Research Papers - Department of Civil Engineering
Permanent URI for this collectionhttps://rda.sliit.lk/handle/123456789/598
Browse
307 results
Filters
Advanced Search
Filter by
Settings
Search Results
Publication Open Access QPred: A Lightweight Deep Learning-Based Web Pipeline for Accessible and Scalable Streamflow Forecasting(Tech Science Press, 2026) Makumbura, R.K; Wijesundara, H; Sajindra, H; Rathnayake, U; Kumar, V; Duraibabu, D; Sen, SAccurate streamflow prediction is essential for flood warning, reservoir operation, irrigation scheduling, hydropower planning, and sustainable water management, yet remains challenging due to the complexity of hydrological processes. Although data-driven models often outperform conventional physics-based hydrological modelling approaches, their real-world deployment is limited by cost, infrastructure demands, and the interdisciplinary expertise required. To bridge this gap, this study developed QPred, a regional, lightweight, cost-effective, web-delivered application for daily streamflow forecasting. The study executed an end-to-end workflow, from field data acquisition to accessible web-based deployment for on-demand forecasting. High-resolution rainfall data were recorded with tipping-bucket gauges and loggers, while river water depth in the Aglar and Paligaad watersheds was converted to discharge using site-specific rating curves, resulting in a daily dataset of precipitation, river water level and discharge. Four DL architectures were trained, including vanilla Long Short-Term Memory (LSTM), stacked LSTM, bidirectional LSTM, and Gated Recurrent Unit (GRU), and evaluated using Nash-Sutcliffe Efficiency (NSE), Coefficient of Determination (R2), Root-Mean-Square-Error-Standard-Deviation Ratio (RSR), and Percentage Bias (PBIAS) metrics. Performance was watershed-specific, as the vanilla LSTM demonstrated the best generalisation for the Aglar watershed (R2 = 0.88, NSE = 0.82, RMSE = 0.12 during validation), while the GRU achieved the highest validation accuracy in Paligaad (R2 = 0.88, NSE = 0.88, RMSE = 0.49). All models achieved satisfactory to excellent performance during calibration (R2 > 0.91, NSE > 0.91 for both watersheds), demonstrating strong capability to capture streamflow dynamics. The highest performing models were selected and embedded into the QPred application. QPred was developed as a lightweight web pipeline, utilising Google Colab as the primary execution environment, Flask as the backend inference framework, Google Drive for artefact storage, and Ngrok for secure HTTPS tunnelling. A user-friendly front end utilises range sliders (bounded by observed minima and maxima) to gather inputs and provides discharge data along with metadata, thereby enhancing transparency. This work demonstrates that accurate, context-aware deep learning models can be delivered through low-cost, web-based platforms, providing a reproducible and scalable pipeline for hydrological applications in other watersheds and for practitioners. CopyrightPublication Embargo Enhancing the effectiveness of satellite precipitation products with topographic and seasonal bias correction(Elsevier B.V., 2026-02) Wanniarachchi, S; Sarukkalige, R; Hapuarachchi, H.A. P; Gomes, P.I.A; Rathnayake, UEstimating precipitation distribution across large regions is crucial for understanding water availability, planning infrastructure, and forecasting flood hazards. Traditional gauge-based methods face challenges, particularly with sparse gauge networks. In response, satellite-based, near-real-time (NRT) precipitation data has gained popularity, especially in poorly gauged watersheds. However, satellite precipitation data quality is often compromised by latency, atmospheric complexities, and topographic effects, resulting in nonlinear errors. To overcome the research gap, this study introduces the Heavy Rain Peak Adjustment (HRPA) method alongside the well-established Seasonal Autoregressive Integrated Moving Average (SARIMA) model for satellite precipitation bias correction. The analysis utilised Global Satellite Mapping of Precipitation (GSMaP-NRT) data and hourly precipitation records from 31 rain gauges in the Ovens River region of Australia. On average, the mean residual of observed and GSMaP-NRT precipitation was −0.02 mm. Additionally, the HRPA method yielded better linear regression R2(0.911), NSE (log) (−0.847), and RMSE (0.628) compared to SARIMA. The results indicate that HRPA outperforms SARIMA, particularly at lower elevations, whereas SARIMA struggles at higher elevations, underscoring its limitations in those areas. Additionally, autocorrelation and partial autocorrelation plots for some stations in hilly areas show significant wave-like patterns, indicating greater uncertainty in satellite precipitation estimates over complex terrain. For several stations, autocorrelations at 24 and 48-hour