Matching Items (11)
Filtering by
- All Subjects: engineering
- Genre: Academic theses
- Member of: ASU Electronic Theses and Dissertations
Description
Due to decrease in fossil fuel levels, the world is shifting focus towards renewable sources of energy. With an annual average growth rate of 25%, wind is one of the foremost source of harnessing cleaner energy for production of electricity. Wind turbines have been developed to tap power from wind. As a single wind turbine is insufficient, multiple turbines are installed forming a wind farm. Generally, wind farms can have hundreds to thousands of turbines concentrated in a small region. There have been multiple studies centering the influence of weather on such wind farms, but no substantial research focused on how wind farms effect local climate. Technological advances have allowed development of commercial wind turbines with a power output greater than 7.58 MW. This has led to a reduction in required number of turbines and has optimized land usage. Hence, current research considers higher power density compared to previous works that relied on wind farm density of 2 to 4 W/m 2 . Simulations were performed using Weather Research and Forecasting software provided by NCAR. The region of simulation is Southern Oregon, with domains including both onshore and offshore wind farms. Unlike most previous works, where wind farms were considered to be on a flat ground, effects of topography have also been considered here. Study of seasonal effects over wind farms has provided better insight into changes in local wind direction. Analysis of mean velocity difference across wind farms at a height of 10m and 150m gives an understanding of wind velocity profiles. Results presented in this research tends to contradict earlier belief that velocity reduces throughout the farm. Large scale simulations have shown that sometimes, more than 50% of the farm can have an increased wind velocity of up to 1m/s
at an altitude of 10m.
at an altitude of 10m.
ContributorsKadiyala, Yogesh Rao (Author) / Huang, Huei-Ping (Thesis advisor) / Rajagopalan, Jagannathan (Committee member) / Calhoun, Ronald (Committee member) / Arizona State University (Publisher)
Created2015
Description
A numerical study of incremental spin-up and spin-up from rest of a thermally- stratified fluid enclosed within a right circular cylinder with rigid bottom and side walls and stress-free upper surface is presented. Thermally stratified spin-up is a typical example of baroclinity, which is initiated by a sudden increase in rotation rate and the tilting of isotherms gives rise to baroclinic source of vorticity. Research by (Smirnov et al. [2010a]) showed the differences in evolution of instabilities when Dirichlet and Neumann thermal boundary conditions were applied at top and bottom walls. Study of parametric variations carried out in this dissertation confirmed the instability patterns observed by them for given aspect ratio and Rossby number values greater than 0.5. Also results reveal that flow maintained axisymmetry and stability for short aspect ratio containers independent of amount of rotational increment imparted. Investigation on vorticity components provides framework for baroclinic vorticity feedback mechanism which plays important role in delayed rise of instabilities when Dirichlet thermal Boundary Conditions are applied.
ContributorsKher, Aditya Deepak (Author) / Chen, Kangping (Thesis advisor) / Huang, Huei-Ping (Committee member) / Herrmann, Marcus (Committee member) / Arizona State University (Publisher)
Created2011
Description
Energy production is driven by economic needs, which sometimes results in the environment and wildlife being an afterthought. Unfortunately, many animals are killed as a result of flying too close to wind turbines, and the addition of animal deterrent devices are a promising alternative. This thesis seeks to provide a solution as a part of post- construction considerations regarding wildlife and wind turbine interactions through the introduction of a blade mounted ecological device. After testing the hypothesis, the data revealed the device is effective for increasing power output when placed at the root, middle, and tip of the blade. The middle position yielded the lowest increase at all speeds tested. The device was designed and attached to blades along the estimated line of separation. The blades were then mounted on a tower and tested with wind speed as an input and power as an output. The data was analyzed by fixing speed as a parameter and then looking at the distribution of the power output data. A comparison of blades with and without the device demonstrates a potential for increasing power output by 144% when the device is attached at the blade’s root, 7.5% in the middle, and 21% near the tip. The analysis for this study was descoped due to the constraints of the system to be scaled up. As such, this analysis will hold for turbines with a blade length of no more than approximately eight feet. Blades of this type would be used in single building energy grid supplement turbines or turbines in areas with power requirements of equal or less than 1kW per turbine installed. Single building energy grid supplement turbines are most often mounted to the tops of buildings and take advantage of higher speeds of wind at those heights. As the ecological devices are designed to be similar to vortex generators, which have been tested on large blades, their addition to large blades could prove to have a similar effect.
