Today, our blog hosts a brief interview with Alwyn Mathew, Research Associate at the University of Cambridge (UoC). UoC and the Department to Civil Engineering is the key D-HYDROFLEX partner for the development of a digital twin for dams.
Alwyn, thank you for accepting our invitation and welcome to the #meet the D-HYDROFLEX team blog series.
Q: Please say a few words about yourself for our readers to get to know you.
A: My name is Alwyn Mathew, and I am currently a Research Associate at the University of Cambridge, working under the supervision of Prof. Ioannis Brilakis. Before this, I served as an Postdoctoral Research Assistant at the University of Dundee in Scotland. I hold a PhD in 3D Computer Vision from the Indian Institute of Technology (IIT) Patna, India. My expertise spans pointcloud processing, 3D mapping, and 3D computer vision, focusing on advancing innovative technologies in these fields.
Q: Could you describe the role of the University of Cambridge on the project?
A: The University of Cambridge is pivotal in advancing research across multiple engineering disciplines, with various research groups involved in EU and non-EU funded projects. The Construction Information Technology (CIT) Laboratory is actively engaged in the D-HYDROFLEX project within the Department of Civil Engineering. CIT is dedicated to providing thought leadership in research, education, and technology transfer, explicitly addressing challenges related to infrastructure sensing (during construction and beyond), data analysis, and knowledge generation. Our lab collaborates with international, national, and local public and private organizations to tackle significant research problems and develop sustainable, practical solutions that push the boundaries of construction IT. We focus on innovating, synthesizing, and adapting next-generation IT methods and systems to enhance automation, construction processes, inspection, and rehabilitation techniques. Additionally, we are committed to educating a new generation of students and fostering multidisciplinary thinking to equip them to solve complex problems at the intersection of signal processing, computer vision, and construction engineering. We also offer continuous education opportunities to ensure practitioners stay at the forefront of IT methodologies and technologies in construction while facilitating technology transfer to disseminate cutting-edge knowledge to the industry effectively.
Q: We constantly hear for digital twins and how these visual models are critical for the energy challenges we are facing. What are your thoughts?
A: Upgrading existing infrastructure and digitizing assets to make them more sustainable and efficient is essential in today’s world. A digital twin, a virtual model of a physical asset, plays a crucial role in this process. It spans the entire lifecycle of an object, using real-time data from sensors to simulate behavior and monitor operations. Infrastructure digitization—from buildings and roads to bridges and dams—is at the forefront of these efforts. However, many large-scale infrastructures, like dams, are often decades old and lack digital records, meaning the digitalization process must start from scratch.
For older infrastructure, such as dams, 3D models often do not exist, and only outdated 2D drawings are available. This presents a significant technical challenge. The geometric model is a fundamental component for building a digital twin, and in the case of older dams, we need to generate an accurate “as-is” 3D model. Our geometric dam digital twin tool addresses this using advanced laser scanning, color imaging, and thermal scans of the dam structure. The collected point cloud data is then clustered and classified into distinct objects, and their relationships are inferred. The final output is a detailed mesh model of the dam, providing a precise geometric foundation for the digital twin.
This process requires thinking outside the box because we are bridging the gap between outdated physical records and modern digital infrastructure systems. We must innovate in both the data acquisition and modeling phases to ensure we can capture, process, and utilize this data in a meaningful and sustainable way.
Q: What are the main challenges of your work on the project, and how do you tackle them?
A: One of the main challenges we face in this project is the unique and complex nature of dams compared to other infrastructures like buildings. Unlike buildings, dams exhibit highly irregular, non-planar surfaces, making it difficult to generalize data across different dams. This lack of uniformity is a significant obstacle, especially when working with the limited publicly available datasets for dams. The challenge of modeling these non-planar surfaces is an exciting open problem in the field, demanding innovative approaches.
Moreover, dams are among the least standardized infrastructure for digital models. This lack of standardization means we must create and develop new modeling standards to support accurate and efficient digital twins for dams. Another critical challenge is the scarcity of data, which pushes us to explore unsupervised machine-learning methods for clustering and classification. Without abundant labeled data, these unsupervised techniques become essential in identifying patterns and relationships within the dam’s point cloud data. To tackle the modeling issues posed by non-planar surfaces, we rely on sophisticated algorithms that dissect and analyze the complex geometries involved. By addressing these challenges with advanced methodologies, we aim to create robust, scalable solutions that can be applied to many dam structures.
Q: What are your expectations from the project?
A: Our work on the D-HYDROFLEX project has allowed us to expand beyond our previous experience in digitizing infrastructure like buildings and roads, introducing us to the unique challenges and opportunities of digitalizing dams. This project is a significant step toward broadening our expertise and applying advanced digitalization techniques in new sectors. By working on dams, we can refine our methodologies and adapt our solutions to handle more complex, non-standardized infrastructures. This opens up opportunities for us to make a broader impact across various infrastructure sectors in the future.
In terms of the energy ecosystem, the project has the potential to be truly transformative. Dams play a critical role in energy generation through hydropower, and optimizing their operations can significantly increase efficiency and sustainability. By creating accurate digital twins of dams, we will be able to monitor real-time conditions, predict potential failures, and optimize maintenance processes. This will lead to better resource management, more reliable energy production, and extended infrastructure lifespans, all of which contribute to a more sustainable energy future. The potential for transformation is a possibility and a reality we are actively working towards.
Furthermore, the knowledge and technological advancements developed through this project can be scaled and adapted to other renewable energy infrastructures. As we refine these digital tools, they could be applied to different sectors, such as wind farms, solar plants, and more, ultimately supporting the broader transition to clean energy and improving the efficiency of energy systems worldwide.
Alwyn, thank you a lot for this interesting interview!
A: Thank you!
Another # meet the D-HYDROFLEX team blog story is completed. Stay tuned to learn more on our workforce team!
Learn more on D-HYDROFLEX project here.
Please subscribe here to our newsletter to keep updated on our activities.
