16168 research outputs found
Sort by
Passive, fractional, battery equivalent-circuit model in time and frequency domains part 1: Linear model
No full textExisting time-domain fractional model simulations of batteries are either limited to short time sequences, frequently less than 100 s, if truly fractional or use low order RC-ladder fractional approximations to reduce computational burden. Here we present an entirely-passive, truly fractional, equivalent-circuit model of a battery. We rely on a Reimann-Louiville fractional order differintegral to account for long time-scales out to 12 days. An analytical solution is provided for the differintegral, subject to the constraint of piecewise constant current. We validate our model fitting against a multi-day sequence of measured time domain data and EIS measured to 10 μ Hz. The spectral content of the current waveform is identified as a crucial factor. The full evaluation of fractional elements leads to residual error of voltage waveforms that is amongst the best in the literature despite the model having only five parameters. In the time-domain, a root-mean-square error (RMSE) as low as 2.8 mV is achieved while maintaining a frequency-domain RMSE of 14 % from measured impedance values over a span of 6 decades. The use of time-weighted regression is shown to be important to the time-domain fit
Online Isolation Forest
No full textThe anomaly detection literature is abundant with offline methods, which require repeated access to data in memory, and impose impractical assumptions when applied to a streaming context. Existing online anomaly detection methods also generally fail to address these constraints, resorting to periodic retraining to adapt to the online context. We propose ONLINE-IFOREST, a novel method explicitly designed for streaming conditions that seamlessly tracks the data generating process as it evolves over time. Experimental validation on real-world datasets demonstrated that ONLINE-IFOREST is on par with online alternatives and closely rivals state-of-the-art offline anomaly detection techniques that undergo periodic retraining. Notably, ONLINE-IFOREST consistently outperforms all competitors in terms of efficiency, making it a promising solution in applications where fast identification of anomalies is of primary importance such as cybersecurity, fraud and fault detection
Workload demands during ball-in-play periods: A comparative study of locomotive and contact metrics in professional male and female rugby union players.
No full textTitle: Workload Demands during Ball-in-Play Periods: A Comparative Study of Locomotive and Contact Metrics in Professional Male for Female Rugby Union Players.
Purpose: Rugby Union is a high collision-based sport demanding the best physical, technical, tactical and decision-making capabilities of players to influence success on the field. Various factors affect the ability of females and males to perform and meet the demands of match-play. This thesis was developed as a pilot study and employed an innovative design that examined and compared contact and workload metrics between female and male professional rugby players, aiming to explore differences in match play.
Methods: Data were collected from 60 professional male and female rugby players, from two professional 2023 New Zealand National Provincial Rugby teams. All participants wore GPS units (Apex Pro Pod, STATSport, Newry, NIR). Each match was filmed and coded using a Sportscode video analysis software package (Sportscode 12.4.3, Sportstec, Australia), where contact and GPS metrics during Ball-in-Play periods were identified and recorded throughout the match. GPS and Sportscode data for each Ball-in-Play period for each match were combined in a bespoke software package. Differences between the sexes for each metric were analysed using Welch’s t-test with the level of significance set at P<0.05 and the magnitudes of the standardised differences were calculated using effect sizes determined by Cohen’s d (Hopkins, 2009, 2017). Tests for equality of variances, normality, and the examination of outliers were conducted in IBM SPSS Statistics (Version 27) and the few metrics that failed normality testing were log-transformed.
Results: Males locomotive metrics (GPS measures) and impacts (accelerometer measures) were typically significantly greater than females showing large to moderate magnitude differences. On the other hand, contacts were often significantly higher for females with small to large magnitude differences.
