Open Research
Permanent link for this community
About
Tuwhera Open Access Research Outputs provides free access to full texts of scholarly works from AUT's Schools, Research Institutes and Centres.
AUT's research is built on a foundation of innovation and excellence, with the aim that its discoveries and applications are shared in ways that enhance wellbeing and prosperity.
Adding your outputs
AUT staff research outputs are added to this collection via Research Elements. All items submitted to this collection are checked to ensure material does not breach publisher copyright and is suitable for archiving prior to being made open access.
Find out more about making your work open access
For help with Research Elements contact the Research and Innovation Office.
Browse
Browsing Open Research by Subject "09 Engineering"
Now showing 1 - 20 of 35
Results Per Page
Sort Options
- ItemA Comprehensive Multi-functional Controller for Hybrid Energy Storage Systems in DC Microgrids(Institute of Electrical and Electronics Engineers (IEEE), 2023-02-09) Lin, Xin; Zamora, Ramon; Baguley, Craig A
- ItemA Novel Approach for Reconstruction of IMFs of Decomposition and Ensemble Model for Forecasting of Crude Oil Prices(Institute of Electrical and Electronics Engineers (IEEE), 2024-02-26) Naeem, Muhammad; Aamir, Muhammad; Yu, Jian; Albalawi, OlayanIn recent eras, the complexity and fluctuations of the global crude oil prices have affected the economic progress of society. It is therefore, the oil price prediction has hauled the attention of scholars and policymakers. Driven by this critical concern for forecasting of crude oil prices, we introduces a novel hybrid model keeping in mind the primary objective of enhancing prediction accuracy while considering the specific characteristics as inherent in the data. To achieve this achievement, the trend is eliminated, allowing the scrutiny of whether the residual component validates the assurance of a series ran by stochastic trends. Following the removal of the trend, the residual component undergoes rigorous evaluation through autoregressive model following the decomposition model. Then we got support from the support vector machine, autoregressive integrated moving average and long-short term memory. The predictions accuracy can be evaluated by using the various performance metrics. The proposed hybrid model’s robustness and forecasting performance are rigorously evaluated through Diebold-Mariano test in comparison to competing models. Furthermore, the forecasting ability is evaluated via directional forecast. Ultimately, the empirical findings explicitly determine the superior predictive capabilities of the proposed hybrid model over alternative approaches.
- ItemA Survey of Indoor Positioning Systems Based on a Six-Layer Model(Elsevier BV, 2023-09-22) Sartayeva, Yerkezhan; Chan, Henry CB; Ho, Yik Him; Chong, Peter HJIndoor positioning has attracted considerable interest in both the industry and academic communities because of its wide range of applications, such as asset tracking, healthcare and context-aware services like targeted advertisements. While there are many indoor localisation methods, each has its advantages and disadvantages, taking into consideration various factors such as the effect of the indoor environment, ease of implementation, computational cost, positioning accuracy, etc. In other words, no single solution can cater for all different situations. Although many survey papers have been published on indoor positioning, new techniques and methods are proposed every year, so it is important to stay abreast of its latest developments. In addition, each survey has its own classification for indoor positioning systems without a common scheme. Inspired by the well-known OSI model and TCP/IP model, it would be desirable to develop a systematic framework for studying indoor positioning systems. In this paper, we make this new contribution by introducing a systemic survey framework based on a six-layer model to give a comprehensive survey of indoor positioning systems, namely: device layer, communication layer, network layer, data layer, method layer and application layer. Complementing the previous survey papers, this paper provides a survey of the latest research works on indoor positioning based on the six-layer model. Our emphasis is on systematic categorisation, machine learning-based enhancements, collaborative localisation and COVID-19-related applications. The six-layer model should provide a useful framework and new insights for the research community.