Keywords: Wind turbine ecosystem, post-construction turbine considerations, wildlife deterrents
Keywords: Wind turbine ecosystem, post-construction turbine considerations, wildlife deterrents
ContributorsBooth, Stephanie (Author) / Trimble, Steve (Thesis advisor) / Middleton, James (Thesis advisor) / Huang, Huei-Ping (Committee member) / Arizona State University (Publisher)
Created2018
Description
Reinforcement learning (RL) is a powerful methodology for teaching autonomous agents complex behaviors and skills. A critical component in most RL algorithms is the reward function -- a mathematical function that provides numerical estimates for desirable and undesirable states. Typically, the reward function must be hand-designed by a human expert and, as a result, the scope of a robot's autonomy and ability to safely explore and learn in new and unforeseen environments is constrained by the specifics of the designed reward function. In this thesis, I design and implement a stateful collision anticipation model with powerful predictive capability based upon my research of sequential data modeling and modern recurrent neural networks. I also develop deep reinforcement learning methods whose rewards are generated by self-supervised training and intrinsic signals. The main objective is to work towards the development of resilient robots that can learn to anticipate and avoid damaging interactions by combining visual and proprioceptive cues from internal sensors. The introduced solutions are inspired by pain pathways in humans and animals, because such pathways are known to guide decision-making processes and promote self-preservation. A new "robot dodge ball' benchmark is introduced in order to test the validity of the developed algorithms in dynamic environments.
ContributorsRichardson, Trevor W (Author) / Ben Amor, Heni (Thesis advisor) / Yang, Yezhou (Committee member) / Srivastava, Siddharth (Committee member) / Arizona State University (Publisher)
Created2018
Description
Laser radars or lidar’s have been used extensively to remotely study winds within the atmospheric boundary layer and atmospheric transport. Lidar sensors have become an important tool within the meteorology and the wind energy community. For example, Doppler lidars are used frequently in wind resource assessment, wind turbine control as well as in atmospheric science research. A Time of Flight based (ToF) direct detection lidar sensor is used in vehicles to navigate through complex and dynamic environments autonomously. These optical sensors are used to map the environment around the car accurately for perception and localization tasks that help achieve complete autonomy.
This thesis begins with a detailed discussion on the fundamentals of a Doppler lidar system. The laser signal flow path to and from the target, the optics of the system and the core signal processing algorithms used to extract velocity information, were studied to get closer to the hardware of a Doppler lidar sensor. A Doppler lidar simulator was built to study the existing signal processing algorithms to detect and estimate doppler frequency, and radial velocity information. Understanding the sensor and its processing at the hardware level is necessary to develop new algorithms to detect and track specific flow structures in the atmosphere. For example, the aircraft vortices have been a topic of extensive research and doppler lidars have proved to be a valuable sensor to detect and track these coherent flow structures. Using the lidar simulator a physics based doppler lidar vortex algorithm is tested on simulated data to track a pair of counter rotating aircraft vortices.
At a system level the major components of a time of flight lidar is very similar to a Doppler lidar. The fundamental physics of operation is however different. While doppler lidars are used for radial velocity measurement, ToF sensors as the name suggests provides precise depth measurements by measuring time of flight between the transmitted and the received pulses. The second part of this dissertation begins to explore the details of ToF lidar system. A system level design, to build a ToF direct detection lidar system is presented. Different lidar sensor modalities that are currently used with sensors in the market today for automotive applications were evaluated and a 2D MEMS based scanning lidar system was designed using off-the shelf components.
Finally, a range of experiments and tests were completed to evaluate the performance of each sub-component of the lidar sensor prototype. A major portion of the testing was done to align the optics of the system and to ensure maximum field of view overlap for the bi-static laser sensor. As a laser range finder, the system demonstrated capabilities to detect hard targets as far as 32 meters. Time to digital converter (TDC) and an analog to digital converter (ADC) was used for providing accurate timing solutions for the lidar prototype. A Matlab lidar model was built and used to perform trade-off studies that helped choosing components to suit the sensor design specifications.
