Browsing by Author "Rathnayake, U"

Now showing 1 - 20 of 63

Open Access
A novel application with explainable machine learning (SHAP and LIME) to predict soil N, P, and K nutrient content in cabbage cultivation
(Elsevier B.V., 2025-03-06) Abekoon, T; Sajindra, H; Rathnayake, N; Ekanayake, I, U; Jayakody, A; Rathnayake, U
Cabbage (Brassica oleracea var. capitata) is commonly cultivated in high altitudes and features dense, tightly packed leaves. The Green Coronet variety is well-known for its robust growth and culinary versatility. Maximizing yield is crucial for food sustainability. It is essential to predict the soil’s major nutrients (nitrogen, phosphorus, and potassium) to maximize the yield. Artificial intelligence is widely used for non-linear predictions with explainability. This research assessed the predictive capabilities of soil nitrogen, phosphorus, and potassium levels with explainable machine learning methods over an 85-day cabbage growth period. Experiments were conducted on cabbage plants grown in central hills of Sri Lanka. SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) were used to clarify the model’s predictions. SHAP analysis showed that high feature values of the number of days and plant average leaf area negatively impacted for nutrient predictions, while high feature values of leaf count and plant height had a positive effect on the nutrient predictions. To validate the results, 15 greenhouse-grown cabbage plants at various growth stages were selected. The nitrogen, phosphorus, and potassium levels were measured and compared with the predicted values. These insights help refine predictive models and optimize agricultural practices. A user-friendly application was developed to improve the accessibility and interpretation of predictions. This tool is a user-friendly platform for end-users, enabling effective use of the model’s predictive capabilities.
Open Access
Analysis of Meandering River Morphodynamics Using Satellite Remote Sensing Data—An Application in the Lower Deduru Oya (River), Sri Lanka
(MDPI, 2022-07-16) Basnayaka, V; Samarasinghe, J. T; Gunathilake, M. B; Muttil, N; Hettiarachchi, D. C; Abeynayaka, A; Rathnayake, U
River meandering and anabranching have become major problems in many large rivers that carry significant amounts of sediment worldwide. The morphodynamics of these rivers are complex due to the temporal variation of flows. However, the availability of remote sensing data and geographic information systems (GISs) provides the opportunity to analyze the morphological changes in river systems both quantitatively and qualitatively. The present study investigated the temporal changes in the river morphology of the Deduru Oya (river) in Sri Lanka, which is a meandering river. The study covered a period of 32 years (1989 to 2021), using Landsat satellite data and the QGIS platform. Cloud-free Landsat 5 and Landsat 8 satellite images were extracted and processed to extract the river mask. The centerline of the river was generated using the extracted river mask, with the support of semi-automated digitizing software (WebPlotDigitizer). Freely available QGIS was used to investigate the temporal variation of river migration. The results of the study demonstrated that, over the past three decades, both the bend curvatures and the river migration rates of the meandering bends have generally increased with time. In addition, it was found that a higher number of meandering bends could be observed in the lower (most downstream) and the middle parts of the selected river segment. The current analysis indicates that the Deduru Oya has undergone considerable changes in its curvature and migration rates.
Open Access
Analysis of Multi-Temporal Shoreline Changes Due to a Harbor Using Remote Sensing Data and GIS Techniques
(MDPI, 2023-05-06) Zoysa, S; Basnayake, V; Samarasinghe, J. T.; Gunathilake, M.B.; Kantamaneni, K; Muttil, N; Muttil, U; Rathnayake, U
Coastal landforms are continuously shaped by natural and human-induced forces, exacerbating the associated coastal hazards and risks. Changes in the shoreline are a critical concern for sustainable coastal zone management. However, a limited amount of research has been carried out on the coastal belt of Sri Lanka. Thus, this study investigates the spatiotemporal evolution of the shoreline dynamics on the Oluvil coastline in the Ampara district in Sri Lanka for a two-decade period from 1991 to 2021, where the economically significant Oluvil Harbor exists by utilizing remote sensing and geographic information system (GIS) techniques. Shorelines for each year were delineated using Landsat 5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper Plus (ETM+), and Landsat 8 Operational Land Imager images. The Normalized Difference Water Index (NDWI) was applied as a spectral value index approach to differentiate land masses from water bodies. Subsequently, the Digital Shoreline Analysis System (DSAS) tool was used to assess shoreline changes, including Shoreline Change Envelope (SCE), Net Shoreline Movement (NSM), End Point Rate (EPR), and Linear Regression Rate (LRR). The results reveal that the Oluvil coast has undergone both accretion and erosion over the years, primarily due to harbor construction. The highest SCE values were calculated within the Oluvil harbor region, reaching 523.8 m. The highest NSM ranges were recorded as −317.1 to −81.3 m in the Oluvil area and 156.3–317.5 m in the harbor and its closest point in the southern direction. The maximum rate of EPR was observed to range from 3 m/year to 10.7 m/year towards the south of the harbor, and from −10.7 m/year to −3.0 m/year towards the north of the harbor. The results of the LRR analysis revealed that the rates of erosion anomaly range from −3 m/year to −10 m/year towards the north of the harbor, while the beach advances at a rate of 3 m/year to 14.3 m/year towards the south of the harbor. The study area has undergone erosion of 40 ha and accretion of 84.44 ha. These findings can serve as valuable input data for sustainable coastal zone management along the Oluvil coast in Sri Lanka, safeguarding the coastal habitats by mitigating further anthropogenic vulnerabilities.
