FMSG: Cyber: Toward Future Underwater Additive Manufacturing of Bio-Based Construction Materials Through AI-Guided Sensing and Material Modeling

Louisiana State University, Baton Rouge LA

Investigators

Abstract

With the global rise in sea levels and frequent extreme weather conditions, effective and efficient underwater construction methods are increasingly essential for the resilience of coastal communities. However, traditional underwater construction approaches face a host of challenges, including severe working conditions, restricted access, and potential ecological damage. This Future Manufacturing Seed Grant (FMSG) funded project will explore the potential of additive manufacturing as an autonomous, advanced construction method to overcome these hurdles. Nevertheless, the complexity of implementing underwater additive manufacturing, from materials selection and process optimization to instrumentation development, presents significant challenges. The key concern is the formulation of concrete additives, typically handled via a trial-and-error method due to the regional variation of materials. Utilizing artificial intelligence-driven material modeling coupled with novel smart sensing systems, this research aims to unravel new insights to enable innovative underwater concrete additive manufacturing. This research could revolutionize underwater construction methods, fostering more efficient, sustainable, and eco-friendly solutions for coastal communities and infrastructure. The research will also be complemented by incorporating courses and outreach programs on artificial intelligence and underwater additive manufacturing topics for graduate, undergraduate, and K-12 students. Specifically, the research team will actively involve underrepresented K-12 students at fundamental project levels and inspire them to further explore STEM fields. The specific goal of this project is to decipher the intricate process-structure-property in underwater concrete additive manufacturing. This endeavor will replace traditional, tedious trial-and-error methods using molecular dynamics simulations, providing a comprehensive understanding of the physical principles governing the interactions between cementitious compositions and various chemical additives. The project will address compatibility issues and potential side effects of chemical admixtures on the cementitious system. Additionally, considering changing materials formulations and underwater environmental conditions that will affect concrete rheological properties, the project will develop a novel multi-sensor system that will be integrated into the concrete 3D printer, providing an accurate real-time monitoring approach. The team will develop an advanced data fusion methodology that merges experimental data, simulation outputs, and sensor results. The study will incorporate fluid mechanics, thermal dynamics, and domain knowledge (such as hydration curves) to build a physics-guided machine learning model. This model will offer a comprehensive understanding of the additive manufacturing process, leading to precise parameter control and improved reliability in underwater concrete additive manufacturing. The project's outcomes will advance the fundamental comprehension of the process-structure-property relationship in additive manufacturing and accelerate the technique development. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.