Conclusion: The differences between males and females were likely a combination of genetic differences, and differences in how the respective games were played, and the males in this study typically had greater full-time professional training support and resources. Due to these differences, altering female threshold standards for specific intensive locomotion metrics to suit the physical capabilities of females, may give more of an accurate representation of game demands. On the other hand, providing greater resources to female rugby union players for similar periods to the males may reduce the differences observed in this study
Identifying and classifying broader scale of disaster impacts to better inform disaster management policies and practice
No full textDisasters are often assessed by factors such as their magnitude, severity and impact on affected communities. Typically, disaster management policies are largely informed by learnings from large-scale disasters. Many studies, however, have shown that cumulative smaller scale disaster impacts have similar outcomes comparable to larger scale disasters. This indirectly contributes to a lack of consistency in and paucity of data related to both smaller-scale disasters and their associated cumulative impacts. This paper argues that greater attention needs to be paid to collecting disaster impact data on a range of disaster-scale events along with their cumulative impacts to better inform disaster management policies and practice. The paper aims to test the hypothesis that a better understanding of a wide range of disaster scale types and their impacts on communities, including small-scale disasters, enables the development of more just, consistent and equitable disaster management policies and practice. Analysing economic costs, evacuation and rainfall data from past flood events that affected Aotearoa-New Zealand over the last three decades, the paper found discrepancies and lack of coherence in disaster data collection and reporting, along with a predominant reliance on rainfall intensity data for disaster risk reduction and planning. The findings also confirmed the importance of identifying economic damages for a range of disaster scales, including those from small-scale disasters. This highlights the need for developing a spectrum to better classify and identify disaster impacts based on various spatial and temporal contexts. The paper concludes by affirming that such database would provide much needed evidence for developing a methodology that enables the identification of the level of disaster impacts; thereby guiding the implementation of more just, consistent and equitable disaster management policies and practice
Identifying temporal and spatial variations in the binding properties of transition metal-organic ligand complexes in cave waters using organic inclusions in speleothem calcite: Implications for paleoclimate research.
No full textSpeleothems such as stalagmites, stalactites and flowstones, contain paleoclimate information encoded in their geochemical properties. One of these properties, transition metal abundance in speleothem calcite, is tightly controlled by complexation reactions in cave water with dissolved organic matter (DOM). Organic compounds are also preserved in speleothem deposits. DOM structure varies from cave system to cave system and over time, but whether these structural variations meaningfully alter the way DOM controls trace metal incorporation into calcite is unclear. This is relevant to the development of speleothem transition metal content as a paleoclimate proxy, particularly over large spatial and temporal scales where DOM structure variations are likely. This study identifies variations in the binding properties of DOM preserved in speleothem calcite over time and space by extracting organic compounds of known age and origin from speleothems collected from New Zealand, Niuean, British, and Australian cave systems. The kinetic properties of DOM complexes from these extracts were measured using small volume diffusive gradients in thin films (SV-DGT) and the variation in the binding strength of complexes with Cu, Co, and Ni was compared across time and space. DOM binding of Cu and Co were found to vary significantly depending on the cave system from which the DOM originated. Whether this variation was linked to climate changes represented in the age of the DOM or to changes in system characteristics related to geographical setting was unclear. Despite this, it is apparent that there are factors related to the origin of DOM in cave systems that influence the kinetic restrictions applied to transition metals in cave water. This variability should be considered when linking trace element contents in cave precipitates to paleoclimate change
Longitudinal relations between belief in climate change and environmental efficacy
No full textClimate change is one the most important issues we face today as it has catastrophic consequences for our environment, communities, and well-being. Thus, an individual’s pro-environmental behaviours are important to mitigate the climate crisis we face today. Research indicates that increased awareness of the climate crisis has led to various climate actions such as the climate strikes in 2019. We test our pre-registered predictions using two waves of an annual national panel study, the New Zealand Attitudes and Values Survey (N = 13,918). Cross-lagged results indicate that the climate change is real item and the personal action item showed partial support for a bidirectional association one-year later. These longitudinal findings reveal a novel link between belief in climate change and environmental efficacy, which is crucial for understanding pro-environmentalism among/of individuals in the community