Open Access
Analysis of recent trends and variability of temperature and relative humidity over Sri Lanka
(India Meteorological Department, 2022-07-01) Rathnayake, U; Gunathilake, M. B; Senatilleke, U; Alyousifi, Y
The world is experiencing adverse consequences of climate change and shifts in climate regimes. Hence, studying the trends and patterns of meteorological variables is of major importance for many parties, including meteorologists, climatologists, agriculturists and hydrologists. Although several researchers have examined the trends and patterns in historical rainfall, only a few have examined the trends in atmospheric temperature. Noteworthy none of the previous studies have attempted to investigate trends in relative humidity over Sri Lanka. Therefore, identifying this existing research gap, this present paper presents a trends and variability analysis of atmospheric temperature and relative humidity of Sri Lanka. The long-term variations of minimum and maximum temperature and relative humidity records at 18 stations distributed in the three climatic zones namely, the dry zone, the intermediate zone and the wet zone in Sri Lanka were investigated for 30 years from 1990 to 2019. Annual and monthly trends were assessed using non-parametric statistical tests, including the Mann Kendall test (MK), Sen’s slope and Spearman’s rho test, while the changing points of temperature and humidity were determined using the Pettit test. In addition, the variability of climate parameters was estimated using the Coefficient of Variation (CoV). Interesting and encouraging results were obtained from the present analysis. Badulla in the intermediate climatic zone was identified with unexpected decreasing temperature trends, while several other areas were identified with expected increasing temperature and relative humidity trends. The adaptation practices based on these results would be interesting to incorporate in achieving sustainable development goals for the country
Open Access
Application of GIS Techniques in Identifying Artificial Groundwater Recharging Zones in Arid Regions: A Case Study in Tissamaharama, Sri Lanka
(MDPI, 2022-12-10) Kariyawasam, T; Basnayake, V; Wanniarachchi, S; Sarukkalige, R; Rathnayake, U
Groundwater resources are severely threatened not only in terms of their quality but also their quantity. The availability of groundwater in arid regions is highly important as it caters to domestic needs, irrigation, and industrial purposes in those areas. With the increasing population and human needs, artificial recharging of groundwater has become an important topic because of rainfall scarcity, high evaporation, and shortage of surface water resources in arid regions. However, this has been given the minimum attention in the context of Sri Lanka. Therefore, the current research was carried out to demarcate suitable sites for the artificial recharging of aquifers with the help of geographic information system (GIS) techniques, in one of the water-scarce regions in Sri Lanka. Tissamaharama District Secretariat Division (DSD) is located in Hambanthota district. This region faces periodic water stress with a low-intensity seasonal rainfall pattern and a lack of surface water sources. Hydrological, geological, and geomorphological parameters such as rainfall, soil type, slope, drainage density, and land use patterns were considered to be the most influential parameters in determining the artificial recharging potential in the study area. The GIS tools were used to carry out a weighted overlay analysis to integrate the effects of each parameter into the potential for artificial groundwater recharge. The result of the study shows that 14.60% of the area in the Tissamaharama DSD has a very good potential for artificial groundwater recharge, while 41.32% has a good potential and 39.03% and 5.05% have poor and very poor potential for artificial groundwater recharge, respectively. The southern part of the DSD and the Yala nature reserve areas are observed to have a higher potential for artificial groundwater recharge than the other areas of Tissamaharama DSD. It is recommended to test the efficiency and effects of groundwater recharge using groundwater models by simulating the effects of groundwater recharge in future studies. Therefore, the results of the current research will be helpful in effectively managing the groundwater resources in the study area.