lags suggest a systematic influence of past residuals on future ones, emphasizing the need for further refinement in satellite precipitation correction methods for these regions.Publication Embargo Long-term recovery from the 2004 Indian ocean tsunami in two Sri Lankan east coast municipalities(Elsevier Ltd, 2026-01) Thamboo, J; Josiah, R; Saja, A; Salah, P; Rossetto, T; Dias, PSri Lanka was the second most affected country after Indonesia, in the 2004 Boxing Day Indian Ocean tsunami (IOT). A study mission was therefore carried out twenty years after the 2004 IOT to assess the recovery of the affected regions, especially in the Eastern region of Sri Lanka, focusing on two of the most affected municipalities, i.e. Kalmunai and Batticaloa. The social and infrastructure characteristics of resettlements/relocations/new settlements in the affected regions, presence of critical infrastructure, preparedness and early warning systems installed have been assessed. It was observed that similar approaches have been adopted to plan the community relocation in both of these municipalities, while the significant reemergence of residential and commercial developments in the coastal stretches of Kalmunai municipality have been noted. Exposure analyses have revealed that there are still some critical infrastructure situated in the tsunami hazard zones. It can be construed that these municipalities have recovered from the physical losses incurred, and spatial planning is in place for future developments considering the tsunami risk. Challenges and opportunities from their differing geographical contexts appear to have been judiciously handled. However, shortcomings are noted in actual implementation due to various reasons, such as limited resources, availability of funding and preference of communities to live close to their original lands. Improving the resilience of infrastructure by designing against the expected tsunami hazard and multi-hazards, regular verification of the early warning systems and evacuation procedures are emphasized to mitigate the impacts from future tsunami.Publication Open Access Waste Polyethylene Reinforced with Coconut Fibers for Sustainable Construction: A Mechanical and Physical Property Evaluation Study(Dr D. Pylarinos, 2025-10-06) Dharmaratne, P. D; Galabada G.H; Malkanthi S.N; U.Halwatura R.This study evaluates the feasibility of using waste polyethylene as a construction material. To achieve this, a series of polymer composites were developed using waste Low-Density Polyethylene (LDPE) reinforced with coconut fiber (coir). The mechanical properties, including the tensile strength, flexural strength, impact strength, and elastic modulus, were assessed, along with the water absorption as a key physical property by following the ASTM standards. The composites were fabricated using the hand layup technique with varying coir-to-LDPE weight ratios and fiber lengths, followed by a hot-press machine manufacturing under controlled conditions. The results demonstrated that different fiber lengths and content levels influenced the mechanical properties, optimizing them at various configurations. A maximum tensile strength of 12.56 MPa was achieved using 40% coir content with 4 cm fiber length. The highest elastic modulus value of 0.46 GPa was achieved at 50% fiber content with 4 cm fibers. At 30% fiber content with 3 cm length, the maximum flexural strength value of 33.77 MPa was obtained. The impact strength reached its maximum value of 1.22 kJ/m² with 40% fiber content and 2 cm fiber length. The high water absorption exhibited by the composites, can be mitigated by applying waterproofing chemicals immediately after fabrication. It was found that the integration of fiber content and length affects the composite's properties. Depending on the required characteristics, appropriate fiber lengths and mix proportions can be selected, making these composites suitable for various applications in the construction industry. Additionally, proper waterproofing immediately after manufacturing the composite is proposed to enhance its performance as a construction material.Publication Embargo Resilience of masonry infilled reinforced concrete school buildings in low to moderate seismic regions: case study of Sri Lankan schools(Springer Science and Business Media, 2025-12-08) Raheem, S; Thamboo, J; Mallikarachi, C; Wijesundara, K; Dias, PThe resilience of school buildings in high seismic regions is widely emphasised and evaluated. However such resilience in low-to-medium seismic regions are generally overlooked due to the lower probability of occurrence and low-to-medium intensities expected. Nonetheless, nominal seismic provisions should be provided for the life safety of pupils occupying these school buildings. Therefore, this study was focused on assessing the level of seismic resilience of school buildings