13474 research outputs found
Sort by
A decomposed streamflow non-gradientbased artificial intelligence forecasting algorithm with factoring in aleatoric and epistemic variables / Wei Yaxing
No full textThe dissertation aims to develop an effectively decomposed time-series nongradient- based artificial intelligence model for forecasting a time-series regression machine learning task. Real-world optimizations, such as forecasting streamflow, are a complicated process that is highly non-linear and multi-modal, demanding the use of a suitable modeling tool, with an emphasis on artificial intelligence algorithms, to get befitting forecast results. Although artificial intelligence algorithms are of primary importance in today's media information and technology, with substantial research in streamflow, handling the limitations must be done appropriately in real-world optimization. The deterministic approach, which utilizes the gradient information in the search process, is prone to trapping at local minima when dealing with complex streamflow forecasting, primarily due to saddle points and local minima in the nonconvex objective function in an artificial neural network. Consequently, the study involved exploiting optimization techniques to enhance the training artificial intelligence algorithm for streamflow forecasting from a gradient-based to a stochastic population-based approach in several aspects, including solution quality, computational effort, and parameter sensitivity on streanflow in Johor, Malaysia. Empirical studies of metaheuristic algorithms performance demonstrated that the hybrid metaheuristic algorithms-artificial neural network outperformed the gradient-based artificial neural network (RMSE=113.92 m3/s) for streamflow forecasting, notably with the firefly approach, with an average RMSE=96.06 m3/s. However, not all accepted metaheuristic algorithms are compatible with enhancing the ANN for streamflow forecasting, demanding a thorough analysis due to performance differences across cases. While the firefly algorithm solution is superior, it has a higher time complexity compared to other algorithms used when there are more hidden layers and neurons. The firefly algorithm remains a feasible alternative for shallow architectural network models, while metaheuristic algorithms such as the Particle swarm algorithm and Bat algorithm are better options for deeper architectural network models. To summarise, metaheuristic algorithms can give a superior optimization approach than the traditional artificial neural network method, providing the computing time is within an acceptable range. Besides, the wavelet transform, which decomposed the original series into unsophisticated sub-series, requires assessing the additional algorithm's capacity, which is the parameter setting sensitivity to perform well in various contexts while tackling a specific problem. Given the multitude of components to manage, streamflow forecasting is preferable to employ an algorithm with low sensitivity to parameter variations. Firefly algorithm outperformed the other metaheuristic algorithms used to solve this proposed hybrid artificial intelligence model regarding parameter sensitivity. A comparison of deep learning convolutional neural network and artificial neural network algorithms was also performed, with findings revealing that convoluted input formation was less stochastic than feedforward formation, particularly for a more complicated series and vice versa, due to its capacity to attract features. The contradictory behavior could be due to the feature attraction ability of convolution neural network to over-fit the simpler sub-series. Finally, one key drawback of estimating streamflow outlined above is that it does not account for variability. As illustrated by the supplementary projected standard deviation and mean variables, generating the forecast in the Aleatoric and Epistemic forms might help improve one's impression of the predicted results
Cross-silo implementation of federated learning in healthcare for 6G: Leveraging VPN-based wireless backhaul networks / Atif Mahmood
No full textThe widespread use of smart devices and the Internet of Everything (IoE) strains current wireless systems. Sixth-generation (6G) research is underway, with Artificial Intelligence (AI) playing a picotal. Terahertz (THz) communication spectrum and Federated Learning (FL) are gaining traction in 6G as advanced AI. FL, a decentralized approach, faces challenges in resource management, statistical and system heterogeneity, and security. Communication bottlenecks occur during FL, and efficient resource management is crucial for faster global convergence. Recent research explores methods like local updates, compression, and decentralized training. Privacy concerns arise, but sharing model updates may expose sensitive information. Solutions address challenges like stragglers and fault tolerance, which are critical for 6G. The Terahertz spectrum, vital for 6G, offers wide bandwidth and potential commercial use, especially in high-speed data and video transmission. Despite limitations, Terahertz technology maximizes spectrum