Open Access
Appraisal of Satellite Rainfall Products for Malwathu, Deduru, and Kalu River Basins, Sri Lanka
(MDPI, 2022-10-20) Perera, H; Senaratne, N; Gunathilake, M. B; Mutill, N; Rathnayake, U
Satellite Rainfall Products (SRPs) are now in widespread use around the world as a better alternative for scarce observed rain gauge data. Upon proper analysis of the SRPs and observed rainfall data, SRP data can be used in many hydrological applications. This evaluation is very much necessary since, it had been found that their performances vary with different areas of interest. This research looks at the three prominent river basins; Malwathu, Deduru, and Kalu of Sri Lanka and evaluates six selected SRPs, namely, IMERG, TRMM 3B42, TRMM 3B42-RT, PERSIANN, PERSIANNCCS, PERSIANN-CDR against 15+ years of observed rainfall data with the use of several indices. Four Continuous Evaluation Indices (CEI) such as Root Mean Square Error (RMSE), Percentage Bias (PBIAS), Pearson’s Correlation Coefficient (r), and Nash Sutcliffe Efficiency (NSE) were used to evaluate the accuracy of SRPs and four Categorical Indices (CI) namely, Probability of Detection (POD), Critical Success Index (CSI), False Alarm Ratio (FAR) and Proportion Correct (PC) was used to evaluate the detection and prediction accuracy of the SRPs. Then, the Mann–Kendall Test (MK test) was used to identify trends in the datasets and Theil’s and Sens Slope Estimator to quantify the trends observed. The study of categorical indicators yielded varying findings, with TRMM-3B42 performing well in the dry zone and IMERG doing well in the wet zone and intermediate zone of Sri Lanka. Regarding the CIs in the three basins, overall, IMERG was the most reliable. In general, all three basins had similar POD and PC findings. The SRPs, however, underperformed in the dry zone in terms of CSI and FAR. Similar findings were found in the CEI analysis, as IMERG gave top performance across the board for all four CEIs in the three basins. The three basins’ overall weakest performer was PERSIANN-CCS. The trend analysis revealed that there were very few significant trends in the observed data. Even when significant trends were apparent, the SRP projections seldom captured them. TRMM-3B42 RT had the best trend prediction performance. However, Sen’s slope analysis revealed that while the sense of the trend was properly anticipated, the amplitude of the prediction significantly differed from that of the observed data.
Open Access
The Assessment of Climate Change Impacts and Land-use Changes on Flood Characteristics: The Case Study of the Kelani River Basin, Sri Lanka
(MDPI, 2022-10-09) Samarasinghe, J. T; Makumbura, R. K; Wickramarachchi, C; Sirisena, J; Gunathilake, M.B; Muttil, N; Yenn Teo, F; Rathnayake, U
Understanding the changes in climate and land use/land cover (LULC) over time is important for developing policies for minimizing the socio-economic impacts of riverine floods. The present study evaluates the influence of hydro-climatic factors and anthropogenic practices related to LULC on floods in the Kelani River Basin (KRB) in Sri Lanka. The gauge-based daily precipitation, monthly mean temperature, daily discharges, and water levels at sub-basin/basin outlets, and both surveyed and remotely sensed inundation areas were used for this analysis. Flood characteristics in terms of mean, maximum, and number of peaks were estimated by applying the peak over threshold (POT) method. Nonparametric tests were also used to identify the climatic trends. In addition, LULC maps were generated over the years 1988–2017 using Landsat images. It is observed that the flood intensities and frequencies in the KRB have increased over the years. However, Deraniyagala and Norwood sub-basins have converted to dry due to the decrease in precipitation, whereas Kithulgala, Holombuwa, Glencourse, and Hanwella showed an increase in precipitation. A significant variation in atmospheric temperature was not observed. Furthermore, the LULC has mostly changed from vegetation/barren land to built-up in many parts of the basin. Simple correlation and partial correlation analysis showed that flood frequency and inundation areas have a significant correlation with LULC and hydro-climatic factors, especially precipitation over time. The results of this research will therefore be useful for policy makers and environmental specialists to understand the relationship of flood frequencies with the anthropogenic influences on LULC and climatic factors.