- ItemAn Adaptive Deep Learning Neural Network Model to Enhance Machine-Learning-Based Classifiers for Intrusion Detection in Smart Grids(MDPI AG, 2023-06-02) Li, Xue Jun; Ma, Maode; Sun, YihanModern smart grids are built based on top of advanced computing and networking technologies, where condition monitoring relies on secure cyberphysical connectivity. Over the network infrastructure, transported data containing confidential information, must be protected as smart grids are vulnerable and subject to various cyberattacks. Various machine learning based classifiers were proposed for intrusion detection in smart grids. However, each of them has respective advantage and disadvantages. Aiming to improve the performance of existing machine learning based classifiers, this paper proposes an adaptive deep learning algorithm with a data pre-processing module, a neural network pre-training module and a classifier module, which work together classify intrusion data types using their high-dimensional data features. The proposed Adaptive Deep Learning (ADL) algorithm obtains the number of layers and the number of neurons per layer by determining the characteristic dimension of the network traffic. With transfer learning, the proposed ADL algorithm can extract the original data dimensions and obtain new abstract features. By combining deep learning models with traditional machine learning-based classification models, the performance of classification of network traffic data is significantly improved. By using the Network Security Laboratory-Knowledge Discovery in Databases (NSL-KDD) dataset, experimental results show that the proposed ADL algorithm improves the effectiveness of existing intrusion detection methods and reduces the training time, indicating a promising candidate to enhance network security in smart grids.
- ItemAnalysis of Improved In-Cylinder Combustion Characteristics with Chamber Modifications of the Diesel Engine(MDPI AG, 2023-03-09) Doppalapudi, AT; Azad, AK; Khan, MMKThis study numerically analyses the effects of chamber modifications to investigate the improvement of in-cylinder combustion characteristics of the diesel engine using a computational fluid dynamics (CFD) approach. Five different modified chambers, namely, the double swirl combustion chamber (DSCC), bathtub combustion chamber (BTCC), double toroidal re-entrant combustion chamber (DTRCC), shallow depth combustion chamber (SCC), and stepped bowl combustion chamber (SBCC) were developed and compared with a reference flat combustion chamber (FCC). The effects of chamber modifications on temperature formation, velocity distribution, injection profiles, and in-cylinder turbulent motions (swirl and tumble ratio) were investigated. During the compression stroke, near top dead centre, the SCC showed a peak temperature of 970 K, followed by the FCC (968 K), SBCC (967 K), and DTRCC (748 K to 815 K). The DSCC and the SCC showed a high swirl ratio above 0.6, whereas the DTRCC and the BTCC showed a high tumble ratio of approximately 0.4. This study found that the SCC, BTCC, and DSCC have better combustion rates than the FCC in terms of temperature, heat release rate, and velocity distribution. However, the DTRCC showed poor temperature formation rates and rapid heat release rates (approx. 150 J/°CA), which can lead to rapid combustion and knocking tendencies. In conclusion, the DSCC and the SCC showed better combustion rates than the other chambers. In addition, turbulent motions inside the chambers avoided combustion in crevice regions. This study recommends avoiding chambers with wider bowls in order to prevent uneven combustion across the cylinder. Furthermore, split bowls such as the DSCC, along with adjusted injection rates, can provide better results in terms of combustion.
- ItemApplication of Multiple Intake Temporal Check All That Apply: A Case Study of Strawberry Yoghurt Formulated with Alternative Sweeteners.(Wiley, 2023-12-12) Chadha, Diksha; Hamid, Nazimah; Kantono, KevinBACKGROUND: It is crucial to reduce the high sugar content of fruit yoghurts in response to the excessive weight gain epidemic. The use of alternative sweeteners in yoghurts is often associated with the negative sensory attributes that can have an impact on yoghurt liking. The main objective of this research was to investigate the effect of alternative sweeteners and strawberry puree addition on the temporal sensory profile of yoghurt using multiple-intake temporal check all that apply (TCATA). A novel approach to the statical analysis of the temporal sensory data was employed by using Aligned Rank Transformation (ART)-ANOVA to investigate the differences between sensory attributes within different product and within different intakes. RESULTS: Results showed that attributes sweet and fruity decreased when the concentration of fruit puree was increased at low concentration of sucrose. Interestingly, when the concentration of fruit puree was increased fruitiness increased and mouthcoating decreased at low concentration of stevia. With successive intakes, attributes sweet, sour, creamy and fruity significantly decreased in yoghurts sweetened with sucrose, xylitol and stevia. Yoghurts containing low concentration of sucrose or xylitol and fruit puree were liked the most. However, stevia-sweetened yoghurts varying in sweetener and puree concentration were not significantly different in liking. In order to investigate the consumer acceptance of yoghurts, a novel approach was used i.e., utilising TCATA temporal data to investigate temporal drivers of liking for each yoghurt type. CONCLUSION: The use of multiple statistical analysis to analyse temporal data suggested that both sweetener and puree concentration need to be considered when developing products using alternative sweeteners. This article is protected by copyright. All rights reserved.
