Lecturer's Publications

Permanent URI for this community

https://digitallibrary.peninsulacollege.edu.my/handle/123456789/17

Browse

Now showing 1 - 20 of 24

Agent-Based Optimization for Multiple Signalized Intersections using Q-Learning
(Faculty of Engineering, Universiti Malaysia Sabah Kota Kinabalu, Malaysia, 2014-12-30) Yit Kwong Chin; Kiam Beng Yeo; Kenneth Tze Kin Teo; Helen Sin Ee Chuo; Min Keng Tan
Relieving urban traffic congestion has always been an urgent call in a dynamic traffic network. The objective of this research is to control the traffic flow within a traffic network consists of multiple signalized intersections with traffic ramp. The massive traffic network problem is dealt through Q-learning actuated traffic signalization (QLTS), where the traffic phases will be monitored as immediate actions can be taken during congestion to minimize the number of vehicles in queue. QLTS is tested under two cases and has better performance than common fixed-time traffic signalization (FTS). When dealing with the ramp flow, QLTS has flexibility to change the traffic signals according to the traffic conditions and necessity.
CUSUM-Variance Ratio Based Markov Chain Monte Carlo Algorithm in Overlapped Vehicle Tracking
(Modelling, Simulation and Computing Laboratory School of Engineering and Information Technology Universiti Malaysia Sabah Kota Kinabalu, Malaysia, 2011) W.Y. Kow; W.L. Khong; Y.K. Chin; I. Saad
Markov Chain Monte Carlo (MCMC) is one of the algorithms that have been widely implemented in tracking vehicle for traffic surveillance purposes. The sampling efficiency of the algorithm is essential to determine the vehicle position accurately. However, the sample size of the algorithm is still remaining an issue as non-optimal sample size will defect the tracking accuracy, especially when the moving vehicle is overlapped. Adaptive sample size of MCMC has been implemented using CUSUM Path Plot and Variance Ratio algorithms to perform vehicle tracking. CUSUM Path Plot determines the samples convergence rate by calculating the hairiness of the sample size whereas Variance Ratio method computes two sets of MCMC to determine the samples steady state. This paper proposes the fusion of CUSUM-Variance ratio algorithm to enhance the tracking efficiency. Experimental results shows that the CUSUM-Variance Ratio method have a better performance in tracking the overlapping vehicle with higher accuracy and more optimal sample size compared to the standalone CUSUM Path Plot and Variance Ratio approaches.
Enhancement of Markov Chain Monte Carlo Convergence Speed in Vehicle Tracking Using Genetic Operator
(Modeling, Simulation & Computing Laboratory, Minerals & Materials Research Unit School of Engineering and Information Technology Universiti Malaysia Sabah Kota Kinabalu, Malaysia, 2012) Wei Yeang Kow; Wei Leong Khong; Yit Kwong Chin; Ismail Saad; Kenneth Tze Kin Teo
Markov Chain Monte Carlo (MCMC) has been essential in tracking vehicle undergoing disturbances for traffic surveillance purposes. It is capable of tracking vehicle by estimating the vehicle’s position with the sampling of probability distributions. However the accuracy of the position estimation is highly dependent on the sampling efficiency of MCMC. Therefore the sample size of the MCMC is adapted to track the vehicle according to the disturbances encountered. The adaptive sample size of MCMC is determined by using the CUSUM path plot and variance ratio convergence diagnostic algorithm. To further enhance the convergence speed, genetic crossover and mutation operator is introduced into the adaptive MCMC. The genetic operator (GO) is capable of reduces the variance between samples and hence allowing faster convergence speed on the MCMC samples. Experimental results have shown that the GO adaptive MCMC tracking algorithm have better tracking performances with consumption of lesser sample size.
Enhancement of Particle Filter Resampling in Vehicle Tracking via Genetic Algorithm
(UKSim-AMSS 6th European Modelling Symposium, 2012) Wei Leong Khong; Wei Yeang Kow; Yit Kwong Chin; Mei Yeen Choong; Kenneth Tze Kin Teo
Vehicle tracking is an essential approach that can help to improve the traffic surveillance or assist the road traffic control. Recently, the development of video surveillance infrastructure has incited the researchers to focus on the vehicle tracking by using video sensors. However, the amount of the on-road vehicle has been increased dramatically and hence the congestion of the traffic has made the occlusion scene become a challenge task for video sensor based tracking. Conventional particle filter will encounter tracking error during and after occlusion. Besides that, it also required more iteration to continuously track the vehicle after occlusion. Thus, particle filter with genetic operator resampling has been proposed as the tracking algorithm to faster converge and keep track on the target vehicle under various occlusion incidents. The experimental results show that enhancement of the particle filter with genetic algorithm manage to reduce the particle sample size.
Evolutionary Optimization Scheme for Exothermic Process Control System
(2011) Tan, M.K.; Chin, Y.K.; Tham, H.J; Teo, K.T.K.
