top of each other, and if the layers do not adhere properly, the object can be weak and prone to breaking. Typically, this problem is caused by incorrect parameter settings. Hence, this study was then carried out to analyse and improve the printing quality in term of tensile strength of the printed part using the response surface methodology (RSM) and the particle swarm optimization (PSO) method. The effect of four input parameters such as layer height, printing speed, infill density, and print temperature was examined on the tensile strength of polylactic acid (PLA) standard samples ASTM D638-IV. The experimental design was performed using face-centred central composite designs (FCCD). The experimental data were statistically analysed to form a regression model of the tensile strength. This model was used to approximate the actual process. The optimization was performed using desirability analysis from RSM and PSO to search for the optimal parameter for maximum tensile strength. Experimental results showed that PSO outperformed RSM with a 1.52 % reduction in tensile strength. The maximum tensile strength obtained from PSO was about 39.069 MPa with the optimal process parameters of layer height of 0.30 mm, printing speed of 30.17 m/s, infill density of 79.72 %, and print temperature of 205.92 °C.

View
Thumbnail Image
Publication

ARx modeling of flexible beam system using bat algorithm

2021-05-03 , Mohd Sazli Saad , Noor Fadhilah Mat Ros , Mohd Zakimi Zakaria , Mohamad Ezral Baharudin , Azuwir Mohd Nor

This paper describes the development of dynamic model of flexible beam system using system identification method based on nature inspired algorithm i.e. bat algorithm. At first, input-output data from the experimental rig of flexible beam were collected such that input signal is taken from piezo actuator and output signal from the laser displacement sensor. Then, linear parametric model structure is accomplished using auto regressive with exogenous inputs (ARX). The optimal parameters of the ARX model are identified using bat algorithm. The best parameter setting estimation is selected based on the best fit criterion i.e mean square error (MSE). The identified model is compared with the model from conventional estimation approach. Simulation results show that bat algorithm can outperform the least square algorithm in parametric modelling of the flexible beam.

View