KYKok Hwa Yu

Kok Hwa Yu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Kok Hwa Yu has a total of 80 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Machine learning-based calibration of IoT turbidity sensors: enhancing sensing accuracy and implementation

    by Ying Nie, Kok Hwa Yu, Choe Peng Leo, Kin Sam Yen, Huan-Bang Li, Takeshi Matsumura

    As one of the important parameter in assessing water quality, measuring turbidity using conventional nephelometers remain costly and impractical for large-scale, real-time applications. IoT-based turbidity sensors offer a scalable alternative, though their performance is often limited by low accuracy, reliability issue, and unit-to-unit variability. To address these issues, this study proposes a machine learning–driven calibration framework for DFRobot SEN0189 sensors. Controlled experiments were carried out using dairy suspensions, with turbidity meter measurements serving as reference values. Four regression models, i.e., linear regression, random forest, support vector machine, and Gaussian process regression, were tested. Among these, Gaussian process regression model demonstrated the best performance, significantly surpassing the other models in predictive accuracy. Field validation in a water reuse system further ascertained the reliability of the calibrated sensors, indicated by the closely aligned predicted turbidity value with the reference turbidity measurements before and after filtration. Overall, the results highlight the potential of machine learning calibration to elevate low-cost IoT turbidity sensors into dependable tools for continuous, real-time water quality monitoring.


  • Strengthening heat transfer of micro heat pipes using biomimetic liquid diode structures

    by Jiaqian Li, Yanbo Sun, Jingtong Wang, Yan Chen, Bo Liu, Lin Guo, Yuying Wang, Hongxing Zhang, Kok Hwa Yu, Gongming Xin


  • JoVE Video Dataset

    by Yang Wang, Kok Hwa Yu, Jing Tao Jia, Ying Nie

    This research focuses on important technologies used in wheeled mobile robot autonomous navigation, such as path planning optimization, system integration, and advancements in visual simultaneous localization and mapping (SLAM) techniques. An enhanced approach is suggested to overcome localization issues in traditional visual odometry brought on by duplicated or unevenly distributed feature points. This approach combines Efficient Perspective-n-Point (EPNP) feature matching, iterative closest point (ICP) pose optimization, and quadtree-based feature management. According to experimental findings, the suggested method greatly increases localization accuracy and stability. A dense point cloud reconstruction technique based on RGB-D data is developed to improve the completeness and detail of environmental representation while mitigating the sparsity often seen in point cloud maps produced by conventional SLAM systems. In order to enhance path quality and computational efficiency, an enhanced rapidly-exploring random tree (RRT) method is presented, which incorporates adaptive step-size management, goal biasing, and B-spline-based path smoothing. Furthermore, real-time local obstacle avoidance in dynamic situations is made possible by the integration of the Timed Elastic Band (TEB) algorithm. Comprehensive real-world tests have confirmed the usefulness of the suggested solutions in terms of efficiency, robustness, and practical applicability after they were implemented on an experimental platform based on the Robot Operating System (ROS).

  • Enhanced Visual SLAM and Path Planning for Autonomous Navigation of Wheeled Mobile Robots

    by Yang Wang, Kok Hwa Yu, Jing Tao Jia, Ying Nie

    This research focuses on important technologies used in wheeled mobile robot autonomous navigation, such as path planning optimization, system integration, and advancements in visual simultaneous localization and mapping (SLAM) techniques. An enhanced approach is suggested to overcome localization issues in traditional visual odometry brought on by duplicated or unevenly distributed feature points. This approach combines Efficient Perspective-n-Point (EPNP) feature matching, iterative closest point (ICP) pose optimization, and quadtree-based feature management. According to experimental findings, the suggested method greatly increases localization accuracy and stability. A dense point cloud reconstruction technique based on RGB-D data is developed to improve the completeness and detail of environmental representation while mitigating the sparsity often seen in point cloud maps produced by conventional SLAM systems. In order to enhance path quality and computational efficiency, an enhanced rapidly-exploring random tree (RRT) method is presented, which incorporates adaptive step-size management, goal biasing, and B-spline-based path smoothing. Furthermore, real-time local obstacle avoidance in dynamic situations is made possible by the integration of the Timed Elastic Band (TEB) algorithm. Comprehensive real-world tests have confirmed the usefulness of the suggested solutions in terms of efficiency, robustness, and practical applicability after they were implemented on an experimental platform based on the Robot Operating System (ROS).