Open Access
Brain Activity Associated with the Planning Process during the Long-Time Learning of the Tower of Hanoi (ToH) Task: A Pilot Study
(MDPI, 2022-10-28) Mitani, K; Rathnayake, N; Rathnayake, U; Linh Dang, T; Hoshino, Y
Planning and decision-making are critical managerial functions involving the brain’s executive functions. However, little is known about the effect of cerebral activity during long-time learning while planning and decision-making. This study investigated the impact of planning and decision-making processes in long-time learning, focusing on a cerebral activity before and after learning. The methodology of this study involves the Tower of Hanoi (ToH) to investigate executive functions related to the learning process. Generally, ToH is used to measure baseline performance, learning rate, offline learning (following overnight retention), and transfer. However, this study performs experiments on long-time learning effects for ToH solving. The participants were involved in learning the task over seven weeks. Learning progress was evaluated based on improvement in performance and correlations with the learning curve. All participants showed a significant improvement in planning and decision-making over seven weeks of time duration. Brain activation results from fMRI showed a statistically significant decrease in the activation degree in the dorsolateral prefrontal cortex, parietal lobe, inferior frontal gyrus, and premotor cortex between before and after learning. Our pilot study showed that updating information and shifting issue rules were found in the frontal lobe. Through monitoring performance, we can describe the effect of long-time learning initiated at the frontal lobe and then convert it to a task execution function by analyzing the frontal lobe maps. This process can be observed by comparing the learning curve and the fMRI maps. It was also clear that the degree of activation tends to decrease with the number of tasks, such as through the mid-phase and the end-phase of training. The elucidation of this structure is closely related to decision-making in human behavior, where brain dynamics differ between “thinking and behavior” during complex thinking in the early stages of training and instantaneous “thinking and behavior” after sufficient training. Since this is related to human learning, elucidating these mechanisms will allow the construction of a brain function map model that can be used universally for all training tasks.
Open Access
Cascaded Adaptive Network-Based Fuzzy Inference System for Hydropower Forecasting
(MDPI, 2022-04-10) Rathnayake, N; Rathnayake, U; Dang, T. L; Hoshino, Y
Hydropower stands as a crucial source of power in the current world, and there is a vast range of benefits of forecasting power generation for the future. This paper focuses on the significance of climate change on the future representation of the Samanalawewa Reservoir Hydropower Project using an architecture of the Cascaded ANFIS algorithm. Moreover, we assess the capacity of the novel Cascaded ANFIS algorithm for handling regression problems and compare the results with the state-of-art regression models. The inputs to this system were the rainfall data of selected weather stations inside the catchment. The future rainfalls were generated using Global Climate Models at RCP4.5 and RCP8.5 and corrected for their biases. The Cascaded ANFIS algorithm was selected to handle this regression problem by comparing the best algorithm among the state-of-the-art regression models, such as RNN, LSTM, and GRU. The Cascaded ANFIS could forecast the power generation with a minimum error of 1.01, whereas the second-best algorithm, GRU, scored a 6.5 error rate. The predictions were carried out for the near-future and mid-future and compared against the previous work. The results clearly show the algorithm can predict power generation's variation with rainfall with a slight error rate. This research can be utilized in numerous areas for hydropower development.
Open Access
Column Study for Adsorption of Copper and Cadmium Using Activated Carbon Derived from Sewage Sludge
(hindawi.com., 2022-03-22) Al-mahbashi, N; Kutty, S. R. M; Jagaba, A. H; Al-Nini, A; Ali, M; Saeed, A. A. H; Ghaleb, A. A. S; Rathnayake, U
mong the water-polluting substances, heavy metals stand out due to their carcinogenic and toxic effects on the creatures and environment. This study aimed to scrutinize the effectiveness of sewage sludge-based activated carbon in the removal of copper and cadmium from aqueous solutions in column study. Detection of breakthrough curves and related parameters was conducted by varying bed depths (3, 6, and 9 cm). The solution with an initial metal concentration (IMC) of 100 ppm was pumped to the column at a flow rate of 2 mL/min. In the process of copper removal, the breakthrough points for depths 3 cm, 6 cm, and 9 cm were achieved at 10 min, 15 min, and 60 min, respectively, whereas breakthrough points of similar depths in cadmium removal process were achieved at 5 min, 10 min, and 30 min, respectively. Adsorption kinetics were analyzed using the Adams–Bohart, Yoon–Nelson, and Thomas kinetics models. The Adams–Bohart model described only the initial part of breakthrough curves. The Thomas model represented the adsorption process with coefficients of determination (R2) ranging between 0.90–0.95 for cadmium removal and 0.89–0.96 for copper removal, while the coefficients of determination of Yoon–Nelson ranged between 0.89–0.94 for cadmium and 0.95–0.97 for copper. Yoon–Nelson was fitted well with copper removal data, while removal of cadmium data was best described by the Thomas model. This study demonstrated that using sewage sludge-based activated carbon to remove heavy metals is an alternative, more cost-effective option to reach the objectives of sustainable development.
