Global ETD Search

1	The Investigation of the Environmental Fate and Transport of 2,4- dinitroanisole(DNAN) in Soils Arthur, Jennifer, Arthur, Jennifer January 2017 (has links) New explosive compounds that are less sensitive to shock and high temperatures are being tested on military ranges as replacements for 2, 4, 6-trinitrotoluene (TNT) and hexahydro-1, 3, 5-trinitro-1, 3, 5-triazine (RDX). One of the two compounds being tested is 2, 4-dinitroanisole (DNAN), which has good detonation characteristics and is one of the main ingredients in a suite of explosive formulations being tested. Data on the fate and transport of DNAN is needed to determine its potential to reach groundwater and be transported off base, a result which could create future contamination problems on military training ranges and trigger regulatory action. In this study, I measured how DNAN in solution interacts with different types of soils from across the United States. I conducted kinetic and equilibrium batch soil adsorption experiments, saturated column experiments with DNAN and dissolution and transport studies of insensitive munitions (IMX-101, IMX -104), which include DNAN, 3-nitro-1,2,4-triazol-5-one (NTO), nitroguanidine (NQ) and hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), under steady state and transient conditions. In the rate studies, change in DNAN concentration with time was evaluated using the first order kinetic equation. Solution mass-loss rate coefficients ranged between 0.0002 h-1 and 0.0068 h-1. DNAN was strongly adsorbed by soils with linear adsorption coefficients ranging between 0.6 and 6.3 L kg-1, and Freundlich coefficients between 1.3 and 34 mg1-n Ln kg-1. Both linear and Freundlich adsorption coefficients were positively correlated with the amount of organic carbon and cation exchange capacity of the soil. In saturated miscible-displacement experiments, it was shown that under flow conditions DNAN transforms readily with formation of amino transformation products, 2-amino-4-nitroanisole (2-ANAN) and 4-amino-2-nitroanisole (4-ANAN). Dissolution miscible-displacement experiments demonstrated that insensitive munition compounds dissolved in order of aqueous solubility as indicated by earlier lab and outdoor dissolution studies. The sorption of NTO and NQ was low, while RDX, HMX, and DNAN all adsorbed to the soils. DNAN transformed in soils with formation of amino-reduction products, 2- ANAN and 4-ANAN. Adsorption parameters determined by HYDRUS-1D generally agreed with batch and column study adsorption coefficients for pure NTO and DNAN. The magnitudes of retardation and transformation observed in these studies result in significant attenuation potential for DNAN in soils, which would reduce risk of groundwater contamination. 2, 4-dinitroanisole (DNAN) Dissolution IMX-101 IMX-104 Insensitive munitions Soil adsorption
2	Design Space Exploration of MobileNet for Suitable Hardware Deployment DEBJYOTI SINHA (8764737) 28 April 2020 (has links) <p> Designing self-regulating machines that can see and comprehend various real world objects around it are the main purpose of the AI domain. Recently, there has been marked advancements in the field of deep learning to create state-of-the-art DNNs for various CV applications. It is challenging to deploy these DNNs into resource-constrained micro-controller units as often they are quite memory intensive. Design Space Exploration is a technique which makes CNN/DNN memory efficient and more flexible to be deployed into resource-constrained hardware. MobileNet is small DNN architecture which was designed for embedded and mobile vision, but still researchers faced many challenges in deploying this model into resource limited real-time processors.</p><p> This thesis, proposes three new DNN architectures, which are developed using the Design Space Exploration technique. The state-of-the art MobileNet baseline architecture is used as foundation to propose these DNN architectures in this study. They are enhanced versions of the baseline MobileNet architecture. DSE techniques like data augmentation, architecture tuning, and architecture modification have been done to improve the baseline architecture. First, the Thin MobileNet architecture is proposed which uses more intricate block modules as compared to the baseline MobileNet. It is a compact, efficient and flexible architecture with good model accuracy. To get a more compact models, the KilobyteNet and the Ultra-thin MobileNet DNN architecture is proposed. Interesting techniques like channel depth alteration and hyperparameter tuning are introduced along-with some of the techniques used for designing the Thin MobileNet. All the models are trained and validated from scratch on the CIFAR-10 dataset. The experimental results (training and testing) can be visualized using the live accuracy and logloss graphs provided by the Liveloss package. The Ultra-thin MobileNet model is more balanced in terms of the model accuracy and model size out of the three and hence it is deployed into the NXP i.MX RT1060 embedded hardware unit for image classification application.</p> Computer Engineering Design Space Exploration DNN iMX RT1060 Data augmentation
