Mines Repository (Colorado School of Mines)
Not a member yet
21416 research outputs found
Sort by
Numerical modeling of gravity die casting processes with solidification present during filling
No full textIncludes bibliographical references.2025 Spring.Computational modeling of U-10wt.%Mo (U-10Mo) and Al-5wt.%Si (Al-5Si) was performed, simulating alloy casting mold filling and solidification with FLOW-3D CAST®’s gravity die casting workspace. Simulations and instrumented castings in the literature have shown casting into a thin plate, monolithic mold cavity using top-gated gravity die casting produces an undesirable temperature gradient in the fluid after filling, which results in porosity formation. Increasing the pouring temperature (TPour) and/or mold temperature (TMold) to reduce porosity formation results in long solidification times, which can cause other casting issues. In the present study, a mold was designed to reproduce simulated fluid flows, solidification evolution, and porosity distributions observed in previous work. TPour, TMold, and filling time (tFill) were varied to determine the sensitivity of these parameters on predicted porosity distributions and solidification times of thin U-10Mo plate castings. Between TPour and TMold, TPour has a greater influence on porosity formation, while TMold has a greater influence on solidification time for the temperature ranges investigated. These results suggest that to decrease porosity while minimizing solidification time, TPour should be increased preferentially over TMold. Reducing tFill reduces the extent of solidification during filling and subsequent porosity formation without increasing solidification time, potentially providing an effective method for minimizing defects associated with the interaction of convection and solidification during filling.
To provide insights into the modeling of defect formation using commercial computational fluid dynamics (CFD) packages, instrumented castings of Al-5Si were performed to compare with model predictions, including temperature evolution in the fluid and defect formations (e.g., porosity) during filling. Applying interfacial heat transfer coefficients (IHTC) from the gravity die casting literature results in an underestimation of the heat transfer between the fluid and mold, preventing accurate predictions of filling-related defects observed in experiments. Increasing the IHTC improves predictions of temperatures and cooling rates observed during casting, but even so, small discrepancies between the modeling and experiments prevent accurate defect predictions during filling. In particular, porosity predictions in the model do not match well with experiments, which is likely due to other mechanisms not considered here such as solid feeding and oxide entrapment. Future casting models should pay special consideration to the IHTC values, as these values determine the amount of solidification that occurs during filling and subsequent defect formations. To improve porosity predictions, future models should include additional porosity formation mechanisms (e.g., oxide entrapment)
Adaptive transmission and distribution dynamic contingency analysis with high penetrations of distributed energy resources
No full textIncludes bibliographical references.2025 Spring.Grid modernization brings forth beautiful effects, such as low carbon emissions, low system power losses, among many others. Many of these improvements can be attributed to smaller-scale power generation resources, which typically are renewable sources of energy. However, these devices are rapidly changing the structure of the power grid and the dynamics associated with it. Several new standards address this, as IEEE 1547 and FERC Order 2222 mandate numerous dynamic standards for distributed energy resources and also allow for their participation in power system markets. These standards will increase grid stability in a future with high penetrations of renewable energy, and make sure that they are treated in a fair manner in power system markets, both of which allow for a more resilient, reliable, and equitable power grid. However, introducing these standards into systems brings forth significant engineering challenges. One challenge is the added computational cost of modeling all distributed energy resources in large scale, bulk power system studies, yet their detailed modeling is crucial due to their proliferation and newfound dynamic behavior. Therefore, there lies a crucial engineering tradeoff for bulk power system studies; accuracy in system studies at the expense of faster, less computationally intensive simulations. This thesis provides a method which offers advantages from both sides of this tradeoff. An extension for transmission-oriented dynamic contingency analysis is shown, which allows for the incorporation of highly detailed, disaggregated feeder models in regions of anticipated high dynamic uniqueness, while not adding the usual computational burden of doing so. First, the effectiveness of short-circuit analysis as a voltage estimation tool in anticipation of a fault contingency are displayed. Subsequently, these voltage estimation results are used to create a multi-fidelity system, where loads in highly voltage-deviating areas are modeled in a distributed manner, and loads in lower voltage-deviating regions are modeled in an aggregated manner. Results show that by performing a dynamic simulation in this manner, simulation complexity mirrors that of a fully aggregated simulation while incorporating the accuracy of a fully disaggregated simulation, constructing the multi-fidelity system in an automated manner each time
Alternative primary treatment for energy-positive wastewater treatment: an exploration of pile cloth media filtration
No full textIncludes bibliographical references.2025 Spring.Signature page also lists author's name as Tayler J. Elwell.Wastewater treatment is essential for safeguarding public health, preserving environmental integrity, and ensuring the sustainable management of water resources. However, many wastewater treatment facilities (WWTFs) in the United States are approaching the end of their operating lifespan and operating at or beyond their design capacities. To address these issues, there is a critical need for innovative research aimed at developing energy-neutral wastewater treatment paradigms. Improving primary treatment methods is one of the most urgent technological upgrades needed in wastewater treatment. The adaptation of pile cloth filtration (PCF) for primary treatment can enhance the performance of primary treatment, increase secondary treatment capacities, enhance renewable energy production, and reduce costs of WWTFs. To address these research needs, this dissertation explores the effectiveness of PCF as an alternative to conventional primary sedimentation (CPS) treatment, specifically focusing on enhancing carbon diversion away from secondary treatment to anaerobic digestion processes.