- ItemApplications of Building Information Modelling in the Early Design Stage of High-Rise Buildings(Elsevier BV, 2023-05-11) Omrany, H; Ghaffarianhoseini, A; Chang, R; Ghaffarianhoseini, A; Pour Rahimian, FHigh-rise buildings consume more energy and have greater environmental impacts, emphasising the need to adopt best practices during the design stage concerning BIM employment. However, despite strong support from the literature, little is known about the applications of BIM in high-rise buildings at the early design stage. Therefore, this paper aims to provide a holistic understanding of the current applications of BIM in high-rise buildings by analysing 60 studies. The findings identified seven research themes, including studies that used BIM for i) optimising building energy efficiency design; ii) collaborative design and planning; iii) life-cycle assessment; iv) designing net-zero energy buildings; v) integrating BIM with smart technologies for designing high-rise buildings; vi) cost analysis, and vii) structural design of high-rise buildings. Furthermore, this study highlights a number of challenges hindering the widespread application of BIM, alongside providing potential directions for the future development of BIM employment in high-rise buildings.
- ItemBattery Energy Storage Capacity Estimation for Microgrids Using Digital Twin Concept(MDPI AG, 2023-06-06) Padmawansa, Nisitha; Gunawardane, Kosala; Madanian, Samaneh; Than Oo, Amanullah MaungGlobally, renewable energy-based power generation is experiencing exponential growth due to concerns over the environmental impacts of traditional power generation methods. Microgrids (MGs) are commonly employed to integrate renewable sources due to their distributed nature, with batteries often used to compensate for power fluctuations caused by the intermittency of renewable energy sources. However, sudden fluctuations in the power supply can negatively impact battery performance, making it challenging to select an appropriate battery energy storage system (BESS) at the design stage of an MG. The cycle count of a battery in relation to battery stress is a useful measure for determining the general health of a battery and can aid in BESS selection. An accurate digital replica of an MG is required to determine the required cycle count and stress levels of a BESS. The Digital Twin (DT) concept can be used to replicate the dynamics of the MG in a virtual environment, allowing for the estimation of required cycle numbers and applied stress levels to a BESS. This paper presents a Microgrid Digital Twin (MGDT) model that can estimate the required cycle count and stress levels of a BESS without considering any unique battery type. Based on the results, designers can select an appropriate BESS for the MG, and the MGDT can also be used to roughly estimate the health of the currently operating BESS, allowing for cost-effective predictive maintenance scheduling for MGs.
- ItemBiocompatible Polymer for Self-Humidification(MDPI AG, 2023-10-16) Al-Jumaily, Ahmed M; Bartual, Sandra; Weerasinghe, NTLung supportive devices (LSDs) have been extensively utilized in treating patients diagnosed with various respiratory disorders. However, these devices can cause moisture depletion in the upper airway by interfering with the natural lubrication and air conditioning process. To remedy this, current technologies implement heated humidification processes, which are bulky, costly, and nonfriendly. However, it has been demonstrated that in a breath cycle, the amount of water vapor in the exhaled air is of a similar quantity to the amount needed to humidify the inhaled air. This research proposes to trap the moisture from exhaled air and reuse it during inhalation by developing a state-of-the-art hydrophilic/hydrophobic polymer tuned to deliver this purpose. Using the atom transfer radical polymerization (ATRP) method, a substrate was successfully created by incorporating poly (N-isopropyl acrylamide) (PNIPAM) onto cotton. The fabricated material exhibited a water vapor release rate of 24.2 ± 1.054%/min at 32 °C, indicating its ability to humidify the inhaled air effectively. These findings highlight the potential of the developed material as a promising solution for applications requiring rapid moisture recovery.