Predicting Steam Turbine Power Generation: A Comparison of Long Short-Term Memory and Willans Line Model
Get PDFThis study focuses on using machine learning techniques to accurately predict the generated power in a two-stage back-pressure steam turbine used in the paper production industry. In order to accurately predict power production by a steam turbine, it is crucial to consider the time dependence of the input data. For this purpose, the long-short-term memory (LSTM) approach is employed. Correlation analysis is performed to select parameters with a correlation coefficient greater than 0.8. Initially, nine inputs are considered, and the study showcases the superior performance of the LSTM method, with an accuracy rate of 0.47. Further refinement is conducted by reducing the inputs to four based on correlation analysis, resulting in an improved accuracy rate of 0.39. The comparison between the LSTM method and the Willans line model evaluates the efficacy of the former in predicting production power. The root mean square error (RMSE) evaluation parameter is used to assess the accuracy of the prediction algorithm used for the generator’s production power. By highlighting the importance of selecting appropriate machine learning techniques, high-quality input data, and utilising correlation analysis for input refinement, this work demonstrates a valuable approach to accurately estimating and predicting power production in the energy industry
Fijian students’ perceptions of their science learning environment in New Zealand secondary schools
Get PDFPacific students in Aotearoa, New Zealand, come from a range of cultural and ethnic backgrounds. This heterogeneous cultural and ethnic mix of students presents opportunities and challenges for teachers. Results from the National Certificate of Educational Achievement (NCEA) show that Pacific students are underperforming in science compared to the general cohort in secondary education. To help meet the needs of Pacific students, many New Zealand educators are undertaking research to explore the perceptions of students learning. Little research has been done on Fijian students to address this. The purpose of this study is to examine Fijian students' perceptions of their current and preferred learning environment and how it can be improved to suit the learning needs of Fijian students.
Talanoa, a Pacific research methodology, informed and guided this study. It is based on Pacific cultures and is primarily used to study Pacific peoples. A mixed-method approach was adopted to respond to the research questions. Talanoa method with a survey called the Constructivist Learning Environment Survey-Fijian students (CLES-FS) was used to collect the data. A total of 305 students participated in this study. A thematic approach was used to analyse the data obtained from the talanoa method. The CLES-FS data was analysed using the Statistical Package for Social Sciences (SPSS), and comparisons were made between the students’ perceptions of their actual and preferred learning environments. Variations between gender, ethnic differences (Fijian Indian and iTaukei) and between different year levels were also analysed.
Key findings show that positive student-teacher relationships, a sense of belonging to the class, and teachers’ understanding of students’ cultural experiences played a vital role in engaging students in their learning. The findings also highlighted that the Fijian students enjoyed working in groups. However, they indicated that excessive use of group work without defined learning intentions had a negative impact on learning. In addition, the study found that the students were reluctant to respond to or ask questions in front of the whole class. Students also identified that, at times, cultural values of respect for elders discouraged them from expressing their views to the teacher. Many students preferred to communicate with the teacher one-on-one about how they would like to learn and have more control over their classroom learning activities.
There was a statistically significant difference in the perceptions of iTaukei and Fijian Indian students regarding building relationships with their teachers and collaborative learning. iTaukei students perceived that they had fewer opportunities to build relationships with their teachers but had more opportunities to engage in collaborative learning than Fijian Indian students. In addition, there was a statistically significant difference in the perceptions of male and female students when it came to learning in a familiar context scale. Female students perceived they were given more opportunities to learn in a familiar context than their male counterparts.
This study has contributed to the understanding of Fijian students' perceptions of the current and preferred learning environments in New Zealand secondary schools. The research also demonstrates that CLES-FS and Talanoa methods can be used effectively for data triangulation. The findings of this study provide new insights into the knowledge gap in the literature about the learning needs of Fijian students studying in New Zealand classrooms. It is envisaged that the findings will help to inform the teachers and leaders of secondary schools who teach Fijian students
Educational engagement practices of Tongan parents in relation to their children's learning
Get PDFThis study explores the parental engagement of Tongan parents in relation to their children’s learning and designs an engagement framework to enhance this engagement. Bronfenbrenner’s (1979) ecological systems theory framework was employed to conceptualise and inform the analysis of parental engagement of Tongan parents in this research.
