Master of Philosophy (M.Phil.)
Permanent URI for this collectionhttp://192.248.9.226/handle/123/18723
Browse
Browsing Master of Philosophy (M.Phil.) by Author "Kumarawadu, S"
Now showing 1 - 3 of 3
- Results Per Page
- Sort Options
- item: Thesis-Full-textObstacle avoidance for unmanned surface vehicles: simulations and experimentsJayendra, RG; Kumarawadu, SSri Lanka ports authority and many other organizations are increasingly interested in the use of Unmanned Surface Vehicles (USV) for harbor security and surveillance applications. USVs can be used to collect information, samples and perform experiments inside a harbor or outside by. Navigating through ships and other objects. This research study is focused on finding algorithms for obstacle avoidance (OA) of USVs. The initial paradigm that is used to establish the solution was the OA of Unmanned Ground Vehicles (UGV). The algorithms developed for UGV were implemented practically with the limitations of hardware. Then, effort is taken to apply those algorithms to the surface vehicles with some modifications. In this study, a novel OA algorithm is proposed for static obstacles based on the Morphin algorithm. This proposed algorithm and the previous algorithm which is developed based on ground vehicles are compared with the potential field method. Static OA without dynamic OA is not helpful for unmanned vehicles on sea. A lot of researches have been carried out to avoid dynamic objects, but have failed to find an optimum solution although comparatively good approaches have been presented. Intelligent techniques have been rarely applied for dynamic obstacle avoidance. In this research, the effectiveness of applying intelligent or mathematical techniques for path prediction of dynamic obstacles is discussed with simulations to pick the best for a given situation. Then a noval projected dynamic obstacle area method is presented to avoid dynamic obstacles effectively. Comparative results are presented at the end to prove the strength "of the noval dynamic obstacle area method.
- item: Thesis-Full-textA Robust natural language question answering system for customer helpdesk applications(2014-05-30) Samarakoon, LT; Kumarawadu, SThis thesis describes a restricted-domain question answering system which can be used in automating a customer helpdesk of a commercial organization. Even though there has been an increasing interest in data-driven methods over the past decade to achieve more natural human-machine interactions, such methods require a large amount of manually labeled representative data on how user converses with a machine. However, this is a requirement that is difficult to be satisfied in the early phase of system development. In addition, the systems should be maintainable by a domain expert who is less technically skilled when compared to a computer engineer. The knowledge based approach that is presented here is aimed at maximally making use of the user experience available with the customer service representatives (CSRs) in the organization and presents how true representative data can be collected. The approach takes into account the syntactic, lexical, and morphological variations, as well as a way of synonym transduction that is allowed to vary over the system's knowledge base. The query understanding method, which is based on a statistical classifier, a ranking algorithm based on Vector Space Model (VSM) and a pattern writing process, takes into account the intent, context, and content components of natural language meaning as well as the word order. A genetic algorithm-based method is presented for finding the domain specific ranking parameters. An evaluation of the approach is presented by deploying a system in a real-world enterprise helpdesk environment in the telecommunication domain. The evaluation shows that the system is able to answer user questions with an accuracy of 94.4%. Furthermore, maintenance of the deployed system is carried out by CSRs successfully.
- item: Thesis-Full-textUnder frequency load shedding for power systems with high variability and uncertaintyBambaravanage, DYT; Kumarawadu, S; Rodrigo, AEmergency load shedding for preventing frequency degradation is an established practice all over the world. The objective of load shedding is to balance load and generation of a particular Power System. In addition to the hydro and thermal generators each with less than 100 MW, today, the Power System of Sri Lanka is comprised of three coal Power Plants: each has a generation capacity of 300 MW, Yugadanavi combined cycle Power Plant (300 MW generation capacity) and a considerably extended transmission network. To cater consumers with high quality electricity, a reliable Power System is a must. Therefore, it has become timely necessity to review the performance of the present CEB Load Shedding Scheme and suggest amendments where necessary. The objective of this research is to explore a better Under Frequency Load Shedding Scheme (UFLSS) which can face probable contingencies and maintain stability of the system while catering more consumers. The suggested UFLSSs can address the recent changes taken place in the Sri Lanka Power System too. A simulation of the Power system of Sri Lanka was designed with software PSCAD. Its validity was checked through implementing actual scenarios which took place in the power system under approximately-equal loaded condition and by comparing the simulated results and actual results. Then a performance analysis was done for the Ceylon Electricity Board (CEB) Under Frequency Load Shedding Scheme which is being implemented in Sri Lanka. Having identified the drawbacks of the CEB Under Frequency Load Shedding Scheme, the new UFLSSs (LSS-I and LSS-II) were suggested. The Load Shedding Scheme – I (LSS-I) is designed based on power system frequency and its derivative under abnormal conditions. Without doing much modification to the prevailing UFLSS, and utilizing the available resources, the suggested LSS-I can be implemented. The LSS-II gives priority for 40% of the system load for continuous power supply, and it is comprised of two stages. During the stage-I, approximately 30% of the load is involved with the Load Shedding action. During the stage-II, the disintegration of the power system is done. This involves the balance 30% of the load. At 48.6 Hz the disintegration of the power system takes place. By disintegrating the power system at the above mentioned frequency, all islands as well as the national grid can be brought to steady state condition without violating the stability constraints of the Sri Lanka power system. During disintegration of the Power System, special attention must be paid for: • Generation & load balance in each island and in the national grid. • Reactive power compensation in islands and in the national grid. • Tripping off of all isolated transmission lines (which are not connected to effective loads). Through simulations the effectiveness of the UFLSSs were evaluated. They demonstrate better performance compared to that of the currently implementing CEB scheme. Results highlight that the UFLSS should exclusively be specific for a particular Power System. It depends on factorssuch as power system practice, power system regulations, largest generator capacity, electricity consumption pattern etc.