- ItemClinical Information System (CIS) Implementation in Developing Countries: Requirements, Success Factors, and Recommendations(Oxford University Press (OUP), 2023-02-07) Tun, Soe Ye Yint; Madanian, SamanehObjective Clinical Information System (CIS) usage can reduce healthcare costs over time, improve the quality of medical care and safety, and enhance clinical efficiency. However, CIS implementation in developing countries poses additional, different challenges from the developed countries. Therefore, this research aimed to systematically review the literature, gathering and integrating research findings on Success Factors (SFs) in CIS implementation for developing countries. This helps to integrate past knowledge and develop a set of recommendations, presented as a framework, for implementing CIS in developing countries. Materials and Methods A systematic literature review was conducted, followed by qualitative data analysis on the published articles related to requirements and SF for CIS implementation. Eighty-three articles met the inclusion criteria and were included in the data analysis. Thematic analysis and cross-case analysis were applied to identify and categorize the requirements and SF for CIS implementation in developing countries. Results Six major requirement categories were identified including project management, financial resources, government involvement and support, human resources, organizational, and technical requirements. Subcategories related to SF are classified under each major requirement. A set of recommendations is provided, presented in a framework, based on the project management lifecycle approach. Conclusion The proposed framework could support CIS implementations in developing countries while enhancing their rate of success. Future studies should focus on identifying barriers to CIS implementation in developing countries. The country-specific empirical studies should also be conducted based on this research’s findings to match the local context.
- ItemComprehensive Analysis of Dust Impact on Photovoltaic Module Temperature: Experimental Insights and Mathematical Modeling(Elsevier, 2023-10-27) Almukhtar, Hussam; Lie, Tek Tjing; Al-Shohani, Wisam AMDust accumulation substantially impacts the efficiency and thermal behavior of photovoltaic (PV) modules. Addressing a current knowledge gap, this article presents a comprehensive assessment of the impact of dust on PV module temperature. This endeavor includes a combination of systematic experiments with a novel developed mathematical model that uniquely incorporates key dust parameters such as emissivity, absorbance, and transmittance. The model enhances accuracy in estimating PV module temperature compared to existing mathematical models, which often overlook heat absorption due to dust and specific dust characteristics. These parameters are measured using specialized devices to ensure realistic values. Outdoor experiments are conducted to validate the model's predictions under real weather conditions, further highlighting the importance of accurate PV temperature estimation in dusty environments. Additionally, a comparative analysis is performed with existing mathematical models for PV temperature prediction, demonstrating the superior performance of the proposed model, which achieves the lowest average prediction error (mean absolute error of 1.4). These findings provide valuable insights into the estimation of PV temperature in dusty conditions, bridging the gap between theoretical modeling and practical application and underscoring the novelty and innovation introduced in this research.
- ItemComprehensive Review of Dust Properties and Their Influence on Photovoltaic Systems: Electrical, Optical, Thermal Models and Experimentation Techniques(MDPI AG, 2023-04-12) Almukhtar, H; Lie, Tek Tjing; Al-Shohani, WAM; Anderson, TA; Al-Tameemi, ZaidAs conventional energy sources decrease and worldwide power demand grows, the appeal of photovoltaic (PV) systems as sustainable and ecofriendly energy sources has grown. PV system installation is influenced by geographical location, orientation, and inclination angle. Despite its success, weather conditions such as dust substantially influences PV module performance. This study provides a comprehensive review of the existing literature on the impact of dust characteristics on PV systems from three distinct perspectives. Firstly, the study looks at the dust properties in different categories: optical, thermal, physical, and chemical, highlighting their significant impact on the performance of PV systems. Secondly, the research reviews various approaches and equipment used to evaluate dust’s impact on PV, emphasizing the need for reliable instruments to measure its effects accurately. Finally, the study looks at modeling and predicting the influence of dust on PV systems, considering the parameters that affect electrical, optical, and thermal behavior. The review draws attention to the need for further research into dust’s properties, including thermal conductivity and emissivity. This analysis highlights the need for further research to develop a scientific correlation to predict the thermal behavior of PV in dusty environments. This paper identifies areas for further research to develop more efficient and effective methods for analyzing this influence and improving PV efficiency and lifespan.