Open Access
Comparing Combined 1D/2D and 2D Hydraulic Simulations Using High-Resolution Topographic Data: Examples from Sri Lanka—Lower Kelani River Basin
(MDPi, 2022-02-17) Samarasinghe, J. T; Basnayaka, V; Gunathilake, M. B; Azamathulla, H. M; Rathnayake, U
The application of numerical models to understand the behavioural pattern of a flood is widely found in the literature. However, the selection of an appropriate hydraulic model is highly essential to conduct reliable predictions. Predicting flood discharges and inundation extents are the two most important outcomes of flood simulations to stakeholders. Precise topographical data and channel geometries along a suitable hydraulic model are required to accurately predict floods. Onedimensional (1D) hydraulic models are now replaced by two-dimensional (2D) or combined 1D/2D models for higher performances. The Hydraulic Engineering Centre’s River Analysis System (HECRAS) has been widely used in all three forms for predicting flood characteristics. However, comparison studies among the 1D, 2D to 1D/2D models are limited in the literature to identify the better/best approach. Therefore, this research was carried out to identify the better approach using an example case study of the Kelani River basin in Sri Lanka. Two flood events (in 2016 and 2018) were separately simulated and tested for their accuracy using observed inundations and satellite-based inundations. It was found that the combined 1D/2D HEC-RAS hydraulic model outperforms other models for the prediction of flows and inundation for both flood events. Therefore, the combined model can be concluded as the better hydraulic model to predict flood characteristics of the Kelani River basin in Sri Lanka. With more flood studies, the conclusions can be more generalized.
Open Access
Comparison of different Artificial Neural Network (ANN) training algorithms to predict atmospheric temperature in Tabuk, Saudi Arabia
(researchgate.net, 2020-06) Perera, A; Azamathulla, H; Rathnayake, U
Use of Artificial neural network (ANN) models to predict weather parameters has become important over the years. ANN models give more accurate results in weather and climate forecasting among many other methods. However, different models require different data and these data have to be handled accordingly, but carefully. In addition, most of these data are from non-linear processes and therefore, the prediction models are usually complex. Nevertheless, neural networks perform well for non-linear data and produce well acceptable results. Therefore, this study was carried out to compare different ANN models to predict the minimum atmospheric temperature and maximum atmospheric temperature in Tabuk, Saudi Arabia. ANN models were trained using eight different training algorithms. BFGS Quasi Newton (BFG), Conjugate gradient with Powell-Beale restarts (CGB), Levenberg-Marquadt (LM), Scaled Conjugate Gradient (SCG), Fletcher-Reeves update Conjugate Gradient algorithm (CGF), One Step Secant (OSS), Polak-Ribiere update Conjugate Gradient (CGP) and Resilient Back-Propagation (RP) training algorithms were fed to the climatic data in Tabuk, Saudi Arabia. The performance of the different training algorithms to train ANN models were evaluated using Mean Squared Error (MSE) and correlation coefficient (R). The evaluation shows that training algorithms BFG, LM and SCG have outperformed others while OSS training algorithm has the lowest performance in comparison to other algorithms used.
Unknown
A comprehensive review to evaluate the synergy of intelligent food packaging with modern food technology and artificial intelligence field
(Springer link, 2024-07-22) Abekoon A; Sajindra, H; Samarakoon, E. R. J.; Jayakody, J.A.D.C.A; Kantamaneni, K; Rathnayake, U; Buthpitiya, B. L. S. K.
