Search for: All records

The explosive growth in supercomputers capacity has changed simulation paradigms. Simulations have shifted from a few lengthy ones to an ensemble of multiple simulations with varying initial conditions or input parameters. Thus, an ensemble consists of large volumes of multi-dimensional data that could go beyond the exascale boundaries. However, the disparity in growth rates between storage capabilities and computing resources results in I/O bottlenecks. This makes it impractical to utilize conventional postprocessing and visualization tools for analyzing such massive simulation ensembles. In situ visualization approaches alleviate I/O constraints by saving predetermined visualizations in image databases during simulation. Nevertheless, the unavailability of output raw data restricts the flexibility of post hoc exploration of in situ approaches. Much research has been conducted to mitigate this limitation, but it falls short when it comes to simultaneously exploring and analyzing parameter and ensemble spaces. In this paper, we propose an expert-in-the-loop visual exploration analytic approach. The proposed approach leverages: feature extraction, deep learning, and human expert–AI collaboration techniques to explore and analyze image-based ensembles. Our approach utilizes local features and deep learning techniques to learn the image features of ensemble members. The extracted features are then combined with simulation input parameters and fed to the visualization pipeline for in-depth exploration and analysis using human expert + AI interaction techniques. We show the effectiveness of our approach using several scientific simulation ensembles. 

The canonical technique for nonlinear modeling of spatial/point‐referenced data is known as kriging in geostatistics, and as Gaussian Process (GP) regression for surrogate modeling and statistical learning. This article reviews many similarities shared between kriging and GPs, but also highlights some important differences. One is that GPs impose a process that can be used to automate kernel/variogram inference, thus removing the human from the loop. The GP framework also suggests a probabilistically valid means of scaling to handle a large corpus of training data, that is, an alternative to ordinary kriging. Finally, recent GP implementations are tailored to make the most of modern computing architectures, such as multi‐core workstations and multi‐node supercomputers. We argue that such distinctions are important even in classically geostatistical settings. To back that up, we present out‐of‐sample validation exercises using two, real, large‐scale borehole data sets acquired in the mining of gold and other minerals. We compare classic kriging with several variations of modern GPs and conclude that the latter is more economical (fewer human and compute resources), more accurate and offers better uncertainty quantification. We go on to show how the fully generative modeling apparatus provided by GPs can gracefully accommodate left‐censoring of small measurements, as commonly occurs in mining data and other borehole assays. 

Two- and three-dimensional rock-penetrating-radar data were acquired on the wall of a pillar in an underground limestone mine. The objective was to test the ability of radar to image fractures and karst voids and to characterize their geometry, aperture, and fluid content, with the goal of mitigating mining hazards. Strong radar reflections in the field data correlate with fractures and a cave exposed on the pillar walls. Large pillar wall topography was included in the steep-dip Kirchhoff migration algorithm because standard elevation corrections are inaccurate. The depth-migrated 250 MHz radar images illuminate fractures, karst voids, and the far wall of the pillar up to approximately 25 m depth into the rock, with a spatial resolution of <0.5 m. Higher frequency radar improved the image resolution and aided in the interpretation, but at the cost of shallower depth of penetration and extra acquisition effort. Due to the strong contrast in physical properties between the rock and the fracture fluid, fractures with apertures as thin as a 50th of a radar wavelength were imaged. Water-filled fractures with mm-scale aperture and air-filled fractures with cm-scale apertures produce strong reflections at 250 MHz. A strong variation in the reflection amplitude along each fracture is interpreted to represent the variable fracture aperture and the nonplanar fracture structure. Fracture apertures were quantitatively measured, but distinguishing water from air-filled fractures was not possible due to the complex radar wavelet and fracture geometry. Two conjugate fracture sets were imaged. One of these fracture sets dominates the rock mass stability and water inrush challenges throughout the mine. All of the detected voids and a large cave are at the intersection of two fractures, indicating preferential water flow and dissolution along conjugate fracture intersections. Detecting, locating, and characterizing fractures and voids prior to excavation can enable miners to mitigate potential collapse and flood hazards before they occur.