Institutional-Repository, University of Moratuwa
Welcome to the University of Moratuwa Digital Repository, which houses postgraduate theses and dissertations, research articles presented at conferences by faculties and departments, university-published journal articles and research publications authored by academic staff. This online repository stores, preserves and distributes the University's scholarly work. This service allows University members to share their research with a larger audience.
 Research Publications |  Thesis & Dissertation |  E- Books |
 UoM Journal Publications |  UoM Conference Proceedings |  Articles published in Scimago's Q1 journals |  UoM Research Reports |  Other Articles authored by UoM staff |
Recent Submissions
item: Conference-Full-text
Investigating the Kanban size of the material flow in apparel manufacturing process
(IEEE, 2024) Navodi, WS; Nixman, L; Ratnayake, V; Abeysooriya, RP
This study seeks to optimise material flow within the apparel manufacturing process by identifying an optimal Kanban size. Work-in-progress (WIP) fluctuations, resulting from various factors such as product type variations, supplier fluctuations, and fluctuating customer demands, pose significant challenges in dynamic manufacturing environments. Managing these fluctuations through the identification of an optimal Kanban size holds promise for improving production processes. A spreadsheet simulation model was devised that employs cost components that include setup time cost, WIP cost, material handling cost, and under-utilisation of resources cost. Through a comparative study, the impact of varying Kanban size on total costs was evaluated. This systematic examination shed light on how modifications in the Kanban size influenced individual cost components within the cost-objective function. The study aims to determine the appropriate bundle size, called Kanban size, to improve material flow efficiency in apparel manufacturing, thus improving operational efficiency and reducing costs. By optimising Kanban size and minimising WIP fluctuations, the study aims to streamline production processes, resulting in shorter lead times, increased productivity, and reduced waste.
item: Conference-Full-text
Development of a finite element model of glass balustrade under soft body impact test
(IEEE, 2024) Wanigasooriya, TWMC; Hidallana-Gamage, HD; Baskaran, K
This paper presents a finite element (FE) model developed to analyze the behavior of glass balustrades under soft body impact test to simulate the human impact loads. The LS-DYNA explicit FE code is used to perform the FE analysis. In the initial stages of the research, field investigations were conducted to study the support conditions, glass types, and the response of glass balustrades to different loading cases in Sri Lanka. A case study of a failure of glass balustrade in a building due to heavy wind loading is discussed in this paper, highlighting the need for a proper and safe design approach for glass balustrades in Sri Lanka. The numerical results obtained from the LS-DYNA FE model were validated against the EN 12600:2002 European standard, which gives the experimental results of the in-plane strains of the tempered glass plate. In addition, the material properties used for the glass plate and impactor and the boundary conditions are thoroughly discussed in this paper. Finally, the results of the field investigation and numerical model reveal critical insights into the design and construction of glass balustrades.
item: Conference-Full-text
FL-CycleGAN: enhancing mobile photography with federated learning-enabled CycleGAN
(IEEE, 2024) Walgama, R; Mahima, KTY
Mobile image photography is continuously emerging as an area of interest, yet achieving professional-level camera quality remains a challenge due to hardware limitations. In order to improve the images taken from mobile phones, deep learningbased image processing techniques such as convolutional neural networks are proposed. However, these networks are typically trained using large amounts of paired data and lack continuous training using images captured from mobile phone users. This is because, in reality, creating paired image datasets from usercaptured images is challenging and may lead to user privacy issues. As a solution to this challenge, this research proposes FLCycleGAN, a novel federated learning-based CycleGAN designed to improve the colors of mobile images continuously using usercaptured images in an unpaired manner. The evaluations on the ZurichRAW to RGB dataset reveal that FL-CycleGAN reconstructs the colors of mobile images with an average PSNR value of 18.46 and SSIM value of 0.707, demonstrating comparable results to state-of-the-art networks based on paired images. Furthermore, FL-CycleGAN reconstructs high-resolution images with a size of 3968 _ 2976 in under 0.005 seconds.
item: Conference-Full-text
Hybrid energy storages for islanded microgrids with high solar PV penetration
(IEEE, 2024) Ediriweera, WEPS; Lidula, NWA
Distribution feeders with high penetration of renewable power generation can be converted into microgrids to provide power with increased reliability and high power quality. Voltage stability during faults in islanded microgrids with a low X/R ratio would be challenging due to the limited
short circuit current contribution of the inverter-based distributed energy resources. Virtual synchronous inertia is used to increase the inertia of inverter-based microgrids to provide the same performance as synchronous generators. This article proposes a designing and controlling system for a supercapacitor-battery hybrid energy storage system for a microgrid in islanded operation. A supervisory controller is used to decide the state of the supercapacitor according to the severity of the fault, and the state of charge level of the energy storage. The performance of the proposed energy storage system is verified using a microgrid derived based on a real distribution system. The results illustrate that during a fault, the proposed hybrid energy storage system can improve the voltage ridethrough capability of the microgrid compared to using only the conventional battery storage systems.
item: Conference-Full-text
Fuzzy-based vehicle yaw stability control system with torque vectoring and active steering
(IEEE, 2024) Herath, P; Rajasuriya, H; Dasanayake, N; Perera, S; Subasinghe, L; Gamage, J
Yaw stability control is essential for the safe operation of a vehicle. This paper presents a robust fuzzy-based integration of active steering with torque vectoring in order to achieve vehicle yaw stability. A single-track vehicle model is used to mathematically model the vehicle, accommodating the active front steering and torque vectoring systems. A fuzzy logic controller is employed to calculate the required corrective steering and differential torque based on the error in yaw rate. The implementation of this system is carried out using MATLAB. Furthermore, an analysis of the effect of the proportion of corrective assistance given by active front steering and torque vectoring is done. The results indicate enhanced vehicle manoeuvrability during cornering, as evidenced by the near alignment of the measured yaw rate with the desired values. The simulations are carried out for two speeds where the effect of the controller was shown significant for both cases. The normalised root mean square error is reducing from 10% to 3.94% in 60 km/h case and from 12.9% to 4.01% in 100 km/h case. The results also show that corrective active steering has a greater effect on yaw stability than an equivalent amount of corrective torque vectoring.