  • Open Access

    Optimizing LSTM networks for improved river hydrological prediction

    by Ying Nie, Kok Hwa Yu, Choe Peng Leo, Kin Sam Yen, Yang Wang

    This study focuses on evaluating the performance of water level prediction utilizing the sparrow search algorithm (SSA) optimization with the long short-term memory (LSTM) model. This hybrid SSA-LSTM model is employed to forecast the Kelantan River water level. Data pre-processing techniques such as mean interpolation and the wavelet denoising technique are utilized to address the missing data issue and ensure data reliability, respectively. The accuracy of the predictions made by the SSA-LSTM model is compared with that of the manually tuned LSTM and PSO-LSTM models. It is demonstrated that the SSA-LSTM model yields superior predictions over both the traditional LSTM and PSO-LSTM models. Utilizing the SSA optimization, this gives rise to improved accuracy, with a reduction in the root mean square error (RMSE) from 0.5829 to 0.5302, which infers a 9.04% reduction as compared with that of the traditional technique. In addition, the coefficient of determination (R2) increases from 0.8160 to 0.8478, thereby indicating improved fitting and generalization abilities. Furthermore, Taylor analysis is performed to assess the water level prediction accuracy, while peak analysis is conducted to evaluate water level prediction during elevated water scenarios.

  • Open Access

    Heat transfer enhancement of microchannel heat sink with water-droplet grooves: geometrical and grooves arrangement effects

    by Yan Hao Soo, Kok Hwa Yu, Mohd Azmi Ismail

    A series of conjugate heat transfer simulations were performed to examine the effects of geometrical modifications and different grooves arrangement on the fluid flow and heat transfer characteristics of coolant flow in microchannel heat sinks containing water-droplet grooves. Numerical investigations were then conducted on the microchannel heat sinks which incorporate the water-droplet grooves for a primary channel inlet velocity in the range of 0.2–1.0 m/s. The presence of water-droplet grooves is found to yield better thermo-hydraulic performance, in comparison with the conventional microchannel heat sink. Using this novel design, the reduction in the thermal resistance is revealed to be more than triple than the increase in the pumping power needed. Based on the simulation results, it is inferred that staggered grooves arrangement would yield a better overall performance, as compared with that of in-line grooves arrangement. In addition, the effects of groove pitch length modification on the fluid flow and thermal are explored, along with the assessment of the aspect ratio influence on the thermo-hydraulic performance. To achieve better heat dissipation, it is preferable to have a shorter groove pitch length and a larger groove aspect ratio of the water-droplet grooved microchannel heat sink.


  • Electro-osmotic flow in channel: effects of superhydrophobic surface structures sizing

    by Mei Jing, Kok Hwa Yu, Mohd Sharizal Abdul Aziz, Wei Shyang Chang

    Numerical study on electro-osmotic flow over superhydrophobic transverse grooves and ribs have been explored. With uncharged liquid-gas interface, smaller electro-osmotic axial flow is attained. The electro-osmotic axial flow is found to be significantly influenced by the gas area fraction and the normalized groove-rib spacing. In the presence of the liquid-gas interfaces, slip flows are observed over these interfaces, consistent with that reported in pressure-driven flow. As the flow is permitted in the vicinity of the liquid-gas interface, flow redistribution is observed along the channel. Higher bulk flow with reducing flow magnitude is observed close to the wall when solid wall is present, and vice versa. Despite the alteration of the axial flow, conservation of flow along axial direction is attained. To estimate the normalized electro-osmotic axial flow for this flow condition, a new correlation formulation is proposed to accurately predict the averaged axial flow magnitude for different values of gas area fraction and normalized groove-rib spacing in the range between 0.1 and 4.

  • Open Access

    Numerical investigation on heat transfer enhancement of graphene coated circular pin fin

    by Muhsen Banur Asoul, Kok Hwa Yu, Khaled Ali Mohammad Al-Attab, Mohd Azmi Ismail, How Tion Puay

    The benefit of the graphene coating is twofold. Firstly, the lower temperature on the graphene coated pin fin can be attained, thus ensuring the application system to operate at lower temperatures, preventing overheating. Secondly, the graphene coated fin could yield more uniform distribution of temperature along the pin fin, thereby producing a more uniform heat dissipation along the fin. With such a coating, an improved cooling system can be created, in alleviating the heat transfer limit of conventional heat sink. Numerical results reported in this study revealed that the graphene coating is consistently beneficial for heat dissipation for all cases investigated, regardless of the air velocity, coating thickness, and pin fin length. At the longest pin fin examined, which is 25D, the temperature difference between the base and the tip of the pin fin can be reduced to 1.70K, as compared to 10.60 K temperature difference for the conventional pin fin. Additionally, thermal resistance reduction of up to 56.60% can be attained using the graphene coating technique.