in low-to-medium seismic regions, where the archetypal school buildings in Sri Lanka and the seismic demand in the country were taken as the case study. A framework to quantify resilience, incorporating social recovery aspects, was adopted to evaluate the seismic resilience. The resilience of the same archetypal school buildings subjected to different nominal retrofitting methods was also assessed to verify the improvement in resilience compared to un-retrofitted buildings. The epistemic and aleatory uncertainties were incorporated by using 25 different recorded seismic accelerograms and Monte-Carlo simulation of material properties (twenty sets of randomised values), respectively; with 500 combinations (aleatoric and epistemic) being analysed for each building type considered. Seismic resilience indices (RIs) obtained indicate that the school buildings with retrofitted configurations are certainly better than un-retrofitted ones, especially for higher hazard levels. Increases in the RIs are in the range of 36.6–91.2% for the highest hazard level. Sensitivity analyses were also carried out to ascertain parameter influence on RIs. The proposed nominal retrofitting solutions for these school building archetypes generate adequate resilience against the seismic hazards demarcated for the country.Publication Embargo Uniaxial compressive response of cement mortar with waste aluminium fibre sourced from electrical distribution cables(Springer Science and Business Media, 2025-01) Perera K.D.Y.G.; Ahamed Y.L.F; Somarathna H.M.C.C; Jayasekara D.A.B.P.M; Mohotti D; Raman S.NElectrical distribution and communication cables cease to function for transmission when their length is insufficient, and it is considered as it approaches the end of their useful lives. Further, the disposal techniques are not eco-friendly. This study aimed to evaluate the feasibility of cement mortar systems with the inclusion of aluminium fibre extracted from electrical distribution cables. Two diameters of 1.35 mm and 1.70 mm and two lengths of 10 mm and 15 mm fibres were used while incorporating four volume ratios, particularly 0.5%, 1.0%, 1.5%, and 2.0% to evaluate the effect of the length, diameter and volume ratios. The compression test and density test were performed to study the behaviour of Metal Fibre Reinforced Mortar (MFRM) systems under both dry and wet states. Compared to conventional mortar, the ultimate compressive strength of MFRM systems was increased up to 39.4% in 1.5% of fibre addition under the 28-day dry state, where the 1.5% volume ratio showed the best performance under compressive loads. Strain at ultimate strength, modulus of elasticity and strain energy also showed improvements with the fibre inclusion up to 74.4%, 87.3%, and 106.6% respectively. Fibres with higher aspect ratios showed significant effectiveness among the aforementioned fibre variations. The overall results highlighted that the MFRM with 1.5% of fibres performed expertly with 15 mm length and 1.35 mm diameter under compression loadsPublication Embargo Machine learning prediction of web-crippling strength in cold-formed steel beams with staggered slotted perforations(Elsevier Ltd, 2025-01) Gatheeshgar, P; Ranasinghe R.S.S; Simwanda, L; Meddage D.P.P.; Mohotti, DThe application of staggered slotted perforations in cold-formed steel (CFS) members is increasingly prominent in modern construction. Understanding the web-crippling strength of CFS beams, especially those with staggered slotted perforations, is crucial in structural engineering. This study employs machine learning (ML) models to predict the web-crippling strength of these beams under one-flange loading conditions, specifically interior-one-flange and end-one-flange loading. The research utilises a comprehensive dataset comprising 576 web-crippling strength results obtained through numerical modelling. The dataset includes parameters such as yield strength, thickness, corner radius, slot length, slot width, and bearing plate length. Four different ML algorithms—k-nearest neighbour (KNN), random forest (RF), support vector regression (SVR), and artificial neural network (ANN)—are developed and evaluated. Performance metrics, including coefficient of determination (R²), mean squared error (MSE), root mean square error (RMSE), mean absolute error (MAE) and mean normalised bias (MNB) are used to assess model accuracy. The random forest model outperforms others in both the training and testing phases. Shapley additive explanation (SHAP) and partial dependence plots further analyse the influence of input features on web crippling strength. This study presents a robust ML-based approach for predicting web crippling strength, providing engineers with a time-efficient alternate method.Publication Embargo Selecting suitable passive design strategies for residential high-rise buildings in tropical climates