This dissertation introduces a modified PCF technology that is designed to dewater solids prior to removal, the dewatering pile cloth filter (DPCF). It then compares the characteristics of municipal wastewater treatment with traditional PCF, DPCF and CPS, followed by secondary activated sludge processes in sequencing batch reactors. Results suggest that implementing PCF treatments can reduce aeration requirements by up to 19%, and the DPCF can generate a primary solids waste stream concentrated to greater than 1% solids. Additionally, the performance of DPCF is characterized upstream of a demonstration scale in a sequencing batch membrane bioreactor (SBMBR), where aeration requirements and energy demands are quantified. Integrating DPCF can reduce SBMBR aeration requirements by up to 21%, while generating a concentrated solid waste stream of greater than 1.5% solids and water recovery of 97%.
A comparative techno-economic analysis is presented to assess the financial distinctions for full-scale implementation of the proposed treatment technology using advanced modeling techniques. This structured approach aims to provide a clearer understanding of the benefits and feasibility of adopting improved PCF as primary treatment in WWTFs. DPCF primary treatment technology in commercial-scale WWTFs has the potential to reduce the capital costs of primary treatment (51%) and secondary treatments (16%), reduce operating costs (21%), and improve biogas production (35%)
Table manners at scale: introducing Lora Lens for efficient analysis & detoxification of language models
No full textIncludes bibliographical references.2025 Spring.Transformer-based language models have achieved significant advancements across numerous natural language processing (NLP) tasks. However, as these models grow in scale and complexity, ensuring interpretability and mitigating toxic outputs become increasingly critical challenges. This thesis addresses these issues by first analyzing the role attention heads play in propagating toxicity within models, leveraging a recently proposed interpretability tool known as Attention Lens. By decoding attention head outputs into human-interpretable tokens, we identify specific attention heads contributing disproportionately to toxic content generation and demonstrate that targeted interventions at the head level significantly reduce toxicity without requiring complete model retraining.
Despite its effectiveness, Attention Lens faces severe limitations in scalability and computational efficiency. To overcome these limitations, we propose and implement the Lora Lens, an innovative adaptation of Attention Lens employing Low-Rank Adaptation (LoRA) to drastically reduce memory footprint and computational cost. Specifically designed for compatibility with large-scale models such as Llama 3 8B, Lora Lens integrates seamlessly with HuggingFace's transformers library and Microsoft's DeepSpeed framework, enabling efficient distributed training. Our results demonstrate that Lora Lens maintains the interpretative capabilities of Attention Lens while significantly enhancing its efficiency and scalability, allowing practical deployment on models with billions of parameters.
Ultimately, this work contributes a practical, scalable interpretability technique, enabling researchers and practitioners to better understand, evaluate, and safely deploy large transformer models
Compositional hindered transport model for primary and enhanced oil recovery from fractured wells in unconventional tight-oil plays, A
No full textIncludes bibliographical references.2025 Spring.This research was conducted as a project within the Unconventional Reservoir Engineering Project (UREP) consortium at the Marathon Center of Excellence for Reservoir Studies (MCERS) in the Petroleum Engineering Department of Colorado School of Mines for the partial fulfilment of the PhD degree requirements in petroleum engineering. The objective of the research was to investigate hindered transport in nanoporous media during primary and enhanced oil recovery from unconventional tight-oil plays.