The primary aim in batch process is to enhance the process operation in order to achieve high quality and purity product while minimizing the production of undesired byproduct. During the process, a large amount of heat is released rapidly when the reactants are mixed together due to exothermic behavior. This causes the reaction to become unstable and consequently the quality and purity of the final product will be affected. Therefore, it is important to have a control scheme which is able to balance the needs of process safety with the product quality and purity. This paper proposes genetic algorithm (GA) as an approach to control the process temperature by changing the coolant temperature because GA does not require the exact process dynamics in advance, which normally in practical, the process dynamics are poorly known in practical. GA is able to evolve itself to obtain an optimum solution to change the coolant temperature. The simulation studies show that the GA will be a good candidate to optimize the process and minimize the temperature overshoot throughout the reaction process. Furthermore, GA is able to evolve itself to obtain an optimum solution to change the coolant temperature.
Exploring Q-Learning Optimization in Traffic Signal Timing Plan Management
(2011) Yit, Kwong Chin; Lai, Kuan Lee; Nurmin Bolong; Soo, Siang Yang; Tze, Kin Teo (Kenneth)
Traffic congestions often occur within the entire traffic network of the urban areas due to the increasing of traffic demands by the outnumbered vehicles on road. The problem may be solved by a good traffic signal timing plan, but unfortunately most of the timing plans available currently are not fully optimized based on the on spot traffic conditions. The incapability of the traffic intersections to learn from their past experiences has cost them the lack of ability to adapt into the dynamic changes of the traffic flow. The proposed Qlearning approach can manage the traffic signal timing plan more effectively via optimization of the traffic flows. Qlearning gains rewards from its past experiences including its future actions to learn from its experience and determine the best possible actions. The proposed learning algorithm shows a good valuable performance that able to improve the traffic signal timing plan for the dynamic traffic flows within a traffic network.
FAULT DETECTION AND LOCALIZATION METHOD FOR INVERTER OPEN-CIRCUIT CONDITION IN THREE-PHASE INDUCTION MOTOR DRIVES
(2019) AMIR RASYADAN BIN KELANA
Recent advancement in semiconductor technologies have extended the use of induction motors in a wide area of variable speed applications. Traditional fixed speed drives are becoming obsolete and majority of the applications now require power electronic inverter-based drives for a more efficient and precise speed control. Nonetheless, the use of power electronic came with an increased possibility of the drive system failure, mainly caused by the switching device itself. There are two common types of inverter faults which are short-circuit and open-circuit fault. Short-circuit fault detection methods are mostly hardware based while open-circuit faults require proper algorithm to be detected. The aim of this research work is to provide a simulation analysis on the inverter fault detection algorithm to detect open-circuit fault together with fault localization algorithm to locate the open-circuit switch in a three-phase induction motor drive. Computer simulation model was built using MATLAB Simulink to simulate the operation of three-phase induction motor drive under healthy and faulty inverter conditions. The sub models of the drive system include an induction motor direct-quadrature (DQ) model, a voltage source inverter (VSI) driven by an open-loop SPWM and a graphical user interface (GUI) to allow user control on the input parameters during simulation run. Dynamic behavior of the motor speed, phase current and electromagnetic torque under healthy and the combinations of single and double open-circuit fault conditions were simulated. The simulated phase currents were then pre-processed to filter the signal from multiple harmonic content induced by the PWM-switching process before being used for features extraction. This work focuses on current vector trajectory based fault detection and localization (FDL) method, two separate algorithms are used to detect the faulty inverter leg and to localize the faulty switch within the faulty leg. Fault detection was done using the modified slope calculation method of the phase currents in the complex DQ plane, while the xx localization strategy is based on the observation of the phase currents polarity. 21 combinations of single and double open-circuit faults were simulated. The results have shown that the FDL method is capable to detect and localize all the 21 fault conditions. Since the fault pattern may appear at different current angle, the detection and localization speed depend on the fault occurrence angle which varies within one phase current period. The algorithm also proved to work under variable load and frequency conditions. Overall, this work has shown the working principle and the capability of the current vector trajectory based FDL algorithm to detect and to localize the 21 open-circuit fault conditions. The simulation analysis done in this work would be useful for future development of the FDL algorithm.
Genetic Algorithm Based PID Optimization in Batch Process Control
(2011) M.K. Tan; Y.K. Chin; H.J. Tham; K.T.K. Teo
