A5054715965- Scheduling and Optimization Algorithms
- Resource-Constrained Project Scheduling
- Sustainable Supply Chain Management
- Metaheuristic Optimization Algorithms Research
- Supply Chain Resilience and Risk Management
- Multi-Criteria Decision Making
- Construction Project Management and Performance
- Supply Chain and Inventory Management
- BIM and Construction Integration
- Solar Radiation and Photovoltaics
- Advanced Multi-Objective Optimization Algorithms
- Quality and Supply Management
- Product Development and Customization
- Blockchain Technology Applications and Security
- Photovoltaic System Optimization Techniques
- Scheduling and Timetabling Solutions
- Energy Load and Power Forecasting
- Microgrid Control and Optimization
- Advanced DC-DC Converters
- Complex Network Analysis Techniques
- Multilevel Inverters and Converters
- Big Data and Business Intelligence
- Optimization and Mathematical Programming
- Assembly Line Balancing Optimization
- IoT and Edge/Fog Computing
UNSW Canberra
2019-2025
Canberra (United Kingdom)
2020-2025
Australian Defence Force Academy
2019-2024
UNSW Sydney
2015-2024
University of Canberra
2015-2024
Rajshahi University of Engineering and Technology
2011-2023
ORCID
2020-2021
In recent years, several multi-method and multi-operator-based algorithms have been proposed for solving optimization problems. Generally, their performance is better than other that based on a single operator and/or algorithm. However, they do not perform consistently well over all the problems tested in literature. this paper, we propose an improved algorithm uses benefits of multiple differential evolution operators, with more emphasis placed best-performing operator. The by 10 5, 10, 15...
Many countries are challenged by the medical resources required for COVID-19 detection which necessitates development of a low-cost, rapid tool to detect and diagnose virus effectively large numbers tests. Although chest X-Ray scan is useful candidate images generated scans must be analyzed accurately quickly if tests processed. causes bilateral pulmonary parenchymal ground-glass consolidative opacities, sometimes with rounded morphology peripheral lung distribution. In this work, we aim...
This paper introduces a novel physical-inspired metaheuristic algorithm called “Light Spectrum Optimizer (LSO)” for continuous optimization problems. The inspiration the proposed is light dispersions with different angles while passing through rain droplets, causing meteorological phenomenon of colorful rainbow spectrum. In order to validate algorithm, three experiments are conducted. First, LSO tested on solving CEC 2005, and obtained results compared wide range well-regarded...
This article proposes a novel framework to improve the prediction accuracy of very short-term (5-min) wind power generation. The consists complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), monarch butterfly optimization (MBO) and long memory (LSTM), called CEMOLS. CEEMDAN is employed extract complex hidden features time-series data into intrinsic functions that are predicted using LSTM models dropout regularization retain long-term relationships between input...
The extensive propagation of industrial Internet Things (IIoT) technologies has encouraged intruders to initiate a variety attacks that need be identified maintain the security end-user data and safety services offered by service providers. Deep learning (DL), especially recurrent approaches, been applied successfully analysis IIoT forensics but their key challenge DL models is they struggle with long traffic sequences cannot parallelized. Multihead attention (MHA) tried address this...
Human activity recognition (HAR) has been regarded as an indispensable part of many smart home systems and healthcare applications. Specifically, HAR is great importance in the Internet Healthcare Things (IoHT), owing to rapid proliferation (IoT) technologies embedded various appliances wearable devices (such smartphones smartwatches) that have a pervasive impact on individual's life. The inertial sensors generate massive amounts multidimensional time-series data, which can be exploited...
The rapid growth of the Internet Things (IoT) technologies has generated a huge amount traffic that can be exploited for detecting intrusions through IoT networks. Despite great effort made in annotating records, number labeled records is still very small, increasing difficulty recognizing attacks and intrusions. This study introduces semi-supervised deep learning approach intrusion detection (SS-Deep-ID), which we propose multiscale residual temporal convolutional (MS-Res) module to...
The selection of an optimal maintenance strategy is one the principal strategic decisions that must be taken in many contexts order to maintain asset with minimum deterioration and deliver maximum output high quality. When considering cost, reliability, safety level industrial assets, decision makers select appropriate strategy, preferably, a known degree uncertainty. This article utilizes new interval type-2 fuzzy (IT2F) multicriteria decision-making method based on analytic hierarchy...
High voltage conversion dc/dc converters have perceived in various power electronics applications recent times. In particular, the multi-port converter structures are key solution DC microgrid and electric vehicle applications. This paper focuses on a modified structure of non-isolated four-port (two input two output ports) electronic interfaces that can be utilized (EV) The main feature this is its ability to accommodate energy resources with different current characteristics. suggested...