The recovery from tight-oil reservoirs is less than 10%; therefore, enhanced oil recovery (EOR) has been an industry focus not only for improving project economics but also for limiting the environmental footprint of oil and gas production from unconventional reservoirs. Furthermore, the industry primarily relies on conventional ideas and models to assess the effectiveness of EOR technologies and agents used in unconventional plays. However, in unconventional reservoirs, the effectiveness of the EOR methods can be improved by deciphering the key mechanism of mass transfer across phase boundaries and the fracture- matrix interfaces. The main defect of this approach is to emphasize displacement processes, which limits the focus of EOR to improve the mobility of fluids in the fracture network and larger pores. Because most of the stranded oil is in the matrix nanopore systems, transferring EOR agents from highly conductive fractures to tight matrix pores, achieving deeper matrix penetration, and alleviating the conditions immobilizing hydrocarbon components must be the key considerations to achieve quantitatively meaningful increases in cumulative recovery. This perception shifts the EOR focus from conventional, advective, displacement considerations in micro and macropore systems to intermolecular and surface phenomena governing the advective-diffusive transport and retention of fluids in fractured nanoporous matrix, which we will call hindered transport in this work.
The fact that produced oils mainly consist of the light end of the hydrocarbons indicates that heavier hydrocarbons may be filtered, hindered, and retained in the reservoir. Hindered transport has also been observed in core experiments and attributed to various mechanisms, such as molecular sieving, steric hindrance, and adsorption on the pore surfaces of the tight matrix. The impact of hindrance on production depends on the sizes, distributions, and connections of the pores and pore-throats, mineralogy of the rock, and composition of the reservoir fluids and injectants. Previous experimental work with mini cores of Niobrara formation has indicated that CO2 injection may reduce hindrance and mobilize heavier hydrocarbons into the oil stream. This work aims to convert the experimental observations to a mechanistic model, develop a numerical model for hindered transport in naturally fractured nanoporous media, provide deeper understanding of the physical phenomena contributing to hinderance, and assess the importance of hindrance on field scale by numerical simulations.
The key aspect of the model developed in this work is that hindrance of heavier hydrocarbons caused by adsorption and sieving does not only affect the accumulation terms but also alters the advective and diffusive flux components of the mass balance equation. Molecular dynamics simulations have indicated that diffusioosmosis caused by adsorption may change the permeability and relative permeability coefficients. Moreover, molecular interactions in crowded media have been reported to cause deviations from regular diffusivity coefficients or cause anomalous diffusion.
While the effects of compositional changes on component fluxes and the contribution of adsorption to the accumulation of components are regularly accounted for, accumulation due to sieving and flux alterations caused by adsorption, sieving, and crowding are not considered in the conventional models of fluid flow in porous media. Therefore, the central contribution of this work is to delineate major hydrocarbon retention mechanisms and propose a mechanistic model for hindered transport of hydrocarbons and CO2 in naturally fractured nanoporous media encountered in unconventional reservoirs. The main task of the study is to construct a one-dimensional, compositional numerical model, which considers fluid transfer between fracture and matrix media with the effects of molecular interactions, surface forces, and steric and dynamic effects on flow and transport. Obtaining the adsorption and sieving parameters used in the mechanistic model is not in the scope of this work. Although substantial work is needed to build a database of the hindered transport properties under practical conditions of interest, the values and ranges provided in the literature, which were obtained by experiments, theoretical models, and molecular simulations in previous studies, suffice for our objective to demonstrate the utility of our model and highlight the relative significance of hindrance in transport in fractured nanoporous media.