to minimize building energy demand(Elsevier Ltd, 2025) Perera, U.S; Weerasuriya A.U; Zhang, X; Ruparathna R; Tharaka M.G.I; Lewangamage C.SPassive design strategies (PDS) are a fitting solution to reduce the ever-growing energy cost of residential high-rise buildings in tropical regions. However, PDSs’ building energy saving potential significantly varies with local climate conditions, but it has been sparsely investigated. Hence, this study investigated the energy-saving efficiency of eight common PDSs integrated into a typical residential high-rise building in three sub-climates: extremely hot humid (0 A), very hot humid (1 A), and warm humid (3 A) defined by ASHRAE for the tropical climate. This study developed a Building Performance Analysis (BPA) workflow with a BIM-based simulation framework and local and global sensitivity analyses for the building energy analysis. The global sensitivity analysis revealed that low e-coating on glasses is the most influential PDS for 0 A and 1 A climates, but it has a negative effect in the sub-climate zone 3 A. The low-conducting exterior walls are the most effective PDS in the sub-climate zone 3 A, but they are poorly performed in the other two sub-climate zones. Based on the energy calculation and sensitivity analysis, this study proposes the best PDS groups, saving up to 40.1 %, 63.5 %, and 31.7 % of average annual building consumption in the sub-climate zones, 0 A, 1 A, and 3 A climates.Publication Open Access Effectiveness of Porous Concrete Pavements in Removing Total Suspended Solids from Urban Stormwater Runoff(International Society for Environmental Information Sciences, 2025-05-25) Wijewickrama D.; Miguntanna N; Siriwardhana K.D; Kalaimathy S.N.; Kantamaneni K; Rathnayake UThis study investigates the effectiveness of total suspended solids removal in porous concrete pavement (PCP) with only changing aggregate size of the mix design and the thicknesses of the pervious concrete pavement specimen. The study used two different aggregate sizes, 10 ~ 14, and 14 ~ 19 mm, with a third mix percentage consisting of 50% of both aggregate sizes. Water content was main-tained low in the mix designs since it influenced the porosity of the concrete and the water flow rate after solidifying the concrete. Slump tests were done to find the workability and all 3 mix designs’ slump was near zero, and casted cubes were used to determine t he compres-sive strength of each mix design. The results revealed that aggregate size had a direct impact on compressive strength, with smaller aggregate mix designs having higher strength. The study validated PCP’s filtration properties as well as the percentage removal of total suspended solids. The removal efficiency was found to increase with the thickness of the PCP and the use of smaller aggregate sizes. Also, data revealed that where higher porosity facilitates improved filtration and reduces Total Suspended Solids (TSS) in st orm water runoff. Furthermore, Infiltration data shows, where higher TSS Reduction Efficiency is associated with improved infiltration capacity, effectively mitigating the impact of stormwater runoff on water quality. According to the study, PCP is a better alternative for stormwater management systems and may be utilized for harvesting and cleaning purposes as non-portable water. The findings of this study might assist in determining the individual performance of each porous concrete pavement type and encourage wider use of these pavements to reduce the need for impermeable surfaces for stormwater management.Publication Embargo Feasibility of hydrological modelling for intermittent streams using HEC-HMS: a process evaluation(Taylor and Francis, 2025-01-08) Perera, M. D.D; Gomes, P.I.AThe use of hydrological software in simulating the rainfall–runoff relationships of intermittent streams is rather unfound due to their dynamic flow regimes. This study assessed the feasibility of using Hydrologic Engineering Center–Hydrologic Modelling System (HEC-HMS), a widely used open-source hydrological modelling software, in discharge simulation of intermittent streams in a dry tropical zone. Individual calibration was required for each stream, even in adjoining sub-catchments with the same geology and climate. Event-based models with transitional periods captured seasonal variations in catchment characteristics. Strong correlations (Pearson’s r > 0.7, P < 0.05) between observed and simulated discharges indicated model success and, after calibration for one season, flows of the next (similar) season could be predicted without further adjustments. Baseflow and channel infiltration were the most sensitive parameters in the wet and dry seasons, respectively. This study demonstrated the possibility of building accurate hydrological models for intermittent streams by incorporating seasonal variations and extensive calibration.