Talaloto, a new Pacific research methodology which I developed for use in this study, was utilised for data collection from Tongan parents, as a way of capturing their authentic voices, grounded within their culture. Talaloto provides a space for personal lived experiences to be told. Focus group interviews were used for gathering data from teachers and students. Initial data collection motivated me to develop and implement an intervention to enhance parental engagement practices, for a sample of Tongan parents, students, and teachers, associated with a local primary school.
The findings showed that parental engagement is an holistic process, requiring collaboration between home, school, and the wider community to enable both participation and interaction. Recognition of and valuing home funds of knowledge as well as the implementation of home visits by teachers, were found to be significant contributors to transforming relationships and lifting student achievement.
The intentional creation of an Engagement Broker role for the purposes of the intervention demonstrated the importance of such a role in providing guidance to all participants as they navigated differing contexts and new roles and relationships.
A key outcome of this study was the development of the ‘Kato Alu ‘framework, where a cultural metaphor is used to conceptualise and contextualise the understanding and use of parental engagement practices by Tongan parents in their children’s learning. I used the four stages of crafting the Kato Alu: Tokonaki (preparation), Lalanga (weaving), Teuteu’i (decoration) and Foaki (presentation), to symbolise the four steps for guiding the parental engagement process. The hope is that this framework may be applied in other Pacific settings and can be adapted for broader use
He raraunga inamata, he toronga ānamata, he huanga āke tonu atu: A multifaceted approach to improving quality of life for upper limb amputees
Get PDFThis thesis investigates innovative approaches to enhance the quality of life for upper limb amputees, integrating advanced prosthetic design, neurocontrol systems, and Māori knowledge systems. With the prevalence of prosthetic rejection and limitations in current designs affecting amputees' daily functionality and psychological well-being, a multifaceted strategy is essential. This research aims to address these challenges by improving prosthetic functionality, user control mechanisms, and incorporating cultural perspectives into prosthetic development.
The study first focuses on the design, development, and evaluation of four low-cost prosthetic hands. Assessing their Degrees of Freedom (DoF), Range of Motion (RoM), and Kapandji scores. The research proposes a novel hybrid coupled-tendon system. This system aims to combine the adaptability of underactuated designs with the precision control of coupled mechanisms, enhancing the prosthetic hands' functionality and grasp versatility.
In exploring control mechanisms, an EEG-based control system was developed to improve the intuitiveness of prosthetic control. Through comparative analysis of motor execution and action observation training regimes, facilitated by a digital twin, the study evaluates the efficacy of these methods in enhancing users' control over the prosthetics. The research identifies limitations in the classification of multiple active grasps and suggests potential improvements through augmented and virtual reality training environments.
This thesis incorporates Māori perspectives by examining how mātauranga Māori can inform and enrich prosthetic design and usage. Through the analysis of pūrākau Māori, this work highlights the importance of considering the mauri, whakapapa of materials in prosthetic manufacturing and the implications of tapu on control systems. This inclusion aims to create culturally resonant devices that acknowledge and integrate the spiritual and cultural dimensions of the user's identity, offering a holistic approach to prosthetic development.
This interdisciplinary research contributes to the field by proposing a comprehensive framework that not only advances prosthetic technology but also aligns with cultural values and practices. By addressing the technical, psychological, and cultural needs of upper limb amputees, the project underscores the potential of combining engineering innovation with cultural wisdom to significantly improve amputees' autonomy and well-being. The findings advocate for a more inclusive and holistic approach to prosthetic development, emphasizing the importance of user-centric design and cultural competency in enhancing the quality of life for individuals with limb loss