Invited Speakers
Automated Essay Scoring (AES) using English Essay Question in Malaysia's School
Assoc. Prof. Ir. Dr. Nasharuddin Zainal
Abstract
Automated Essay Scoring (AES) is a one of the important research areas in educational technology. Research about AES is started in the early 1960s and keep growing in development along with the advances of the computing technology. AES is more widely use by the educational system for assessment and classroom activities, it being use in all layers of education. The potential for Automated Essay Scoring is being used widely becomes a reality as more and more classroom activities, assessment, and test preparation materials are supplied online. These essay questions might include everything from arithmetic, where students are asked to explain how they arrived at their answers, to science, where they might be asked to define words or explain experiments, or history, where they must ask to explain or discuss an event. In this study we try to design the AES using Natural Language Processing (NLP) and machine learning techniques to evaluate the student's answer by comparing it with the answer scheme. It is not simply a string-matching program and need a proper system to process the students answer. We propose the 4 main process to the system that is Spelling Checking, Pre-Processing, Latent Semantic Analysis and Thesaurus Checking. All this process is combined and analyze to become a good AES system. Each of this process mean give somethings important to the system flow on how the essay going to be evaluate. The system has been done some experiment and the system get the average correlation between system scores and human appraiser scores is only 70%. However, there needs to be more research on the future use of this Automatic Essay Scoring system.
Battery Lifespan Prognostics for Sustainable Operation and Management of Battery Storage Systems
Associate Professor Dr. Afida Ayob
Abstract
The increasing energy demands of a growing population and the challenges of climate change provide a strong driving force for transportation electrification and smart grid development. As one of the most widely used energy storage devices, lithium-ion batteries play an important role in those fields. A long battery lifetime is critical to achieving the economic viability in electric vehicles, renewable energy, and smart grid infrastructure. One of the most urgent issues in lithium-ion batteries is degradation. Battery degradation is a complex electrochemical process, which includes many electrochemical side reactions, such as solid electrolyte interphase, electrolyte oxidation, salt decomposition, particle fracture, and active material dissolution. Accurate predictions of the remaining battery lifetime at different operating conditions are essential for the battery management system to avoid potentially dangerous battery failures and guarantee reliable and efficient operation. The development of battery remaining useful lifetime (RUL) prognostic techniques, challenges and future research directions are identified and discussed.
Nutrient classification using machine-learning algorithms on optical absorbance data for hydroponics system
Ts. Dr. Nur Hidayah Azeman
Abstract
Hydroponics is a soilless cultivation system that has received significant attention as an approach to attain food demands and overcome soil constraints. Nitrogen, phosphorus, and potassium (NPK) are essential macronutrients that plays a vital role in plant growth. Hydroponic cultivation faces challenging issues such as high initial and operational costs compared to soil-based cultivation and imbalanced ions in the nutrient solution. Determining nitrogen in nitrate is crucial, as either an insufficient or excessive supply of nitrate ions may reduce the plant yield or lead to environmental pollution. This work discusses the advancement of hydroponics systems toward precision farming, where the performance of feature reduction techniques and conventional machine learning (ML) algorithms in determining nitrate concentration levels were demonstrated. The reduction technique and ML algorithms were evaluated using a high-dimensional spectroscopic dataset containing measured nitrate-nitrite mixed solution absorbance data.
High Flow Rate Peristaltic Electromagnetic Micropump Fabricated Using Multiple Photolithography
Assoc. Prof. Dr.Jumril Yunas
Abstract
This paper reports a channel fabrication method using multiple photolithography for peristaltic electromagnetic micropump. The process is aimed to create a PDMS based microchannel and pump chamber with sufficient cross-sectional width and height to enable the high-transport rate of fluidic sample. The process is also intended to replace the SU-8 photoresist weakness that cannot be removed with PDMS compatible chemicals. This technique includes 4 main steps. The initial process uses the SU-8 photoresist to create the pathway and followed with the second photolithography using the AZ 4620 photoresist to fill up the channel. The next process is to look at the PDMS to close the photoresist and form a chamber. The last process is to remove the photoresist in the channel using acetone, so that the channel will be formed. The coating properties and surface quality were analyzed using scanning electron microscopy (SEM). The results showed that the fabricated channel and chamber can be formed properly without any leaking that can be used for microfluidic delivery system that ensures an easier, simpler and more advantageous process compared to conventional lithography process.