utilization and enhances transmission security. The study addresses the challenges of federated learning (FL) in three phases. Firstly, it provides a concise overview of the THz spectrum in fixed wireless communication, highlighting its suitability for fixed wireless links within 1 km. The study examines THz properties, estimates data rates, and suggests its use to enhance Federated Learning (FL) communication. In the healthcare sector, FL relies on THz-based wireless backhaul with a VPN for hospitals, laying the groundwork for THz utilization in 6G wireless backhaul. This introduces an innovative network architecture for secure cross-silo FL, focusing on healthcare enhancement. In the second phase, data from a new HFMD biosensor is classified using centralized machine learning to analyze the HFMD dataset and create a benchmark. Initial experiments with Support Vector Machine (SVM) yielded a 72% accuracy rate, and a neural network achieved 80%. Federated learning with four clients reached a maximum accuracy of 91%, addressing healthcare data security concerns. The third phase focuses on improving dataset convergence and distribution, introducing an efficient network architecture for
Federated Learning in the subsequent section on the wireless backhaul-based federated network for healthcare dataset experiments. Here, an efficient network for healthcare records analysis is introduced. The framework is built upon VPN technology, and a cross-silo-based federated learning approach is applied over the wireless backhaul network. A comparison is made between the Terahertz-based wireless backhaul channel bandwidth, the E/V band (mmWave), and the microwave. A significant enhancement in convergence time is observed when utilizing the higher bandwidth wireless channel. In summary, this research indicates that optimal accuracy is attained when utilizing FedAvg and THz as the communication channel, with convergence times ranging from 55 seconds to 24 seconds for FedAvg and transitioning from THz to downgraded MW. This highlights the pivotal significance of a higher bandwidth communication link. Relying solely on the dataset for determining device rules is not advisable. A three-step strategy is employed to bolster security, encompassing a private network within the telecom network, a private network with licensed frequency channels, and a private network with a licensed band. This security is further strengthened through VPN-based measures
Enhancing power distribution system resilience against urban flash floods using hierarchical combination of Monte Carlo technique and reinforcement learning / Suhail Afzal
No full textPower system is recognized as one of the lifeline systems of a community that is appreciably reliable, but vulnerable to natural hazards such as hurricanes, winter storms, and floods. In recent decades, urban flash floods have become more common because of climate change and are worsening with the intensification of short-duration rainfall extremes. This catastrophic natural hazard presents a significant threat to a power distribution system, thus enhancing system resilience against flash floods is essential and imperative. To achieve this goal, existing literature offers a wealth of knowledge, however in these works, two-dimensional (2D) surface flow models are used to solve the hydraulics. These 2D hydrodynamic models consist of a set of nonlinear differential equations and require high-resolution topographic data of the floodplain. Though these models can provide descriptions of overland flow propagation, they fail to provide overflow locations which are crucial in flash flood modelling. Additionally, these 2D flood models are computationally expensive, hence cannot be run in real-time. Furthermore, the flooding fragility model offered by the Federal Emergency Management Agency (FEMA) of the United States is adapted to formulate failure scenarios, that is inappropriate for the Malaysian distribution network. Moreover, researchers have proposed various service restoration models and techniques prioritizing critical load advocating the resilient operational prowess of diverse sets of distributed generators (DGs). However, mostly dispatchable DGs are modelled, and the time-based model has not been extensively taken into consideration. In addition to this, varying load profile, temporal fault incursion and DG profiles are also not investigated. Therefore, this study presents a probabilistic flood model that is easy to develop and can handle heavy uncertainties related to urban flash flooding. In this respect, a linear regression model is proposed to estimate flood elevation in an urban floodplain and the Monte Carlo technique is employed to predict overflow locations in a grid-based environment. Considering rainfall intensity, soil moisture, and curvature of the surface, reinforcement learning is then leveraged to trace the flow path of floodwater from these overflow locations, to identify distribution substations at the risk of inundation. The proposed flood model is applied to IEEE 33-bus and a real 23-bus distribution systems considering a hypothetical terrain and validated