- ItemConsensus on a Netball Video Analysis Framework of Descriptors and Definitions by the Netball Video Analysis Consensus Group(BMJ, 2023-02-08) Mackay, L; Jones, B; Janse Van Rensburg, DC; Hall, F; Alexander, L; Atkinson, K; Baldrey, P; Bedford, A; Cormack, S; Clarke, J; Croft, H; Denton, K; Fox, AS; Hadley, P; Handyside, R; Hendricks, S; Kerss, J; Leota, L; Maddern, B; McErlain-Naylor, SA; Mooney, M; Pyke, D; Pistorius, D; Ramagole, DA; Ryan, D; Scott, F; Scott, T; Snow, J; Spencer, K; Thirlby, J; Viljoen, CT; Whitehead, SUsing an expert consensus-based approach, a netball video analysis consensus (NVAC) group of researchers and practitioners was formed to develop a video analysis framework of descriptors and definitions of physical, technical and contextual aspects for netball research. The framework aims to improve the consistency of language used within netball investigations. It also aims to guide injury mechanism reporting and identification of injury risk factors. The development of the framework involved a systematic review of the literature and a Delphi process. In conjunction with commercially used descriptors and definitions, 19 studies were used to create the initial framework of key descriptors and definitions in netball. In a two round Delphi method consensus, each expert rated their level of agreement with each of the descriptors and associated definition on a 5-point Likert scale (1 - strongly disagree; 2 - somewhat disagree; 3 - neither agree nor disagree; 4 - somewhat agree; 5 - strongly agree). The median (IQR) rating of agreement was 5.0 (0.0), 5.0 (0.0) and 5.0 (0.0) for physical, technical and contextual aspects, respectively. The NVAC group recommends usage of the framework when conducting video analysis research in netball. The use of descriptors and definitions will be determined by the nature of the work and can be combined to incorporate further movements and actions used in netball. The framework can be linked with additional data, such as injury surveillance and microtechnology data.
- ItemConstructing New Backbone Networks via Space-Frequency Interactive Convolution for Deepfake Detection(Institute of Electrical and Electronics Engineers (IEEE), 2023-10-16) Guo, Zhiqing; Jia, Zhenhong; Wang, Liejun; Wang, Dewang; Yang, Gaobo; Kasabov, NikolaThe serious concerns over the negative impacts of Deepfakes have attracted wide attentions in the community of multimedia forensics. The existing detection works achieve deepfake detection by improving the traditional backbone networks to capture subtle manipulation traces. However, there is no attempt to construct new backbone networks with different structures for Deepfake detection by improving the internal feature representation of convolution. In this work, we propose a novel Space-Frequency Interactive Convolution (SFIConv) to efficiently model the manipulation clues left by Deepfake. To obtain high-frequency features from tampering traces, a Multichannel Constrained Separable Convolution (MCSConv) is designed as the component of the proposed SFIConv, which learns space-frequency features via three stages, namely generation, interaction and fusion. In addition, SFIConv can replace the vanilla convolution in any backbone networks without changing the network structure. Extensive experimental results show that seamlessly equipping SFIConv into the backbone network greatly improves the accuracy for Deepfake detection. In addition, the space-frequency interaction mechanism does benefit to capturing common artifact features, thus achieving better results in cross-dataset evaluation. Our code will be available at https://github.com/EricGzq/SFIConv.
- ItemControl Strategies and Stabilization Techniques for DC/DC Converters Application in DC MGs: Challenges, Opportunities, and Prospects—A Review(MDPI AG, 2024-01-31) Nduwamungu, Aphrodis; Lie, Tek Tjing; Lestas, Ioannis; Nair, Nirmal-Kumar C; Gunawardane, KosalaDC microgrids (DC MGs) offer advantages such as efficiency, control, cost, reliability, and size compared to AC MGs. However, they often operate with numerous constant power loads (CPLs), exhibiting a negative incremental impedance characteristic that can lead to instability. This instability weakens stability boundaries and reduces system damping, especially when dealing with pulsed power loads (PPLs) on electric aircraft, ships, and cars. Linear controllers may not ensure stability across various operations, causing voltage dips and potential system instability. To secure DC/DC converter functionality and comply with impedance specifications, it is crucial to consider minor loop gain in control strategies and stabilization techniques. Employing diverse methods to decrease minor loop gain in DC/DC converters is essential. A comprehensive evaluation, including strengths, weaknesses, opportunities, and threats (SWOT) analysis, is conducted to assess control strategies, stabilization techniques, and stability standards for different DC/DC converters, identifying SWOT.