This study reviews recent advancements in food science and technology, analyzing their impact on the development of intelligent food packaging within the complex food supply chain. Modern food technology has brought about intelligent food packaging, which includes sensors, indicators, data carriers, and artificial intelligence. This innovative packaging helps monitor food quality and safety. These innovations collectively aim to establish an unbroken chain of food safety, freshness, and traceability, from production to consumption. This research explores the components and technologies of intelligent food packaging, focusing on key indicators like time–temperature indicators, gas indicators, freshness indicators, and pathogen indicators to ensure optimal product quality. It further incorporates various types of sensors, including gas sensors, chemical sensors, biosensors, printed electronics, and electronic noses. It integrates data carriers such as barcodes and radio-frequency identification to enhance the complexity and functionality of this system. The review emphasizes the growing influence of artificial intelligence. It looks at new advances in artificial intelligence that are driving the development of intelligent packaging, making it better at preserving food freshness and quality. This review explores how modern food technologies, especially artificial intelligence integration, are revolutionizing intelligent packaging for food safety, quality, reduced waste, and enhanced traceability.
Unknown
Computation of Time of Concentration Based on Two-Dimensional Hydraulic Simulation
(MDPI, 2022-10-07) Zolghadr, M; Rafiee, M.R; Esmaeilmanesh, F; Fathi, A; Tripathi, R.P; Rathnayake, U; Gunakala, S. R; Azamathulla, H. M
: Time of concentration (TC) is a parameter in runoff estimation, used to study and design different types of projects in watersheds. Any error in TC calculation leads to an inaccurate estimation of the design flow, which can lead to over-sizing or under-sizing of designed facilities that can have great economic and environmental consequences. Therefore, choosing the correct method to estimate TC is of great importance. Due to the diversity of estimation methods in the literature, the obtained TC values are different. This study aims to present a new method to calculate TC, based on its main concept, i.e., the time required for a water parcel to reach its outlet from the farthest hydrological point of a watershed. A two-dimensional hydraulic simulation was used to model the water parcel travel. A watershed was selected as a case study, and its time of concentration was determined by salt solution tracing. The field measurement results were used for calibration of the numerical simulation model. Meanwhile, 31 empirical relations in the literature were reviewed to determine the most accurate ones. Estimated TC values were compared with the measured ones, and the relative error percentage was used to evaluate the accuracy of the result. In the empirical TC methods, the maximum error was above 300%, and the minimum error was 6.7% for the field studied area. The relative errors of hydraulic simulation outputs were between 3 and 27%. The results showed that only three empirical methods, namely Simas and Hawkins, SCSlag, and Yen and Chow, had the least errors respectively equal to 6.7%, 8.660%, and 13.5%, which can be recommended for the studied area and those with similar hydrological characteristics. On the other hand, hydraulic simulation is also introduced as an efficient method to determine TC which can be used in any desired watershed.
Unknown
Data exploration on the factors associated with cost overrun on social housing projects in Trinidad and Tobago
(Elsevier Ltd, 2024-02) Chadee, A. A; Allis, C; Rathnayake, U; Martin, H; Azamathulla, H. M
This data article explores the factors that contribute to cost overrun on public sector projects within Trinidad and Tobago. The data was obtained through literature research, and structured questionnaires, designed using open-ended questions and the Likert scale. The responses were gathered from project actors and decision-makers within the public and private construction industry, mainly, project managers, contractors, engineers, architects, and consultants. The dataset was analysed using frequency, simple percentage, mean, risk impact, and fuzzy logic via the fuzzy synthetic evaluation method (FSE). The significance of the analysed data is to determine the critical root causes of cost overrun which affect public sector infrastructure development projects (PSIDPs), from being completed on time and within budget. The dataset is most useful to project and construction management professionals and academia, to provide additional insight into the understanding of the leading factors associated with cost overrun and the critical group in which they occur (political factors). Such understanding can encourage greater decisions under uncertainty and complexity, thus accounting for and reducing cost overrun on public sector projects. © 2023
Unknown
Deep Machine Learning-Based Water Level Prediction Model for Colombo Flood Detention Area
(MDPI, 2023-02-08) Herath, M; Jayathilaka, T; Hoshino, Y; Rathnayake, U
Machine learning has already been proven as a powerful state-of-the-art technique for many non-linear applications, including environmental changes and climate predictions. Wetlands are among some of the most challenging and complex ecosystems for water level predictions. Wetland water level prediction is vital, as wetlands have their own permissible water levels. Exceeding these water levels can cause flooding and other severe environmental damage. On the other hand, the biodiversity of the wetlands is threatened by the sudden fluctuation of water levels. Hence, early prediction of water levels benefits in mitigating most of such environmental damage. However, monitoring and predicting the water levels in wetlands worldwide have been limited owing to various constraints. This study presents the first-ever application of deep machine-learning techniques (deep neural networks) to predict the water level in an urban wetland in Sri Lanka located in its capital. Moreover, for the first time in water level prediction, it investigates two types of relationships: the traditional relationship between water levels and environmental factors, including temperature, humidity, wind speed, and evaporation, and the temporal relationship between daily water levels. Two types of low load artificial neural networks (ANNs) were developed and employed to analyze two relationships which are feed forward neural networks (FFNN) and long short-term memory (LSTM) neural networks, to conduct the comparison on an unbiased common ground. The LSTM has outperformed FFNN and confirmed that the temporal relationship is much more robust in predicting wetland water levels than the traditional relationship. Further, the study identified interesting relationships between prediction accuracy, data volume, ANN type, and degree of information extraction embedded in wetland data. The LSTM neural networks (NN) has achieved substantial performance, including R2 of 0.8786, mean squared error (MSE) of 0.0004, and mean absolute error (MAE) of 0.0155 compared to existing studies.