on a real urban area. Further to this, site surveys and historical data are used to develop flooding fragility curves for indoor electrical substations to determine their probability of failure. Finally, the spatiotemporal impact of flash flooding on a modified IEEE 33-bus test system is captured using the proposed flood model. The evolving substation faults are then included in the proposed resilience-oriented time horizon-based service restoration model that also considers dynamic load demand, heavy uncertainties related to renewable generation, and dependencies of a distribution network in and outside the power system. This work will assist decision-makers and utility operators in enhancing power system resiliency against urban flash floods while overcoming the barriers of limited data and time
Techniques in the treatment of sense in the translation of English and Chinese hymns / Rebecca Mccort
No full textThis study aims to identify and compare the use of translation techniques from existing taxonomies that prevail in the treatment of sense in English-Chinese (En-Ch) and Chinese-English (Ch-En) hymn translation, and to descriptively identify any additional novel translation techniques that prevail in the selected hymn set. Twenty pairs of ST and TT hymns were selected from The New Hymnal: English-Chinese bilingual edition (1999). Of these, ten pairs were En-Ch translations, and ten were Ch-En. Source text (ST) and target text (TT) hymn lyrics were analyzed to identify use of Low’s (2017) song lyric translation techniques, as well as techniques for written translation proposed by Vinay & Darbelnet (1995), Baker (1992), and Newmark (1988). Novel translation techniques not included in any of these taxonomies were identified through descriptive analysis. Results revealed that Dilution, Modulation, Near-synonym, Compensation, and Transposition are the five most commonly used translation techniques for En-Ch translations, whereas Dilution, Near-synonym, Changing the kind of utterance, Modulation, and Compensation are the five most used techniques for Ch-En translations. Novel techniques identified in En-Ch translations include Partial transfer, More descriptive term, Change to figure of speech, Intra-line position change, and Pronoun change. Novel techniques in Ch-En translations include Partial transfer, Intra-line position change, Change to figure of speech, Pronoun change, Statement to address, and Phrase to sentence. Though failing to qualify as translation techniques, Addition and Replacement also play a significant role in both directions of translation. Findings emphasize the suitability of Low’s (2017) techniques for song translation and suggest which techniques from this song-translation-specific taxonomy are most useful in producing successful song translations. The research also highlights the range of techniques translators draw on for flexible handling of semantics, revealing the intentional and artful nature of song translation
Lexical borrowing in English newspapers in Bangladesh / Md. Nurul Islam
No full textBangladesh has experienced innumerable language contact situations before and after its independence. There have been many instances of merging Urdu, Perso-Arabic, and Hindi (or Sanskrit) words with the Bengali language. As a result, articles and other write-ups in the Bangladeshi English newspapers contain many common Bengali loanwords. The aim of this study is to examine lexical innovation in Bangladeshi newspapers and to determine their functions in relation to the socio-cultural and political contexts in which they are found. The data were collected from three national dailies, namely, The Daily Star, The Independent and The News Today. This study strove to create a corpus of contact expressions in these newspapers, delineate the types of Bengali and other loanwords used in the corpus and the reasons for their use, and explore the results in relation to the language ecology. The features were analysed and interpreted using ‘a combination of two approaches’ – the World Englishes frameworks of Kachru (1988) and Strevens (1977). Hence, this study can be considered a qualitative study carried out through textual analysis. The analysis of data shows that lexical borrowing and hybridisation are the two features found in the English dailies in Bangladesh contributing to the highest number of loanwords in English and are often used by the people in different domains within Bangladesh. Moreover, these features are used to provide for ideas and concepts when there is no equivalent in the receiving language. Lexical borrowing is found in only two sub-categories – independent and reduplicating loans, each providing a variety of terms that reflect the notion of Bengaliness. The vocabularies found in different domains attest to their Bengali socio-cultural and political elements, such as references to religion, animals, festivals, traditional sports, music, culinary, clothing and so forth. Furthermore, the two sub-categories – independent and reduplicating lexical borrowings – contribute to the existence of multilingualism in Bengali-English words because they are loanwords