- ItemDC Circuit Breaker Evolution, Design, and Analysis(MDPI AG, 2023-08-23) Moradian, Mehdi; Lie, Tek Tjing; Gunawardane, KosalaWhile traditional AC mechanical circuit breakers can protect AC circuits, many other DC power distribution technologies, such as DC microgrids (MGs), yield superior disruption performance, e.g., faster and more reliable switching speeds. However, novel DC circuit breaker (DCCB) designs are challenging due to the need to quickly break high currents within milliseconds, caused by the high fault current rise in DC grids compared to AC grids. In DC grids, the circuit breaker must not provide any current crossing and must absorb surges, since the arc is not naturally extinguished by the system. Additionally, the DC breaker must mitigate the magnetic energy stored in the system inductance and withstand residual overvoltages after current interruption. These challenges require a fundamentally different topology for DCCBs, which are typically made using solid-state semiconductor technology, metal oxide varistors (MOVs), and ultra-fast switches. This study aims to provide a comprehensive review of the development, design, and performance of DCCBs and an analysis of internal topology, the energy absorption path, and subcircuits in solid-state (SS)-based DCCBs. The research explores various novel designs that introduce different structures for an energy dissipation solution. The classification of these designs is based on the fundamental principles of surge mitigation and a detailed analysis of the techniques employed in DCCBs. In addition, our framework offers an advantageous reference point for the future evolution of SS circuit breakers in numerous developing power delivery systems.
- ItemDevelopment and High-Fidelity Simulation of Trajectory Tracking Control Schemes of a UUV for Fish Net-Pen Visual Inspection in Offshore Aquaculture(Institute of Electrical and Electronics Engineers (IEEE), 2023-11-30) Tun, Thein Than; Huang, Loulin; Preece, Mark AnthonyOffshore aquaculture fish farming faces labor shortage, safety, productivity and high operating cost issues. Unmanned underwater vehicles (UUVs) are being deployed to mitigate these issues. One of their applications is the fish net-pen visual inspection. This paper aims to develop and simulate with high-fidelity several trajectory tracking control schemes for a UUV to visually inspect a fish net-pen in a standard task scenario in offshore aquaculture under 0.0 m/s, 0.5 m/s and 0.9 m/s underwater current disturbances. Three controllers, namely 1) Proportional-Derivative control with restoring force & moment compensation (Compensated-PD), 2) Proportional-Integral-Derivative control with restoring force & moment compensation (Compensated-PID), and 3) computed torque (or) inverse dynamics control (CTC/IDC) were conducted on a 6 degrees-of-freedom (DoF) BlueROV2 Heavy Configuration dealing with 12 error states (pose and twist). A standard task scenario for the controllers was formulated based on the Blue Endeavour project of the New Zealand King Salmon company located 5 kilometres due north of Cape Lambert, in northern Marlborough. This simulated experimental study gathered and applied many available and physically quantifiable parameters of the fish farm and a UUV called BlueROV2 Heavy Configuration. Results show that while utilizing the minimum thrust, CTC/IDC outperforms Compensated-PID and Compensated-PD in overall trajectory tracking under different underwater current disturbances. Numerical results measured with root-mean-square-error (RMSE), mean-absolute-error (MAE) and root-sum-squared (RSS) are reported for comparison, and simulation results in the form of histograms, bar charts, plots, and video recordings are provided. Future work will explore into advanced controllers, with a specific emphasis on energy-optimal control schemes, accompanied by comprehensive stability and robustness analyses applied to linear and nonlinear UUV models.