The primary aim in batch process is to enhance the process operation in order to achieve high quality and purity product while minimising the production of undesired by-product. However, due to the difficulties to perform online measurement, batch process supervision is based on the direct measurable quantities, such as temperature. During the process, a large amount of exothermic heat is released when the reactants are mixed together. The exothermic behaviour causes the reaction to become unstable and consequently the quality and purity of the final product will be affected. Therefore, it is important to have a control scheme which is able to balance the needs of process safety with the product quality and purity. Since the chemical industries are still applying PI and PID to control the batch process, researchers are keen to optimize PID parameters using artificial intelligence (AI) techniques. However, most of these PID optimization techniques need online process model to predetermine the optimizer parameters. However in practice, the dynamic model of the batch process is poorly known. As a result, majority of the studies focused on acceptable performance instead of optimum performance of the batch process control. This paper proposes a new genetic algorithm (GA) optimizer which consists of additional information of the online estimated model parameters in addition to the PID parameters as the string of the GA. The simulation results show that the proposed GA auto-tuning method is a better candidate than the regular GA where the estimated model parameters in fitness function is capable to control the process temperature while avoiding model mismatch and disturbance condition.
Image Segmentation via Normalised Cuts and Clustering Algorithm
(2012) Mei, Yeen Choong; Wei, Yeang Kow; Yit, Kwong Chin; Lorita Angeline; Tze, Kin Teo (Kenneth)
Image segmentation has been widely applied in image analysis for various areas such as biomedical imaging, intelligent transportation systems and satellite imaging. The main goal of image segmentation is to simplify an image into segments that have a strong correlation with objects in the real world. Homogeneous regions of an image are regions containing common characteristics and are grouped as single segment. One of the graph partitioning methods in image segmentation, normalised cuts, has been recognised producing reliable segmentation result. To date, normalised cuts in image segmentation of various sized images is still lacking of analysis of its performance. In this paper, segmentation on synthetic images and natural images are covered to study the performance and effect of different image complexity towards segmentation process. This study gives some research findings for effective image segmentation using graph partitioning method with computation cost reduced. Because of its cost expensive and it becomes unfavourable in performing image segmentation on high resolution image especially in online image retrieval systems. Thus, a graph-based image segmentation method done in multistage approach is introduced here.
Implementing Manifold Learning in Adaptive MCMC for Tracking Vehicle under Disturbances
(2012) Wei, Yeang Kow; Yit, Kwong Chin; Wei, Leong Khong; Hui, Keng Lau; Tze, Kin Teo (Kenneth)
In recent years, tracking vehicle with overlapping and maneuvering disturbances has become a challenging task in visual tracking. Markov Chain Monte Carlo (MCMC) is proved to be effective in tracking vehicle under disturbances by probabilistically estimating the vehicle position. However the sampling based tracking algorithm is highly depending on the sampling efficiencies where adequate chain length is necessary to sustain the tracking accuracy. Therefore variance ratio (VR) based MCMC has been implemented in this study to adapt the chain length according to the disturbances encountered. Isomap manifold learning is further implemented to update the vehicle model and accurately track the vehicle with maneuvering disturbances. Multiple vehicle models with different viewing angles are represented by Isomap under low dimensional manifold. The suitable vehicle model will be selected according to the estimated vehicle position. Experimental results have shown that Isomap-VR-MCMC have better tracking performances compared to VR-MCMC with smaller RMSE value.
International Journal of Simulation Systems, Science & Technology
(United Kingdom Simulation Society, 2011-06) Guest Editor: Ismail Saad
Mathematical Model Implementation of SPWM fed Three-phase Induction Motor Drive Using MATLAB Simulink
(Intelligent Automotive Systems Research Cluster, Electrical Electronic and Automation Section, University Kuala Lumpur Malaysian Spanish Institute, Kulim, 09000, Kedah, Malaysia*, 2020) Amir Rasyadan; Sazali bin Yaacob; Pranesh Krishnan
Three-phase induction motors are used in a vast area of applications mainly due to their simplicity, ruggedness and high reliability. With recent advancement in semiconductor technologies, the use of fixed speed induction motor drive is becoming obsolete, majority of the applications now requires inverter-based drives for variable speed operation. In the study of induction motor drive operation, mathematical models are often used to simulate the steady state and transient behavior of induction motor. However, to develop such model is not a straightforward task. Knowing only the equations by themselves are not always enough without some knowledge on solving mathematical equations with the use of computer simulation software. This work presents an approach to implement the mathematical model of a Sinusoidal Pulse Width Modulation (SPWM) fed three-phase induction motor drive in MATLAB Simulink. The sub models include induction motor DQ model and a voltage source inverter (VSI) fed by SPWM signal generator. The presented model implementation is able to simulate the dynamic behavior of an induction motor operation, this would be useful for further studies on the development of induction motor drive system.