from Bengali, Sanskrit/Hindi, Urdu and Arabic/Perso-Arabic languages. Two types of hybrid words – open and closed set – result in the very unique phenomenon of English-Bengali and Bengali-English loan blend. Furthermore, coinages are also commonly found in the Bangladeshi English newspapers and form part of the Bengali-English lexicon. In addition, English words are created with slight semantic modification. Semantic shift and blending provide the least number of lexical items in the data. The selected three English newspapers of Bangladesh use a number of acronyms in English which have unique features and contribute to a high number of loanwords in English. Overall, the seven features of language contact which have contributed to 618 loanwords are used in the English newspapers in Bangladesh. Among 618 loanwords, 25 words are codified as English words of Bengali origin by the latest edition of the Oxford English Dictionary. The study concludes that the Bangladeshi English newspapers exhibit a novelty at the lexical level due to the strong influence of the Bengali language and that the use of loanwords is comparable to that available in other varieties of the English language
Techno-economic assessment and environmental analysis of bio-compressed natural gas production from palm oil mill effluent / Nasrin Abu Bakar
No full textThere were 451 palm oil mills in Malaysia and these mills generated about 60 million m3 of palm oil mill effluent (POME) in the year 2021. Due to its high organic content, POME needs to be treated before being discharged to the water bodies within regulatory discharge limits. Conventionally, anaerobic digestion method via open ponding or tank systems are used for this purpose where the treatment efficiency is low and the biogas produced is not recovered. In the last two decades, capturing biogas using closed anaerobic digester are becoming acceptable practice as an integrated treatment of POME and biogas capture as a mean to reduce greenhouse gases (GHG). The captured biogas is typically used for heat and power generation. In recent time, upgrading biogas for biomethane or bio-compressed natural gas (Bio-CNG) production has emerged as an alternative to biogas utilisation in Malaysia. However, a detailed technical, economic and environmental assessment for commercial Bio-CNG production, is yet to be established in Malaysia, which forms the justification of this work. Potential biogas volume, installed electricity capacity and Bio-CNG production from entire palm oil mills (451 mills) in Malaysia in 2021 were approximately 1648 million m3, 508 MW and 988 million m3 Bio-CNG, respectively. A total of 135 mills were installed with biogas plant in 2021 and therefore only 33% of the full energy potential was realized. In terms of utilization, 87 mills utilise the biogas for electricity generation, 15 mills for steam or combined heat and power generation, and only a single mill for Bio-CNG production. However, 32 mills were just flaring the biogas generated without energy recovery and used it for the purpose of methane (CH4) emission mitigation strategy. As a proof of concept, a 400 m3/hr Bio-CNG plant was developed and evaluated in a palm oil mill located at Kuala Kubu Bahru, Selangor. The Bio-CNG production process which was based on combined biological and physical methods, and membrane technology achieved hydrogen sulphide (H2S) and carbon dioxide (CO2) removal efficiencies of 99 and 92.2%, respectively. The produced Bio-CNG was found to contain about 94 vol.%, 3 vol.%, 0.5 vol.% and 3 vol.% of CH4, CO2, O2 and N2 respectively with H2S at a trace level of 3 ppm, resulting in significant increase of calorific value from 20 MJ/m3 to 35 MJ/m3. These properties are also comparable to natural gas quality. The economic analysis conducted for Bio-CNG plant integrated with existing biogas plant indicated that an approximate 14% internal rate of return with a payback period about 6 years for a mill with 54 tonnes per hour capacity. The Life Cycle Impact Assessment carried out showed that the environmental impacts of the Bio-CNG production were global warming, fine particulate matters formation, fossil and mineral resources scarcity. These were due to the plants’ heavy dependence on the grid-connected electricity. In conclusion, Bio-CNG is technically, economically, and environmentally viable business alternative to biogas offsite utilisation
Solid-state microwave disruption technique for the extraction of bioactive compounds from Orthosiphon stamineus (Java tea) / See Tiam You