The size, weight and cost of these lidar sensors are still very high and thus making it harder for automotive manufacturers to integrate these sensors into their vehicles. Ongoing research in this field is determined to find a solution that guarantees very high performance in real time and lower its cost over the next decade as components get cheaper and can be seamlessly integrated with cars to improve on-road safety.
This thesis begins with a detailed discussion on the fundamentals of a Doppler lidar system. The laser signal flow path to and from the target, the optics of the system and the core signal processing algorithms used to extract velocity information, were studied to get closer to the hardware of a Doppler lidar sensor. A Doppler lidar simulator was built to study the existing signal processing algorithms to detect and estimate doppler frequency, and radial velocity information. Understanding the sensor and its processing at the hardware level is necessary to develop new algorithms to detect and track specific flow structures in the atmosphere. For example, the aircraft vortices have been a topic of extensive research and doppler lidars have proved to be a valuable sensor to detect and track these coherent flow structures. Using the lidar simulator a physics based doppler lidar vortex algorithm is tested on simulated data to track a pair of counter rotating aircraft vortices.
At a system level the major components of a time of flight lidar is very similar to a Doppler lidar. The fundamental physics of operation is however different. While doppler lidars are used for radial velocity measurement, ToF sensors as the name suggests provides precise depth measurements by measuring time of flight between the transmitted and the received pulses. The second part of this dissertation begins to explore the details of ToF lidar system. A system level design, to build a ToF direct detection lidar system is presented. Different lidar sensor modalities that are currently used with sensors in the market today for automotive applications were evaluated and a 2D MEMS based scanning lidar system was designed using off-the shelf components.
Finally, a range of experiments and tests were completed to evaluate the performance of each sub-component of the lidar sensor prototype. A major portion of the testing was done to align the optics of the system and to ensure maximum field of view overlap for the bi-static laser sensor. As a laser range finder, the system demonstrated capabilities to detect hard targets as far as 32 meters. Time to digital converter (TDC) and an analog to digital converter (ADC) was used for providing accurate timing solutions for the lidar prototype. A Matlab lidar model was built and used to perform trade-off studies that helped choosing components to suit the sensor design specifications.
The size, weight and cost of these lidar sensors are still very high and thus making it harder for automotive manufacturers to integrate these sensors into their vehicles. Ongoing research in this field is determined to find a solution that guarantees very high performance in real time and lower its cost over the next decade as components get cheaper and can be seamlessly integrated with cars to improve on-road safety.
ContributorsBhaskaran, Sreevatsan (Author) / Calhoun, Ronald J (Thesis advisor) / Dahm, Werner (Committee member) / Huang, Huei-Ping (Committee member) / Chen, Kang Pin (Committee member) / Choukulkar, Aditya (Committee member) / Arizona State University (Publisher)
Created2018
Description
First, in a large-scale structure, a 3-D CFD model was built to simulate flow and temperature distributions. The flow patterns and temperature distributions are characterized and validated through spot measurements. The detailed understanding of them then allows for optimization of the HVAC configuration because identification of the problematic flow patterns and temperature mis-distributions leads to some corrective measures. Second, an appropriate form of the viscous dissipation term in the integral form of the conservation equation was considered, and the effects of momentum terms on the computed drop size in pressure-atomized sprays were examined. The Sauter mean diameter (SMD) calculated in this manner agrees well with experimental data of the drop velocities and sizes. Using the suggested equation with the revised treatment of liquid momentum setup, injection parameters can be directly input to the system of equations. Thus, this approach is capable of incorporating the effects of injection parameters for further considerations of the drop and velocity distributions under a wide range of spray geometry and injection conditions. Lastly, groundwater level estimation was investigated using compressed sensing (CS). To satisfy a general property of CS, a random measurement matrix was used, the groundwater network was constructed, and finally the l-1 optimization was run. Through several validation tests, correct estimation of groundwater level by CS was shown. Using this setup, decreasing trends in groundwater level in the southwestern US was shown. The suggested method is effective in that the total measurements of registered wells can be reduced down by approximately 42 %, sparse data can be visualized and a possible approach for groundwater management during extreme weather changes, e.g. in California, was demonstrated.