Minimizing Network Coding Nodes in Multicast Tree Construction via Genetic Algorithm
(2012) Shee Eng Tan; Zhan Wei Siew; Yit Kwong Chin; Scott Carr Ken Lye; Kenneth Tze Kin Teo
Network coding is a method that increases network throughput by encoding several packets with single packet size length and forwards the packet in a single transmission time slot. At the same time, network coding increases the complexity of packets management and delay of network due to the waiting time for network coding opportunity. A solution based on improved genetic algorithm is proposed to optimize the network coding node resources in network coding. Genetic algorithm will search a multicast tree that fulfils the desired throughput with a desired multicast rate. Mutation rate of the genetic algorithm will change based on the previous solution to avoid from being stuck on the local optima. The simulation result shows that with given multicast rate, improved genetic algorithm is able to search and construct multicast tree with minimal usage of network coding nodes.
Multiple Intersections Traffic Signal Timing Optimization with Genetic Algorithm
(2011) Y.K. Chin; K.C. Yong; N. Bolong; S.S. Yang; K.T.K. Teo
Traffic congestion in the urban area occurs more frequent than the past due to rapidly increasing on road vehicle usage rates. It could seriously hinder the development of urban area if a well management system has not being established. These scenarios necessitate the development of advance traffic management systems to increase the performance of signalized intersection. Traffic signal timing management (TSTM) system which comprise of genetic algorithm based optimization is proposed. Using a proper TSTM system, network traffic flow can be improved with considerably less cost than other infrastructural improvements. The proposed genetic algorithm based optimization approach allows signal timing parameters such as offset, cycle time, green split and phase sequence to be optimized with objective of minimum delay and better traffic fluency. The proposed GATSTM system has the ability to handle and manage the dynamic changes of the traffic networks condition by calibrating the system parameters accordingly.
OPTIMIZATION OF EMBEDDED PLATFORM IN COLLABORATIVE CAMERA APPLICATION FOR DISTRIBUTED FACE DETECTION
(2018-05) MOHD KHAIRI BIN KAMARUDDIN
This thesis presented an optimization technique for embedded platform in a collaborative camera for distributed face tracking system. In this work various method,hardware and its configuration for face detection and recognition has been reviewed and implemented. Local Binary Pattern is used as feature extraction method for face detection algorithm. Then Fisherface (Linear Discriminant Analysis) has been implemented for face recognition algorithm. In terms of embedded platform, Raspberry Pi2 has been chosen as a preferred platform to run the system because it is compatible with OpenMP multiprocessing method. The advantage of Raspberry Pi 2 is it has the most number of core and the lowest cost when compare to the other embedded platform which was tested. Several enhancements have been done to the embedded platform which enable multiprocessing for OpenCV libraries using OpenMP and Random Access Memory tuning whereby mounting the log files of the operating system to the cache memory. Every node contains algorithms using Local Binary Pattern as feature extraction method for face detection and Fisherface as recognition technique. Alternate loop has been applied to the system avoiding vision algorithm for every two out of three times. By doing this, the processing time is reduced without compromising the performance of the nodes. This enhancement also further gives a big impact on reducing face detection speed rate. The result shows the optimization of the embedded platform allow the program to increase the number of frame per second from 3 frame per second to 13 frame per second without compromising the number of detection per second which is from 4 to 3 detections. Futher work on the algorithm shows its capability of the collaborative camera to share and centralize the logging information between cameras regarding the moving of samples and gives a warning if suspicious unregistered sample intrude to surveillance area.
Optimization of Traffic Flow Within an Urban Traffic Light Intersection with Genetic Algorithm
(2010) K. T. K. Teo; W. Y. Kow; Y. K. Chin
Traffic flow control optimization in the traffic light systems is studied for improvement in this paper. Traffic light systems are built to control the traffic flows at the intersections to ensure the fluency of traffic flow within the traffic network. The increasing traffic flows that cannot be supported by the current traffic light systems cause lengthen of queue length at the intersection. The effect of queue length, green time, cycle time and amber time in the traffic system is observed and studied through simulations. Longer green time will pass through more vehicles, but it will increase the cycle time at the same time which causes more vehicles to accumulate at the intersection during the waiting time. Genetic algorithm is introduced in this paper for the optimization of the traffic flow control as its ability to find the optimized solution in its self tuning process. Genetic algorithm taking current queue length as its input then it will output the optimized green time for the intersection. The result of Genetic algorithm is further improved with the introduced of the incoming traffic flow during red time of each phase.
