Browsing by Author "Thiruchittampalam, S"

Now showing 1 - 7 of 7

item: Article-Full-text
Developing a reliable monitoring framework to detect environmental changes near mining areas : a remote sensing approach
(2011) Thiruchittampalam, S; Jayawardena, CL
Agriculture and mining are two key industries that govern the growth of the socio-economic systems of a country. Detrimental effects on the socio-environment during mining activities are unavoidable. Hence, a proper balance between social, environmental, and economic aspects needs to be promoted in mining industries. Developing robust real-time monitoring systems will prevent overexploitation of resources, negative impact on the socio-environmental system, illegal mining activities, and abandoned mines without proper reclamation. This study aims to recommend utilising no-cost satellite data to assess effects on the environment due to mining. Further, the machine learning approach in this study is expected to lead to an autonomous realtime monitoring mechanism. Based on accessibility and stakeholder interest, a cluster of abandoned and operating quarry sites in the Anuradhapura District, Sri Lanka, was selected as the study area. Freely available high spatial resolution satellite images were obtained from Google Earth and Copernicus Open Access Hub (Sentinel 2 satellite data). Imageries acquired were subjected to object-based land cover classification. Machine learning algorithms, namely, Decision Tree, Random Forest, and Support Vector Machine, were used in the classification process. The best performing algorithm in land cover classification was chosen for multitemporal analysis of the area of interest. In addition, false colour composites and spectral indices were generated using Sentinel 2 images to differentiate human-induced negative impacts and natural changes. The results show that Support Vector Machine outperformed other algorithms in classifying land cover near mining areas. Further, multitemporal analysis of land cover changes using this algorithm implies autonomous monitoring using satellite data was viable. Additionally, auxiliary information such as false colour composites and spectral indices confirmed that the increased proportion of water bodies in the area was due to leaving the abandoned mines without proper rehabilitation. This study provides evidence that the fusion of machine learning based classification with spectral indices helps develop robust monitoring systems.
item: Conference-Full-text
Evaluation of ventilation network through hybrid analytical-numerical approach in underground working block
(Department of Earth Resources Engineering, University of Moratuwa, 2021-12) Thanujan, T; Brinthan, K; Thiruchittampalam, S; Jayawardena, CL; Dissanayake, DMDOK; Jayawardena, CL
The mine environment is complex and highly dynamic due to the developments over time and surrounding climatic changes. Heedlessness to supply adequate quantity and quality of air will catalyse short and long-term ailments to the workers. Therefore, this study emerges as the new research frontier in incorporating software-assisted numerical simulation with analytical computations. This investigation assesses the existing ventilation parameters at the Bogala underground graphite mine for the propriety of the working environment. The uttermost bottom block between 240 and 275 fathoms (FM) levels was examined. The parameters were obtained through the in-field ventilation survey. Measured air quantity, psychometry, and air quality values were analysed and fed to the computer-simulated model. Moreover, the reentry time for a development drive at 275 FM level was estimated using the throwback method. Adequacy assessment unveils that all the parameters besides air quality are inadequate at most stations for optimal mine conditions to attain maximum efficiency. Furthermore, the re-entry time after the development blast at the selected drive is meager and necessitates re-calculation for each blast. Moreover, stale air mixing and air recirculation are extant at 240 FM and 275 FM levels, respectively. Thus, mine ventilation at Bogala needs to be optimised, admitting workers' health, safety and comfort, and productivity of the mine.
item: Conference-Full-text
Noise and vibration control in crusher plant activities to enhance health and safety of workers
(Department of Earth Resources Engineering, 2018-08) Thiruchittampalam, S; Kinoj, A; Ekanayake, EMCK; Vithurshan, S; Hemalal, PVA; Samaradivakara, GVI; Rohitha, LPS; Chaminda, SP; Abeysinghe, AMKB; Samaradivakara, GVI
Fulfillment of human needs necessitates activities that have their positive and negative aspects. Crusher plant operation is not an exception. One of the major concerns related to these activities is its impact on worker health and safety. Although noise and vibration related legislation focus on reduction at the source, provision of personal protective equipment is considered as an effective method of control. The focus of this study was to look for control strategy for noise and foot transmitted vibration hazards related to fixed machineries in crusher plants. Six crusher plant sites were selected based on their varying production capacity for the study. Accordingly, selected sites were visited to carry out area noise and wholebody vibration surveys. Plant arrangement, details of crusher units, prevailing meteorological conditions and ground conditions were also recorded. Daily exposure levels were determined and sound and whole-body vibration contour maps were developed based on Control of Noise and Vibration at Work Regulations Act 2005. Relationship of areas of hazardous zones w i t h production capacity was determined using statistical tools and the level of worker awareness was studied through one to one interviews. Cost effective controls are recommended by studying the best practices.
item: Conference-Full-text
A proximity based rehabilitation approach for abandoned quarries in rural Sri Lanka
(IEEE, 2018-05) Jayawardena, CL; Thiruchittampalam, S; Dassanayake, ABN; Abeysinghe, AMKB; Wimalarathna, W; Chathuranga, D
Site closure and rehabilitation is seldom practiced in Sri Lankan quarrying industry. Hence, open craters filled with water and garbage, having steep unstable slopes represent most of the abandoned quarry sites, raising public health and safety concerns additionally to unpleasant environments. Traditional rehabilitation methods are rarely applicable for such circumstances due to unique socio-economic conditions and stakeholder aspirations. This study is an assessment on the feasibility of outcrop excavations adjoining the abandoned and isolated quarry sites, to establish relatively flat ground conditions, in rural Sri Lanka. The results reveal possibilities of establishing mega-quarry sites on selected locations in Anuradhapura District to produce large quantities of aggregates and useful land masses for future development while ensuring safe environment for the local communities.
