STTR Phase I: Underground Live - an innovative, advanced analytical tool for characterizing the subsurface and reducing underground construction risk

The broader impact/commercial potential of this Small Business Technology Transfer (STTR) Phase I project is to reduce the cost of underground civil infrastructure construction and make new infrastructure more sustainable by developing a computational software-based tool to create and update ground models in real time using data-driven advanced analytics. Civil infrastructure is increasingly moving underground, including roadways, transit, utilities, and facilities.

However, risk in underground construction remains a critical barrier to attracting investment. A significant risk in building underground is the high uncertainty in ground conditions and physical properties influencing design and construction, resulting in increased costs due to over-design and/or delays and failures during construction. This project strives to improve the understanding of ground conditions by providing a solution to update the ground models during construction in a routine and autonomous manner, making full use of the wealth of data collected during construction.

This Small Business Technology Transfer (STTR) Phase I project aims to develop a technical solution that automates the process of back-analyzing ground properties and updating ground models in real time. Several technical challenges will be addressed. Current backanalyses practice is extraordinarily labor-intensive and expensive in managing and integrating data from underground infrastructure projects.

The dynamic environment during construction requires 4D inversion analysis on potentially hundreds of unique tunnel-structure interactions. Furthermore, the efficacy of the inversion in estimating geotechnical parameters, which has been demonstrated for only limited situations during the fundamental development of the techniques, needs to be validated.

The goals of the proposed research are to (1) develop algorithms to automatically integrate data from geotechnical instrumentation and monitoring, construction process monitoring, existing infrastructure, apriori geostatistical model, etc., (2) develop algorithms to dynamically update the geotechnical parameter inversion in spatial proximity of tunnel construction to adjacent structures, sensors, and ground conditions, (3) characterize the geotechnical parameter inversion efficacy across a broad variety of ground conditions and tunneling-structure interactions, and (4) learn the influence of sensing layout and optimized sensing on inversion efficacy. This Phase I work will lay the foundation for the development of a ‘live’ ground modelling tool.

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.