Unknown
Dynamic control of urban sewer systems to reduce combined sewer overflows and their adverse impacts
(10.1016/j.jhydrol.2019.124150, 2019-09-15) Rathnayake, U; Anwar, A. H. M. F
Sewer network planners use control algorithms, based on optimization techniques, to control urban wastewater systems. These control algorithms have been used to ease the stress on the sewer networks and then, to reduce or to minimize the combined sewer overflows (CSOs). CSOs are not only risking human health but also adversely affecting the aquatic lives. Therefore, many cities try to avoid CSOs. However, this cannot be done to the perfect level due to the capacity limitations of the existing combined sewer networks. In addition, climate variabilities have caused unpredictable precipitation increments and therefore, the control is extremely difficult. Therefore, considering the spatial and temporal variations of runoffs and qualities of stormwater generated from the pre- cipitation, an enhanced optimal control algorithm is illustrated in this paper to control the existing combined sewer networks. Minimizing the pollution load to the receiving water and minimizing the cost of wastewater treatment and pump operation are the two objective functions in the developed optimization algorithm. The algorithm was then successfully applied to a real-world combined sewer network in Liverpool, United Kingdom. Results reveal that the developed optimal control model is capable of handling the dynamic control settings of combined sewer system to minimize the two objective functions simultaneously. With a little computational appreciation, the developed optimal control model can be well-used in the real-time control of combined sewer networks
Unknown
An Efficient Automatic Fruit-360 Image Identification and Recognition Using a Novel Modified Cascaded-ANFIS Algorithm
(MDPI, 2022-06-10) Rathnayake, N; Rathnayake, U; Dang, T. L; Hoshino, Y
Automated fruit identification is always challenging due to its complex nature. Usually, the fruit types and sub-types are location-dependent; thus, manual fruit categorization is also still a challenging problem. Literature showcases several recent studies incorporating the Convolutional Neural Network-based algorithms (VGG16, Inception V3, MobileNet, and ResNet18) to classify the Fruit-360 dataset. However, none of them are comprehensive and have not been utilized for the total 131 fruit classes. In addition, the computational efficiency was not the best in these models. A novel, robust but comprehensive study is presented here in identifying and predicting the whole Fruit-360 dataset, including 131 fruit classes with 90,483 sample images. An algorithm based on the Cascaded Adaptive Network-based Fuzzy Inference System (Cascaded-ANFIS) was effectively utilized to achieve the research gap. Color Structure, Region Shape, Edge Histogram, Column Layout, Gray-Level Co-Occurrence Matrix, Scale-Invariant Feature Transform, Speeded Up Robust Features, Histogram of Oriented Gradients, and Oriented FAST and rotated BRIEF features are used in this study as the features descriptors in identifying fruit images. The algorithm was validated using two methods: iterations and confusion matrix. The results showcase that the proposed method gives a relative accuracy of 98.36%. The Fruit-360 dataset is unbalanced; therefore, the weighted precision, recall, and FScore were calculated as 0.9843, 0.9841, and 0.9840, respectively. In addition, the developed system was tested and compared against the literature-found state-of-the-art algorithms for the purpose. Comparison studies present the acceptability of the newly developed algorithm handling the whole Fruit-360 dataset and achieving high computational efficiency.