item: Conference-Full-text
Smartphone vs. consumer-grade gnss for field studies: a statistical comparison
(IEEE, 2022-07) Jayawardena, C; Jayasundara, R; Thiruchittampalam, S; Thanujan, T; Rathnayake, M; Adhikariwatte, V; Hemachandra, K
Smartphones have replaced not only the telephone but also most other devices such as camera, computer, torch, Global Navigation Satellite System (GNSS) devices etc., used in our day-to-day activities due to their multifunctionality. However, the accuracy of the inbuilt GNSS receivers on smartphones can be ambiguous and has concerns when used in field investigations due to internal and external factors. Despite the fact that these factors are well known, the degree of influence of these factors on the accuracy of components of location data provided by smartphones is unexplored. This study aims to statistically assess the influence of three factors (time of the day, landscape characteristics, and ability to track signals of multiple GNSS constellations) on the accuracy of spatial data provided by smartphone GNSS. The horizontal and vertical accuracies of smartphones are within 40 m and 15 m, respectively. The results reveal that the ability to track multi-constellation and landscape properties, respectively, has a significant influence on the horizontal and vertical accuracies of smartphone location information. Further, this study also provides insights to improve the reliability of spatial data collection using smartphones in outdoor environments.
item: Conference-Full-text
Spatio-Temporal analysis with machine learning for sustainable management of abandoned quarries
(Division of Sustainable Resources Engineering, Hokkaido University, Japan, 2024) Gouthaman, V; Jayakody, JANS; Jayasinghe, JASHR; Thiruchittampalam, S; Jayawardena, CL; Iresha, H; Elakneswaran, Y; Dassanayake, A; Jayawardena, C
The abandoned quarries demand not only appropriate rehabilitation but also continuous monitoring. If these sites are left unmanaged, they can create significant environmental and ecological impact due to changes in land-use and land-cover. Monitoring of abandoned quarries are often overlooked due to challenges such as accessibility, safety and costs involved. Hence, there hardly exists a systematic monitoring approach or appropriate guidelines for managing quarry sites upon termination of the extraction activities. Thereby the presence of hazardous environments will be unavoidable with a substantial resistance on the quarry industry for not so sustainable closure procedures and following up actions. To overcome these challenges, sufficient monitoring of quarry sites and surroundings to enforce appropriate rehabilitation strategies with post monitoring that has minimal on-site involvements would be essential. For such purposes, analysing remotely sensed data would be applicable, based on the quality of data collection, processing and sufficient ground truthing. Accordingly, this study aims to develop an automated classification approach for mapping the land cover in the regions of abandoned quarries. It employs a comprehensive methodology that includes data preparation, feature extraction and selection, hyperparameter optimization, and identification of the algorithm that exhibits better accuracy. The efficacy of machine learning models - decision tree (DT), random forest (RF), and support vector machine (SVM) - were critical to analyse Landsat 8 and Sentinel 2 satellite images at selected sites in Anuradhapura, Sri Lanka. The outcome reveals that the SVM model produced the highest accuracy of 91.30% with a kappa index of 0.898. This superior performance of Sentinel 2 images could be attributed to their higher spatial resolution compared to Landsat 8 and SVM’s efficient handling of high-dimensional data. Furthermore, SVM’s robustness against overfitting using regularization, and its flexibility in dealing with complex separations through kernel functions would have facilitated the computations in addition to textural features and spectral indices incorporated to augment the model training procedure. It was also evident that augmenting the number of features can help alleviate the misclassifications that occur when exclusive use of spectral data. Utilizing the developed machine learning algorithms, a temporal analysis was performed on land cover from 2018 to 2022 to obtain a comprehensive overview on the changes. This analysis underscores the potential for monitoring land cover changes in abandoned quarries for effective management and rehabilitation strategies with minimal human intervention.
item: Conference-Abstract
A statistical appraisal on accuracy of smartphone location services: a case study at the University of Moratuwa, Sri Lanka
Jayawardena, CL; Jayasundara, DRT; Thiruchittampalam, S; Jayakodi, JDSU; Senanayake, IP
Smartphones have become an essential companion in most of the communities. Yet, we may not be quite aware of the capabilities and services that these devices could provide. As a result, features such as location services are underutilized and mostly used for navigation and location sharing. This study explores the limitations of embedded GPS receivers in smartphones with reference to the performance of a consumer-grade hand-held GPS device. The location coordinates obtained with the GPS unit and six smartphones on five locations over ten consecutive days revealed that over 70% of smartphone records provide the location coordinates within 0 to 10 m accuracy. Furthermore, at certain locations, over 75% of records have maintained the coordinate accuracy within 0 to 5 m. Hence, the use of smartphone location information in place of standalone GPS readings, can be recommended for moderate location accuracy requirements, such as geo-tagged data collection. Nevertheless, hand-held GPS units provide better approximations than the smartphones, for elevation readings at the studied locations. Accordingly, further investigations are recommended, to evaluate the discrepancies in elevation records, provided the ambiguities generated while recording the elevation measurements from the hand-held GPS units are minimized.