Smoothing of Hip Angle Kinematics Data During Parachute Landing Using Functional Data Analysis Approach
Dr. Syazwana Aziz
Abstract
Smoothing is one of the fundamental procedures in functional data analysis (FDA). There are three main factors in the data smoothing process through the basic method of B-spline expansion namely order, roughness penalty and smoothing parameter. Therefore the main objective of this research are to identify the optimal value of derivatives, roughness penalty and smoothing parameter by applying the kinematics data, specifically hip flexion/extension angle during the parachute landing activity. There have several series of tests conducted to obtain appropriate parameters to be applied to the overall kinematics and kinetics data. According to result shows that the optimal smoothing parameter was smoothing B-spline basis order 6, 4th order of roughness penalty with lambda λ=1e-12.
Vital signs monitoring, systems, and technologies towards application to Internet of medical things
Dr. Rosmina Jaafar
Abstract
Vital signs are the measurements of human body's most basic functions. These measurements are taken to help assess the general physical health of a person, give clues to possible diseases, and show progress toward recovery from specific illnesses. There are four to six main vital signs commonly acquired. The first four vital signs often referred as the primary vital signs that include body temperature, blood pressure, pulse (heart rate), and breathing rate. The secondary vital sign is extended to the fifth vital sign, which is commonly obtained by medical doctors in the hospital as the pain score, and the sixth vital sign, which is often discipline-dependent parameter such as oxygen saturation that is a vital parameter observed in the cases of COVID-19. Self-monitoring of the vital signs is helping people to better monitor their health status both at an activity/fitness level for self-health tracking and at a medical level providing more data to clinicians with a potential for early diagnostic and can provide treatment guidance. Current technology revolution such as during the era IR4.0 has caused advancement of many medical devices. This provides the possibility of vital signs to be handy and be acquired via the miniaturization of electronic devices that enabled the design of many reliable and adaptable devices, which contribute to a world-wide change in the health monitoring approach. This talk will present a review of important aspects of both primary and secondary vital signs, the vital signs sensing technologies plus their system architectures and specifications. For each vital sign the description will be made concerning its origin and effect on health, monitoring needs, acquisition methods and the recent scientific developments on the related area. For the vital signs acquisition technologies, the focus of description will be emphasized on the main system architectures. Furthermore, discussion will be made on the evolution of medical electronic devices based on the basic prototypes developed over the years. Finally, discussion will focus on the likely market trends and
future challenges for the emerging technologies incorporating the internet of medical things.
Optimized piezoelectric energy harvesting system for Micro-devices application integrated with Internet of Things (IoT) platform toward sustainable energy utilization
Dr. Mahidur R Sarker
Abstract
In the past few years, due to its significant deployment to meet global demand, the Internet of Things (IoT) has gained a lot of interest. Power-efficient devices such as sensors and IoT have been shown to play an
important role in our daily lives. However, due to the lower lifetime of different energy resources needed to power the sensors over time, the operating time of IoT devices are restricted. Piezoelectric Energy Harvesting system (PEHS) has been demonstrated to show the potential to be used to supply the low
energy required to power these types of devices. However, multiple factor contributes towards the optimal operation of PEHS such transducer elements selection, modules interaction, generated power control and optimization. The general aim of this research is to investigate and elucidate the interaction of the main components and elements of PEHS. The elucidation of the components interaction can be used to produce digital twins of the system which later can be simulated and investigated further. Subsequently, the digital twin of the PEHS can then be utilized in order to create control and optimization strategy that can be used to ensure that sufficient energy can be provided by the PEHS such that it can power IOT devices accordingly. We proposed the use of tuned Propotional-Integral-Derivative (PID) controller to help optimize the PEHS system output. Several tuning approach, including the heuristic tuning, Ziegler-Nichols (ZN) tuning, and iterative learning algorithm (ILA) will be used to obtain the optimum value of the PID controller parameters. The final power generated from the optimally controlled PEHS is expected to be able to power the IOT devices sufficiently to sustain its independent operation.
