Dynamic Modelling for Resilient Smart Grids: Enhancing Robustness and Security in Energy Distribution Systems
The emergence of smart grid systems has transformed the landscape of energy generation, storage, and distribution. However, the inherent complexity and dynamical nature of smart grids pose new challenges in system reliability and optimality. Dynamical modelling on complex networks offers some key advantages in addressing these challenges. Built on the CI’s research in graph neural networks and neural dynamical models, this project aims to develop dynamically modelling techniques in smart grids that will optimise energy generation and distribution, and ensure reliable operation of future energy distribution systems.
The project consists of the following components:
1. Modelling of energy generation and consumption patterns: We will develop data-driven adaptive models to capture spatial and temporal dynamical patterns in smart grids, incorporating factors such as regional weather conditions and consumer behaviours.
2. Resilient smart grids: We will develop mathematical and computational models that help reveal network vulnerability, including nonlinear cascading effects of failures. The models enable simulation to suggest risk mitigation mechanisms.
3. Demand and market dynamic prediction: We will develop models to explain and simulate market dynamic and effectiveness of consumer behaviour incentives. The model will enable optimising energy generation schedule and market interventions.
Natural grasslands are threatened by a range of invasive weeds. This project aims to develop tools with novel image processing and machine learning techniques to characterize grasslands and target invasive weed species at landscape and paddock scales, leveraging a suite of readily available remotely sensed imagery. The project will deliver landscape-scale surveillance capability to characterize the extent and composition of grassland communities, including weed species, to improve management of grassland weeds, leading to more effective weed containment and reduced impact of weeds in natural grasslands and pastures.
RoboSort: Autonomous Recycling with Multi-Sensor Robotic Arm for Efficient Recyclable Material Detection and Grasping
The development of the proposed RoboSort system holds immense potential for revolutionizing waste management and recycling. By leveraging advanced sensing technologies and robotic precision, the system offers several significant benefits. Firstly, it enhances the efficiency and accuracy of recyclable material identification, leading to reduced contamination and increased recycling rates. This, in turn, contributes to resource conservation and minimizes the environmental impact of waste. Additionally, the robotic arm’s ability to pick up recyclables from a pile of trash autonomously streamlines the sorting process, reducing the need for manual labour and lowering associated costs. Moreover, the system promotes a safer working environment by minimizing human exposure to potentially hazardous materials within the waste stream. Overall, the adoption of this robotic system has the potential to enhance recycling practices, conserve valuable resources, and promote sustainable waste management practices, fostering a cleaner and greener future for our planet.
Data-driven Industry 4.0 Immersive Visualization System for Sustainable Advancing Smart Manufacturing:
A pivotal aspect of this project involves developing a user-friendly visualisation interface, empowering manufacturers to effortlessly access and comprehend the data and insights generated by the system. Such user-centric visualisations will enable manufacturers to make informed decisions and promptly implement changes to enhance their operations. This project exemplifies technological innovation with Industry 4.0 data collection, processing, and visualisation system tailored for smart manufacturing. The primary focus is on aiding local industries within the Australian manufacturing sector in enhancing their productivity and sustaining their competitiveness through the adoption of cutting-edge Industry 4.0 technologies in their day-to-day operations. The project’s core will be designing and developing prototypes capable of seamlessly collecting and processing data from diverse sources, including sensors, machines, and production lines under injecting environmental 3D point cloud data into digital immersive model to render spatial relationships and affordable physical layout. This data will then undergo comprehensive analysis to uncover valuable patterns and trends, providing actionable insights to optimise production processes to make well informed task decomposition.
In essence, this project will be providing a practical solution of Industry 4.0 technologies to real-world manufacturing challenges. Simultaneously, the project will contribute significantly to fostering sustainable industries by helping local manufacturers bolster their competitiveness and embrace advancements in the rapidly evolving global manufacturing landscape. It promotes resource optimisation, renewable energy integration, and data-driven circular economy practices, all contributing to a greener and more sustainable future.
Plastic Waste Management and Recycling with Hyperspectral/LiDAR Imaging from Unmanned Aerial Vehicles (UAVs) in Landfills and Agricultural Farmlands:
As the world grapples with plastic wastes, UAV imagery can be an effective tool to identify plastic waste hotspots and monitor plastics in landfills for efficient plastic waste recycling. UAV imagery can also be used to track large-scale plastic usage in agricultural farmland to prevent them ending up in landfills and hence, contribute to plastic circular economy.