Unknown
Estimation of Potential Evapotranspiration across Sri Lanka Using a Distributed Dual-Source Evapotranspiration Model under Data Scarcity
(Hindawi, 2022-04-04) Senatilleke, U; Abeysiriwardana, H. D; Makumbura, R. K; Faisal Anwar, A. H. M; Rathnayake, U
Evapotranspiration estimations are not common in developing countries though most of them have water scarcities for agricultural purposes. erefore, it is essential to estimate the rates of evapotranspiration based on the available climatic parameters. Proper estimations of evapotranspiration are unavailable to Sri Lanka, even though the country has a signi cant agricultural contribution to its economy. erefore, the Shuttleworth–Wallace (S-W) model, a process-based two-source potential evapotranspiration (PET) model, is implemented to simulate the spatiotemporal distribution of PET, evaporation from soil (ETs), and transpiration from vegetation canopy (ETc) across the total landmass of Sri Lanka. e country was divided into a grid with 6km × 6km cells. e meteorological data, including rainfall, temperature, relative humidity, wind speed, net solar radiation, and pan evaporation, for 14 meteorological stations were used in this analysis. ey were interpolated using Inverse Distance Weighting (IDW), Universal kriging, and iessen polygon methods as appropriate so that the generated thematic layers were fairly closer to reality. Normalized Dierence Vegetation Index (NDVI) and soil moisture data were retrieved from publicly available online domains, while the threshold values of vegetation parameters were taken from the literature. Notwithstanding many approximations and uncertainties associated with the input data, the implemented model displayed an adequate ability to capture the spatiotemporal distribution of PET and its components. A comparison between predicted PET and recorded pan evaporations resulted in a root mean square error (RMSE) of 0.75 mm/day. e model showed high sensitivity to Leaf Area Index (LAI). e model revealed that both spatial and temporal distribution of PETis highly correlated with the incoming solar radiation uxes and aected by the rainfall seasons and cultivation patterns. e model predicted PET values accounted for 80–90% and 40–60% loss of annual mean rainfall, respectively, in the drier and wetter parts of the country. e model predicted a 0.65 ratio of annual transpiration to annual evapotranspiration.
Unknown
Evaluation of Future Streamflow in the Upper Part of the Nilwala River Basin (Sri Lanka) under Climate Change
(MDPI, 2022-03-16) Chathuranika, I. M; Gunathilake, M. B; Azamathulla, H. M; Rathnayake, U
Climate change is a serious and complex crisis that impacts humankind in different ways. It affects the availability of water resources, especially in the tropical regions of South Asia to a greater extent. However, the impact of climate change on water resources in Sri Lanka has been the least explored. Noteworthy, this is the first study in Sri Lanka that attempts to evaluate the impact of climate change in streamflow in a watershed located in the southern coastal belt of the island. The objective of this paper is to evaluate the climate change impact on streamflow of the Upper Nilwala River Basin (UNRB), Sri Lanka. In this study, the bias-corrected rainfall data from three Regional Climate Models (RCMs) under two Representative Concentration Pathways (RCPs): RCP4.5 and RCP8.5 were fed into the Hydrologic Engineering Center-Hydrologic Modeling System (HEC-HMS) model to obtain future streamflow. Bias correction of future rainfall data in the Nilwala River Basin (NRB) was conducted using the Linear Scaling Method (LSM). Future precipitation was projected under three timelines: 2020s (2021–2047), 2050s (2048–2073), and 2080s (2074–2099) and was compared against the baseline period from 1980 to 2020. The ensemble mean annual precipitation in the NRB is expected to rise by 3.63%, 16.49%, and 12.82% under the RCP 4.5 emission scenario during the 2020s, 2050s, and 2080s, and 4.26%, 8.94%, and 18.04% under RCP 8.5 emission scenario during 2020s, 2050s and 2080s, respectively. The future annual streamflow of the UNRB is projected to increase by 59.30% and 65.79% under the ensemble RCP4.5 and RCP8.5 climate scenarios, respectively, when compared to the baseline scenario. In addition, the seasonal flows are also expected to increase for both RCPs for all seasons with an exception during the southwest monsoon season in the 2015–2042 period under the RCP4.5 emission scenario. In general, the results of the present study demonstrate that climate and streamflow of the NRB are expected to experience changes when compared to current climatic conditions. The results of the present study will be of major importance for river basin planners and government agencies to develop sustainable water management strategies and adaptation options to offset the negative impacts of future changes in climate.