No full textNatural products that can be extracted from herbal plant leaves with pharmaceutical effect tend to be less harmful and have less or no side effect as compared to synthetic drugs. Extraction of herbal plant leaves involved lengthy drying time and is energy intensive. Hence, fresh and dried Orthosiphon stamineus (Java tea) leaves which contains medicinal values was processed into herbal tea to reduce the processing time and energy consumption while preserving its bioactive compounds over a prolong storage period. In this work, a solvent free biomass disruption process using microwave which is named as solid-state microwave disruption (SSMD) was established. Microwave process not only can achieve higher drying efficiency but also can disrupt cell plant for better elution process of bioactive compounds. A simple elution process was employed to elute the bioactive compounds and the total amount of bioactive compound of eupatorin, 3’- hydroxy-5,6,7,4’-tetramethoxyflavone and sinensetin obtained was used to illustrate the performance of SSMD. The microwave method was optimized and modelled using modified energy parameters method of absorbed energy density (AED) and absorbed power density (APD). The optimum extraction yield of 6.61 mg/g obtained by SSMD is comparable with those obtained by ultrasonic assisted extracion (UAE), microwaveassisted extraction (MAE) and Soxhlet extraction (SE) at 6.47 mg/g, 6.59 mg/g and 6.90 mg/g, respectively. It was found that for both fresh and dried Java tea leaves, the irradiated microwave flux was able to penetrate the biomass’s plant cell wall and interact with the intracellular moisture within the plant cell. As the moisture reacted with microwave, more plant cell disrupted with reduced moisture content. The microwave method was optimized and modelled. The model gives an accurate prediction and successfully optimized the degree of disruption. The extraction yield of SSMD-elution for both fresh and dried samples was more than 95% of SE yield. However, the fresh sample required shorter overall heating time as compared to that of the dried sample. Moreover, SSMD shortened the traditional long drying process which normally required more than 24 hours. In addition, the SSMD process was scaled-up using the energy-based parameter for both fresh and dried sample. The large scale process (5 times larger) shows that the model has good predictability and the optimum condition obtained can be scaled up. The scale up process further highlighted the feasibility of SSMD such as time saving in overall drying, disruption and extraction process. On the other hand, the simple SSMD process eases the selection of solvent, space saving and lower solvent consumption. Also, the simple optimization and modelling process using energy based method confirm the application of SSMD in bioactive compound extraction and also for herbal tea processing. The solvent free disruption and extraction process is established via modelling and optimization study which would contribute greatly to the development of pharmaceutical and food processing industrial
Motivation factors affecting project manager’s job performance in Malaysia and Singapore / Ng Zhe Xhiang
No full textThe construction industry significantly contributes to a country's economic sector, relying heavily on the job performance of its project members. However, the construction industry in Malaysia and Singapore still faces room for improvement in job performance. Researchers have identified failures in achieving time, cost, and quality goals in numerous construction projects in these countries are mainly due to lack of motivation practice in the industry. Thus, the aim of this study is to explore suitable motivation practices that can enhance the job performance of project managers (PMs) in Malaysia and Singapore. A literature review was conducted by studying previous research articles, journals, and books concerning motivation and job performance. This study adopted a quantitative approach, sending out 256 sets of questionnaire surveys to project managers in Malaysia and Singapore through email. However, only 33.6% received responses. Collected data was analysis using mean analysis to determine the motivation factors affecting job performance, strategies to improve job performance, and the performance of PMs' roles. High salary levels, fringe benefits, and job safety were ranked as the top three motivation factors. Next, the Mann-Whitney U test was used to determine significant differences in agreement levels on motivation factors and job performance. Subsequently, Pearson’s correlation test was conducted to identify the correlation between motivation factors and job performance. It was found that most of the motivation factors has positive correlation with the role of project manager. This research is beneficial to PMs and other construction stakeholders in Malaysia and Singapore. By instilling motivation in employees, it is possible to improve the job performance of individuals
Investigation of muscle strength in long covid patients based on mechanomyography / Harinivas Rao Suba Rao