- ItemDual Knowledge Distillation on Multiview Pseudo Labels for Unsupervised Person Re-Identification(Institute of Electrical and Electronics Engineers (IEEE), 2024) Zhu, Wenjie; Peng, Bo; Yan, Wei QiUnsupervised person re-identification (Re-ID) has made significant progress by leveraging valuable pseudo labels from completely unlabeled data. However, the predominant use of pseudo labels heavily relies on clustering results, which may lead to the accumulation of supervision deviation due to inevitable noise. In this paper, we propose a novel framework, namely Dual Knowledge Distillation on Multiview Pseudo Labels (DKD-MPL), to address this challenge. Specifically, the proposed DKD-MPL framework consists of two modules: Global Knowledge Distillation (GKD) and Self-Knowledge Distillation (SKD). In the GKD module, the pseudo labels obtained from the epoch-wise clustering procedure serve as the logits for the teacher model, while the mini-batch query images' pseudo labels act as the logits for the student model. Within the SKD module, we facilitate self-knowledge distillation by considering the pseudo labels generated by positive anchors and query images as two augmentations of the mini-batch data. As a result, DKD-MPL facilitates the exploitation of both global and local complementary knowledge across different views of pseudo labels, thereby mitigating supervision deviation. To demonstrate the effectiveness of DKD-MPL, we provide a theoretical analysis of the proposed loss and conduct extensive experiments on four popular datasets, e.g., Market-1501, DukeMTMC-reID, MSMT17, and VeRi-776. The results indicate that our method surpasses unsupervised approaches and achieves comparable performance to supervised person Re-ID methods.
- ItemDynamic- Structured Reservoir Spiking Neural Network in Sound Localization(Institute of Electrical and Electronics Engineers (IEEE), 2024) Roozbehi, Zahra; Narayanan, Ajit; Mohaghegh, Mahsa; Saeedinia, Samaneh-AlsadatSound source localization is a critical problem in various fields, including communication, security, and entertainment. Binaural cues are a natural technique used by mammalian ears for efficient sound source localization. Spiking neural networks (SNNs) have emerged as a promising tool for implementing binaural sound source localization approaches. However, optimizing the topology and size of SNNs is crucial to reduce computational costs while maintaining accuracy. This paper proposes a real-time structure of a reservoir SNN (rSNN) called Adaptive-Resonance-Theory-based rSNN (ART-rSNN) for localizing sound sources in the time domain by integrating an energy-based localization method. The dataset used in this work is recorded by two different omnidirectional microphones from a real environment. The dataset includes various sound events such as speech, music, and environmental sounds. The proposed ART-rSNN architecture can dynamically adjust the location of its neurons to amplify estimated energy near the sound source, resulting in higher localization accuracy. Our proposed method outperforms several conventional and state of the art algorithms in terms of accuracy and is able to detect the front and back direction of azimuth angle. This work demonstrates the potential of dynamic neuron arrangements in SNNs for improving sound source localization in practical applications.
- ItemElectricity Price Forecasting in New Zealand: A Comparative Analysis of Statistical and Machine Learning Models with Feature Selection(Elsevier BV, 2023-06-19) Kapoor, Gaurav; Wichitaksorn, NuttananIn this study, we present an empirical comparison of statistical models and machine learning models for daily electricity price forecasting in the New Zealand electricity market. We demonstrate the effectiveness of GARCH and SV models and their t-distribution variants when paired with feature selection techniques, including LASSO, mutual information, and recursive feature elimination. A key aspect of our study is the inclusion of a diverse set of explanatory variables in all models. We compare these models against a range of popular machine learning models, including LSTM, GRU, XGBoost, LEAR, and a four-layer DNN, where the latter two are considered benchmarks. Our results reveal that GARCH and SV models, particularly their t variants, perform exceptionally well when paired with feature selection techniques and explanatory variables. In most scenarios considered, these models outperform machine learning models when coupled with LASSO feature selection. This contribution provides a comprehensive evaluation of the performance of different models and feature selection techniques for electricity price forecasting in the New Zealand electricity market. Our best-performing model improves the symmetric mean absolute percentage error (sMAPE) and mean absolute scaled error (MASE) by 2% to 3% over the LEAR benchmark model, highlighting the practical relevance of our findings.