OPTIMIZATION OF URBAN MULTI-INTERSECTION TRAFFIC FLOW VIA Q-LEARNING
(2013) Yit, Kwong Chin; Heng, Jin Tham; N.S.V. Kameswara Rao; Nurmin Bolong; Tze Kin Teo (Kenneth)
Congestions of the traffic flow within the urban traffic network have been a challenging task for all the urban developers. Many approaches have been introduced into the current system to solve the traffic congestion problems. Reconfiguration of the traffic signal timing plan has been carried out through implementation of different techniques. However, dynamic characteristics of the traffic flow increase the difficulties towards the ultimate solutions. Thus, traffic congestions still remain as unsolvable problems to the current traffic control system. In this study, artificial intelligence method has been introduced in the traffic light system to alter the traffic signal timing plan to optimize the traffic flows. Q-learning algorithm in this study has enhanced the traffic light system with learning ability. The learning mechanism of Q-learning enables traffic light intersections to release itself from traffic congestions situation. Adjacent traffic light intersections will work independently and yet cooperate with each others to a common goal of ensuring the fluency of the traffic flows within the traffic network. The simulated results show that the Q Learning algorithm is able to learn from the dynamic traffic flow and optimize the traffic flow accordingly.
Overlapping Vehicle Tracking via Adaptive Particle Filter with Multiple Cues
(2011) W.L. Khong; W.Y. Kow; Y.K. Chin; I. Saad; K.T.K. Teo
Vehicle tracking is a vital approach to assist the onroad traffic surveillance system. Since the on-road vehicles is increasing, occlusion and overlapping of vehicles is often happen in the traffic surveillance scene. Therefore, segmentation and tracking of the occlusion or overlapped vehicle can be a challenging task in surveillance system via image processing. In this paper, a multiple cues overlapping vehicle tracking algorithm is proposed to continuously track the occluded vehicle effectively. The earlier vehicle tracking systems are normally based on colour feature which will leads to inaccurate results when the background colour is complex or too similar with the target vehicle. On the other hand, shape feature will increase the accuracy but consume more computation time in the resampling process during overlapping. The experimental results show that enhancement of the particle filter resampling process with multiple cues is capable to track the overlapped vehicle with higher accuracy and without compromising the processing time.
Performance Analysis of Intelligent Transport Systems (ITS) with Adaptive Transmission Scheme
(2012) Scott Carr Ken Lye; Shee Eng Tan; Yit Kwong Chin; Bih Lii Chua; Kenneth Tze Kin Teo
Various means of modern transports have already taken the initiative to incorporate computing and communication technology. This cross field area is coined as intelligent transportation systems (ITS). Transport systems of the future require fast and precise monitoring to ensure safety of such systems is guaranteed. Thus, the communication aspect demands seamless and minimal error whilst delivering vital data. However, the dynamic surroundings always post a challenge to wireless communications, taking account into a myriad of interference such as multipath fading, shadowing, dispersing and path loss. Furthermore, there is also case of mobility. In this paper, the performance of wireless communications with adaptive modulation and coding in vehicular scenarios were compared against rigid transmission techniques. Simulations were conducted over different mobility channel models against Signal-to-Noise ratio (SNR) to provide thorough analysis. Performance measure of Bit Error Rate (BER) and Packet Error Rate (PER) were used to provide understanding of the overall picture.
Q-Learning Based Traffic Optimization in Management of Signal Timing Plan
(2011) Yit Kwong Chin; Nurmin Bolong; Aroland Kiring; Soo Siang Yang; Kenneth Tze Kin Teo
Occurrences of traffic congestions within the urban traffic network are increasing in a rapid rate due to the rising traffic demands of the outnumbered vehicles on road. The effectiveness of management from traffic signal timing planner is the key solution to solve the traffic congestions, but unfortunately the current traffic light signal system is not fully optimized based on the dynamic traffic conditions on the road. Adaptable traffic signal timing plan system with ability to learn from their past experiences is needed to overcome the dynamic changes of the urban traffic network. The ability of Q-learning to prospect gains from future actions and obtain rewards from its past experiences allows Q-learning to improve its decisions for the best possible actions. A good valuable performance has been shown by the proposed learning algorithm that able to improve the traffic signal timing plan for the dynamic traffic flows within a traffic network.

Browse

Browsing Lecturer's Publications by Title

Results Per Page

Sort Options