This Faculty Early Career Development (CAREER) grant will establish an integrated research and education program aimed at the next generation of physics-based data analytics to enable holistic (multiscale and multiphysics) characterization of the subsurface in real time. Coupled physics processes driven by engineered stimulation of the subsurface underlie many emerging technologies germane to advanced geo-infrastructures; examples include sustainable energy mining from enhanced geothermal systems and intelligent mitigation of the affiliated environmental impacts. Optimal design and closed-loop control of such operations require real-time feedback on the nature of progressive variations in the target subterranean regions. 3D in-situ tracking of multiphysics processes in such environments is, however, exceptionally challenging; engineered treatments (such as fluid or gas injection) are often induced in a complex domain whose structure and material properties are unknown (or uncertain) across multiple scales. Nevertheless, existing approaches to in-situ monitoring mostly rely on simplistic characterization of the subsurface, and mainly ignore the multiscale and coupled-physics nature of the induced processes in data inversion. Moreover, these tools are by and large computationally expensive and inapplicable for real-time sensing, or only amenable to ad hoc sensory configurations. Therefore, there exists a critical need for fast (yet robust) holistic data processing tools that transcend some of these limitations. In light of the fast-paced developments in sensing instruments, furnishing high-resolution spatiotemporal measurements and big data sets, such advances in data analytics is paramount for engineered systems of the future.
The research component of this project aims to establish a comprehensive (analytical, computational, and experimental) platform for: (1) real-time geometric reconstruction of advancing interfaces and volumetric process zones in multiphasic subterranean domains of a-priori unknown properties, (2) high-fidelity hydro-mechanical characterization of thus-recovered regions, and (3) verification and validation of these developments in a laboratory setting for better understanding of injection-induced multiphasic variations in randomly fractured rock masses pertinent to enhanced geothermal systems. This will be accomplished by taking advantage of the most recent advances in applied mathematics, geophysics, biomedical engineering, and sensor technology. In particular, the inverse solution is built upon three fundamental lynchpins: (i) inverse scattering and the theory of transmission eigenvalues, (ii) Marchenko integral equations and the generalized autofocusing concept, and (iii) non-iterative solutions to full-field inversion. The education component of this project aims at: (1) integration of the interdisciplinary knowledge underpinning state-of-the-art data processing tools for complex environments into the curriculum of science and engineering students at CU-Boulder and outreach activities, and (2) cultivating an effective knowledge transfer from research to academia and practice that includes all the stakeholders. In this vein, a three-tier educational program will be developed, involving: (i) engineering outreach to K-12 students with emphasis on underrepresented minorities, (ii) introducing WISE: wave-based inversion in subterranean environments as a new thrust in the engineering program at CU Boulder, and (iii) multilateral collaborations among scientists, engineers, and practitioners at regional, national and international levels.
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.