Besides discussing the basis, providing a mechanistic model, and developing small- and large-scale computational models, this work demonstrates the significance of molecular level phenomena and diffusion-driven exchange between matrix and fracture media. Importance of the matrix-block sizes (natural fracture densities) and considering transient matrix diffusion in numerical models are shown. The results emphasize the use of the appropriate multicomponent diffusive flux relations and highlight the inadequacy of the bulk and binary diffusion coefficients to model multicomponent diffusion in crowded media. It is shown that, hinderance causes considerable changes in compositional distributions under the same production conditions. The pressure dependency of the molecular diffusion, adsorption, and sieving parameters is more impactful at lower pressure ranges or over large pressure changes
Numerical methods to analyze and post-process solutions of plasma and gas dynamics problems
No full textIncludes bibliographical references.2025 Spring.Numerical methods to analyze and solve solutions to PDEs continue to evolve leading to a vast array of numerical techniques from scientific computing, uncertainty quantification, signal processing, and machine learning. Rooted in fundamental physics and constitutive relations, these tools are essential for in-depth numerical data extraction and analysis. In this doctoral work, we leverage the dispersion relation to: 1) calculate the growth rate of a collisionless plasma, described by the Vlasov-Poisson system for particle-in-cell plasma dynamics, and apply dimension reduction, and 2) analyze and mitigate spurious oscillations in hyperbolic PDEs, particularly the Euler equations for inviscid, compressible gas dynamics, through the use of post-processing filtering. For the plasma dynamics problem, we use active subspace analysis to analyze the key parameter contributions in the growth rate of steady-state plasma distributions. The global sensitivity analysis sheds light on the induced uncertainty due to perturbations in a collisionless plasma. As for hyperbolic PDEs, we use the Discontinuous Galerkin (DG) method, which we further improve by combining discontinuity detection and post-processing filtering to resolve the solution about the shock waves that arise in Euler equations. We present a hybrid filter for discontinuities using Smoothness-Increasing Accuracy-Conserving (SIAC) and data-driven techniques, which reduces the error about discontinuities by applying a convolutional neural network trained to resolve SIAC-filtered discontinuity data. In addition, we examine the effectiveness of the SIAC filter for time-stepping schemes using spectral analysis of the numerical dispersion to limit the spread of spurious oscillations in time for discontinuous solutions
Comanche group, Grant County, New Mexico
No full textMine report no. 1829.Typescript (carbon copy).Includes maps.Title supplied by cataloger.Report on Comanche group, southern portion -- Report on Comanche Copper Company's properties -- Letter of correspondence, dated, 1912 -- 1 partial sheet of handwritten notes -- Photocopy of plat (dated 1909) -- Two topographic quadrangle maps (dated 1950 and 1951)
Investigating the effect of biochar on the frost durability and durability assessment of concrete
No full textIncludes bibliographical references.2025 Spring.As the key binding constituent, cement plays a critical role in determining the engineering properties and durability of concrete. Since its production is responsible for 8% of the global CO2 emissions, there is significant interest in technologies that reduce cement’s carbon footprint. One emerging solution is the integration of biochar, a carbon negative, renewable material produced from the thermochemical conversion of biomass. Biochar’s impact on the mechanical and fresh properties of concrete has been widely studied, while biochar’s effect on concrete durability remains largely unknown. This thesis seeks to address this research gap by investigating the effect of biochar on the frost durability and durability assessment of biochar concrete.
In the first effort, biochar was used as a partial replacement of cement in mortars to investigate the validity of bulk electrical resistivity as an assessment of biochar mortar permeability. In conventional cementitious composites, bulk electrical resistivity measurements can be used as an approximation of permeability because electrical charge is only carried via electrolytic conductivity; dissolved charged ions in the pore solution migrate more easily (higher conductivity) in more permeable composites. However, cementitious composites that incorporate conductive materials can display both electrolytic conductivity and electric conductivity, both of which can affect bulk electrical resistivity measurements. Resistivity measurements cannot distinguish between these conductive pathways; thus, decreased resistivity may indicate increased electric conductivity through the solid phase or increased permeability, which is typically associated with decreased durability. This work found that while biochar can reduce matrix permeability, it also influences the electrical properties of mortars through changes to both electrical and electrolytic conductivity, making bulk electrical resistivity measurements a poor indicator of permeability changes.
In the second effort, biochar was used as a partial replacement of sand to evaluate compatibility between biochar and an air-entraining admixture and to explore biochar concrete’s resistance to rapid freeze-thaw testing. Entrained air voids are typically integrated into concrete to provide resistance against frost damage, but reduced concrete permeability can also enhance this protection. To entrain the required 6 vol% air in concrete, concrete mixes with biochar required 970 - 2800% more air-entrainer compared to a control. Additionally, despite improved permeability at some replacement levels, non-air-entrained biochar concrete failed in less than 15 freeze-thaw cycles, compared to conventional concrete which failed in 97 freeze-thaw cycles. Air-entrained biochar concrete showed improvement, but still failed between 165 and 185 freeze-thaw cycles. This degraded freeze-thaw performance was attributed to biochar concrete’s higher sorptivity, associated with pore refinement. When water freezes in small pores, higher tensile pressures are developed than when water freezes in large pores, leading to greater damage
Prudential Mining, Tunneling and Transportation Company: Georgetown, Colorado, The
No full textMine report no. 2132.Includes illustrations and a map entitled, Plat of the Prudential Mining, Tunneling and Transportation Company's property.Physical copy held in the Russell L. & Lyn Wood Mining History Archive is a gift of the Russell L. & Lyn Wood Mining History Archive Fund