No full textMechanomyography (MMG) is an alternate tool that can noninvasively test muscular strength. The main focus of this study is the development of MMG for assessing muscle strength in patients with long-term conditions. Long COVID is a post-COVID syndrome that affects those who have had COVID-19 symptoms for at least three to six months. The physiological consequences of age, gender, and level of physical exercise are very important in defining an individual's muscle strength, especially for post-COVID-19 individuals. Additionally, MMG has additional potential advantages because it gives a greater signal-to-noise ratio and is less sensitive to the location of the sensor on the muscle. The MMG prototype can measure muscle strength potential by obtaining an MMG signal from the surface of an identified muscle from time to time with the aid of a smartphone application, comparing post-COVID-19 with non-covid-healthy individuals. Twenty healthy and post-COVID-19 category 1 and 2 participants respectively volunteered in the study. All of them performed knee extensions and half-squats as the baseline measures, followed by a series of loaded (2kg ankle load during knee extension and 10kg body load during half-squats) in each set to induce muscle fatigue, with an MMG sensor attached externally over their quadriceps muscle. All participants who engaged in more vigorous contractile activity experienced higher muscle fatigue, portrayed by a reduction in MMG values throughout the session, with post-COVID-19 subjects reporting lower MMG values than healthy subjects (p=0.034) [post-COVID-19 MMG value:(0.80±0.18) g; healthy non-covid MMG value:(0.83±0.14) g, p=0.034]. The MMG signal from axis Y is reliable for measuring muscle strength potential, which oscillates laterally and is shown to be more sensitive than the axes X and Z. In addition, to support this research, a thigh angle was calculated, and this was portrayed by a reduction in the triaxial ADXL345 signal throughout the session, and a greater deviation in thigh angle values throughout the session, with post-COVID-19 subjects reporting greater thigh angle difference compared to healthy subjects
A framework for walkable neighbourhood in Depok, Sleman Regency, Yogyakarta / Linda Hijriyah
No full textThe world population and urbanization would increase for the next several decades mainly in Asian developing countries. 70% of the world's population will be living in urban areas by 2030. It increases the number of built-up areas and people have relied on motorized transportation. Indonesia has inadequate physical infrastructure, and its development conditions do not meet as ideal neighbourhoods. Pedestrians, specifically, are thus the most vulnerable users. In Indonesia, they have the greatest fatality rate compared to other users, which is higher than other Asian developing countries. Therefore, walking is a significant issue in the built environment mainly at the neighbourhood scale as it is the forefront and first unit in the larger territories.
Focusing on the context of Sleman Regency Yogyakarta, this research aims to develop a framework for achieving a walkable neighbourhood in Sleman Regency, Yogyakarta. The objectives of this research are to identify the influencing factors of the built environment on the walkability of neighbourhood; to investigate the physical walkability factors; and to define the perception of residents on the physical walkability factors; to establish a framework on the strategies to enhance the walkability of neighbourhood in Sleman Regency. Those are conducted within five phases where a convergent mixed method approach is applied. Findings from the literature review and expert interview for the influencing factors of the built environment on the walkability of neighbourhood are synthesized, which are then applied to investigate the physical and perceived walkability factors that affect the selected existing neighbourhood in Sleman Regency as a case study, that is Depok Sub-regency. Map-based approaches by using satellite imagery, direct observation, and UMI walkability application are employed to investigate the physical walkability factors. Subsequently, the semi-structure interview with a total of 40 residents is conducted to define the perception of residents. Afterward, the findings from the physical and perceived walkability factors are synthesized to establish a framework for the strategies to enhance the walkability of neighbourhood in Sleman Regency Yogyakarta.
The results show that the neighbourhood is the Car-dependent neighbourhood, with walkscores ranging from 0 to 28. It means residents would be relying chiefly on their own motorized transportation in their journey within the neighbourhood. The investigation also finds that the physical walkability factors in the microscale have issues with the sidewalk and also other factors including street trees, traffic calming, and street furniture. The type, number, distribution of amenities, and connectivity of the street network affect the accessibility within neighbourhood which subsequently influences pedestrian facility to the walkability of neighbourhood for people who walk towards the journey to access their destinations within neighbourhood. The residents' perception is that physical walkability factors need to be improved to induce walking either walking as a mode of transportation or physical activity in terms of its safety, comfortability, distinctiveness, accessibility, directness, connectedness, legibility, and familiarity. Through the formulated framework as guidance generated from this study, the existing neighbourhood could be enhanced by using the fundamental factors, to achieve a walkable neighbourhood in Sleman Regency Yogyakarta, within the framework