ContributorsLee, Joon Young (Author) / Lee, Taewoo (Thesis advisor) / Huang, Huei-Ping (Committee member) / Lopez, Juan (Committee member) / Phelan, Patrick (Committee member) / Chen, Kangping (Committee member) / Arizona State University (Publisher)
Created2015
Description
This study uses Computational Fluid Dynamics (CFD) modeling to analyze the
dependence of wind power potential and turbulence intensity on aerodynamic design of a
special type of building with a nuzzle-like gap at its rooftop. Numerical simulations using
ANSYS Fluent are carried out to quantify the above-mentioned dependency due to three
major geometric parameters of the building: (i) the height of the building, (ii) the depth of
the roof-top gap, and (iii) the width of the roof-top gap. The height of the building is varied
from 8 m to 24 m. Likewise, the gap depth is varied from 3 m to 5 m and the gap width
from 2 m to 4 m. The aim of this entire research is to relate these geometric parameters of
the building to the maximum value and the spatial pattern of wind power potential across
the roof-top gap. These outcomes help guide the design of the roof-top geometry for wind
power applications and determine the ideal position for mounting a micro wind turbine.
From these outcomes, it is suggested that the wind power potential is greatly affected by
the increasing gap width or gap depth. It, however, remains insensitive to the increasing
building height, unlike turbulence intensity which increases with increasing building
height. After performing a set of simulations with varying building geometry to quantify
the wind power potential before the installation of a turbine, another set of simulations is
conducted by installing a static turbine within the roof-top gap. The results from the latter
are used to further adjust the estimate of wind power potential. Recommendations are made
for future applications based on the findings from the numerical simulations.
dependence of wind power potential and turbulence intensity on aerodynamic design of a
special type of building with a nuzzle-like gap at its rooftop. Numerical simulations using
ANSYS Fluent are carried out to quantify the above-mentioned dependency due to three
major geometric parameters of the building: (i) the height of the building, (ii) the depth of
the roof-top gap, and (iii) the width of the roof-top gap. The height of the building is varied
from 8 m to 24 m. Likewise, the gap depth is varied from 3 m to 5 m and the gap width
from 2 m to 4 m. The aim of this entire research is to relate these geometric parameters of
the building to the maximum value and the spatial pattern of wind power potential across
the roof-top gap. These outcomes help guide the design of the roof-top geometry for wind
power applications and determine the ideal position for mounting a micro wind turbine.
From these outcomes, it is suggested that the wind power potential is greatly affected by
the increasing gap width or gap depth. It, however, remains insensitive to the increasing
building height, unlike turbulence intensity which increases with increasing building
height. After performing a set of simulations with varying building geometry to quantify
the wind power potential before the installation of a turbine, another set of simulations is
conducted by installing a static turbine within the roof-top gap. The results from the latter
are used to further adjust the estimate of wind power potential. Recommendations are made
for future applications based on the findings from the numerical simulations.
ContributorsKailkhura, Gargi (Author) / Huang, Huei-Ping (Thesis advisor) / Rajagopalan, Jagannathan (Committee member) / Forzani, Erica (Committee member) / Arizona State University (Publisher)
Created2017
Description
The fast pace of global urbanization makes cities the hotspots of population density and anthropogenic activities, leading to intensive emissions of heat and carbon dioxide (CO2), a primary greenhouse gas. Urban climate scientists have been actively seeking effective mitigation strategies over the past decades, aiming to improve the environmental quality for urban dwellers. Prior studies have identified the role of urban green spaces in the relief of urban heat stress. Yet little effort was devoted to quantify their contribution to local and regional CO2 budget. In fact, urban biogenic CO2 fluxes from photosynthesis and respiration are influenced by the microclimate in the built environment and are sensitive to anthropogenic disturbance. The high complexity of the urban ecosystem leads to an outstanding challenge for numerical urban models to disentangling and quantifying the interplay between heat and carbon dynamics.This dissertation aims to advance the simulation of thermal and carbon dynamics in urban land surface models, and to investigate the role of urban greening practices and urban system design in mitigating heat and CO2 emissions. The biogenic CO2 exchange in cities is parameterized by incorporating plant physiological functions into an advanced single-layer urban canopy model in the built environment. The simulation result replicates the microclimate and CO2 flux patterns measured from an eddy covariance system over a residential neighborhood in Phoenix, Arizona with satisfactory accuracy. Moreover, the model decomposes the total CO2 flux from observation and identifies the significant CO2 efflux from soil respiration. The model is then applied to quantify the impact of urban greening practices on heat and biogenic CO2 exchange over designed scenarios. The result shows the use of urban greenery is effective in mitigating both urban heat and carbon emissions, providing environmental co-benefit in cities. Furthermore, to seek the optimal urban system design in terms of thermal comfort and CO2 reduction, a multi-objective optimization algorithm is applied to the machine learning surrogates of the physical urban land surface model. There are manifest trade-offs among ameliorating diverse urban environmental indicators despite the co-benefit from urban greening. The findings of this dissertation, along with its implications on urban planning and landscaping management, would promote sustainable urban development strategies for achieving optimal environmental quality for policy makers, urban residents, and practitioners.