Wireless Airborne IoT Network for Rural Water Quality Monitoring
Assoc. Prof. Dr. Nor Fadzilah Abdullah
Abstract
Currently, the Chini Lake shores house around 500 indigenous people distributed across six villages with limited access to cellular tower communication coverage. This is mainly due to the challenging terrain profile and dense foliage. Therefore, it is important to establish a reliable line-of-sight (LoS) data transmission via a low-altitude platform (LAP). The high availability but low-cost wireless communication infrastructure such as LoRa is the perfect solution in this scenario. This will allow better coverage due to good propagation characteristics at lower frequency bands as well as the elevated platform. The solution shall also equip a wireless machine-to-machine (M2M) network, sensors technologies and a big data analytic enablement platform. In addition, the characterization of the wireless channel behaviour in Malaysia’s tropical rural areas, where the propagated wireless signal suffers from several imperfections, such as attenuation, diffraction, scattering and absorption due to the presence of various surrounding elements is also being investigated. The outcome of this research is expected to offer a new understanding of the propagation behaviour of current and future wireless IoT technologies, thus helping the network engineer to perform accurate planning and deployment in a rural environment. With this solution, the indigenous Orang Asli community in Chini Lake, Pahang, Malaysia will have access to digital content, as well as water level alerts for mitigation of flooding and drought situations, and Internet access for the promotion of local products and services.
The fabrication of 4H-Silicon Carbide (4H-SiC) FinFET:
Challenges and prospects for sustainable power devices
Dr. Muhammad Idzdihar Idris
Abstract
Silicon-based devices are excellent in both material and electrical properties. They have been used in many applications in our daily life, including smartphones, computers and other electronic devices. However,
silicon-based devices could not function in specific environments, especially in extreme and hostile environments. As an alternative, semiconductors with a wider band gap, such as silicon carbide (SiC) and gallium nitrides (GaN) are good candidates for those environments. The ability of SiC devices to operate
in extreme environments such as high temperature, high voltage and high radiation has resulted in significant interest at both research and commercial levels. However, one of the essential parameters, which is the field effect mobility (µFE) of the 4H-SiC metal oxide semiconductor field effect transistors, is significantly lower than that observed in bulk due to the high density of traps between the dielectric and the underlying semiconductor. Various advanced processing techniques have been proposed to mitigate the interfacial problem at SiC/SiO2. In this work, a new approach to utilizing the FinFET structure was proposed to increase the total drain current due to the increase of the active area. This work also discusses the challenges in the fabrication process of 4H-SiC FinFET and the effect of different trench widths on the normalized drain current for single and multiple-Fin. It was found that 4H-SiC MOSFETs with a FinFET structure utilizing aluminium dioxide, Al2O3, as a dielectric demonstrated an enhanced peak µFE of 140 cm2 V−1s−1 and smaller subthreshold slope, S, in comparison to planar MOSFETs with same gate dielectric process. The anisotropic channel mobility of 4H-SiC FinET fabricated on different sidewall plane orientations from (11) ̅20/ 112 ̅0 to 11 ̅00/1 ̅100 is also reported. It is thereby shown that the FinFET structure of 4H-SiC significantly impacts the electrical performance by attaining higher electron mobility than the conventional MOSFET.
Model Reduction of Discrete-Time Bilinear Second Order-Structured Systems in Infinite and Finite Frequency Intervals
Assoc. Prof. Ir. Dr. Rosmiwati Mohd Mokhtar
Abstract
Model reduction technique approximates the higher-order model by a lower-order model in such a manner that the input-output behaviour of the original system is preserved at a desired level of accuracy. In this study, the balancing-type model order reduction of discrete-time bilinear second order-structured systems are considered. These systems are independently linear in the state and the input, but not jointly. The formulation involves derivative pairs of each state and reduction techniques for such systems require retention of state pairs in a reduced order model to make reduction effort meaningful. Retention of the state's structure has been achieved by partitioning the system’s Gramians into position and velocity portions. Balancing Gramians with different combinations yields different second-order balanced truncation techniques. The proposed approach is evaluated on a bilinear system model for infinite and finite
interval applications. Results certified the proposed technique can perform model reduction and at the same time preserve its structure, especially in limited frequency intervals.
Organized by
Supported by