3	Design Space Exploration of MobileNet for Suitable Hardware Deployment Sinha, Debjyoti 05 1900 (has links) Indiana University-Purdue University Indianapolis (IUPUI) / Designing self-regulating machines that can see and comprehend various real world objects around it are the main purpose of the AI domain. Recently, there has been marked advancements in the field of deep learning to create state-of-the-art DNNs for various CV applications. It is challenging to deploy these DNNs into resource-constrained micro-controller units as often they are quite memory intensive. Design Space Exploration is a technique which makes CNN/DNN memory efficient and more flexible to be deployed into resource-constrained hardware. MobileNet is small DNN architecture which was designed for embedded and mobile vision, but still researchers faced many challenges in deploying this model into resource limited real-time processors. This thesis, proposes three new DNN architectures, which are developed using the Design Space Exploration technique. The state-of-the art MobileNet baseline architecture is used as foundation to propose these DNN architectures in this study. They are enhanced versions of the baseline MobileNet architecture. DSE techniques like data augmentation, architecture tuning, and architecture modification have been done to improve the baseline architecture. First, the Thin MobileNet architecture is proposed which uses more intricate block modules as compared to the baseline MobileNet. It is a compact, efficient and flexible architecture with good model accuracy. To get a more compact models, the KilobyteNet and the Ultra-thin MobileNet DNN architecture is proposed. Interesting techniques like channel depth alteration and hyperparameter tuning are introduced along-with some of the techniques used for designing the Thin MobileNet. All the models are trained and validated from scratch on the CIFAR-10 dataset. The experimental results (training and testing) can be visualized using the live accuracy and logloss graphs provided by the Liveloss package. The Ultra-thin MobileNet model is more balanced in terms of the model accuracy and model size out of the three and hence it is deployed into the NXP i.MX RT1060 embedded hardware unit for image classification application. Design Space Exploration DNN iMX RT1060 Data augmentation
4	Railway curve squeal: Statistical analysis of train speed impact on squeal noise Asplund, Ruben January 2024 (has links) No description available. ml operation maintenance railway maintenance railway curve machine learning statistics railway train passenger train x60 squeal noise curve squeal squealing noise squeal wheel squeal squealing programming system development computer engineering data visualization data preprocessing industrial computer systems matlab python vibration vibrations vibration signals signals and systems filtering signal processing computer systems communication system infrastructure technology infrastructure track t-test z-test random forest xgboost feature importance speed detection band-pass filter script yaw angle angle of attack decision tree gbdt filter temperature humidity dew point curve radius cloud accelerometers sensors track rail statistical analysis acoustics sound ai artificial intelligence gradient boosting decision trees curve software development drift underhåll järnvägsunderhåll järnvägskurva maskininlärning statistik järnväg tåg x60 skrikljud snäva kurvor kurvradie kurvskrik hjulskrik programmering systemutveckling datateknik data industriella datorsystem matlab python vibration signaler vibrations signaler signaler och system filtrering signalbehandling datorsystem kommunikationssystem infrastrukturteknik infrastruktur spår t-test z-test random forest Södertälje band-pass filter hastighets detektion attackvinkeln anfallsvinkel IMx SKF temperatur fuktighet daggpunkt accelerometer järnvägskurvskrik tåg hastighetens statistisk analys akustik ljud ai kurva kurvor mjukvaruutveckling artificiell intelligens Infrastructure Engineering Infrastrukturteknik Signal Processing Signalbehandling Communication Systems Kommunikationssystem Computer Systems Datorsystem

1

Page generated in 0.0229 seconds