ContributorsLi, Peiyuan (Author) / Wang, Zhihua (Thesis advisor) / Vivoni, Enrique (Committee member) / Huang, Huei-Ping (Committee member) / Myint, Soe (Committee member) / Xu, Tianfang (Committee member) / Arizona State University (Publisher)
Created2021
Description
To date, there is not a standardized method for consistently quantifying the performance of an automated driving system (ADS)-equipped vehicle (AV). The purpose of this dissertation is to contribute to a framework for such an approach referred to throughout as the operational safety assessment (OSA) methodology. Through this research, safety metrics are identified, researched, and analyzed to capture aspects of the operational safety of AVs, interacting with other salient objects. This dissertation outlines the approach for developing this methodology through a series of key steps including: (1) comprehensive literature review; (2) research and refinement of OSA metrics; (3) generation of MATLAB script for metric calculations; (4) generation of simulated events for analysis; (5) collection of real-world data for analysis; (6) review of OSA methodology results; and (7) discussion of future work to expand complexity, fidelity, and relevance aspects of the OSA methodology. The detailed literature review includes the identification of metrics historically used in both traditional and more recent evaluations of vehicle performance. Subsequently, the metric formulations are refined, and robust severity evaluations are proposed. A MATLAB script is then presented which was generated to calculate the metrics from any given source assuming proper formatting of the data. To further refine the formulations and the MATLAB script, a variety of simulated scenarios are discussed including car-following, intersection, and lane change situations. Additionally, a data collection activity is presented, leveraging the SMARTDRIVE testbed operated by Maricopa County Department of Transportation in Anthem, AZ to collect real-world data from an active intersection. Lastly, the efficacy of the OSA methodology with respect to the evaluation of vehicle performance for a set of scenarios is evaluated utilizing both simulated and real-world data. This assessment provides a demonstration of the ability and robustness of this methodology to evaluate vehicle performance for a given scenario. At the conclusion of this dissertation, additional factors including fidelity, complexity, and relevance are explored to contribute to a more comprehensive evaluation.
ContributorsComo, Steven Gerard (Author) / Wishart, Jeffrey (Thesis advisor) / Yang, Yezhou (Thesis advisor) / Chen, Yan (Committee member) / Favaro, Francesca (Committee member) / Arizona State University (Publisher)
Created2022
Description
Machine learning has demonstrated great potential across a wide range of applications such as computer vision, robotics, speech recognition, drug discovery, material science, and physics simulation. Despite its current success, however, there are still two major challenges for machine learning algorithms: limited robustness and generalizability.
The robustness of a neural network is defined as the stability of the network output under small input perturbations. It has been shown that neural networks are very sensitive to input perturbations, and the prediction from convolutional neural networks can be totally different for input images that are visually indistinguishable to human eyes. Based on such property, hackers can reversely engineer the input to trick machine learning systems in targeted ways. These adversarial attacks have shown to be surprisingly effective, which has raised serious concerns over safety-critical applications like autonomous driving. In the meantime, many established defense mechanisms have shown to be vulnerable under more advanced attacks proposed later, and how to improve the robustness of neural networks is still an open question.
The generalizability of neural networks refers to the ability of networks to perform well on unseen data rather than just the data that they were trained on. Neural networks often fail to carry out reliable generalizations when the testing data is of different distribution compared with the training one, which will make autonomous driving systems risky under new environment. The generalizability of neural networks can also be limited whenever there is a scarcity of training data, while it can be expensive to acquire large datasets either experimentally or numerically for engineering applications, such as material and chemical design.
In this dissertation, we are thus motivated to improve the robustness and generalizability of neural networks. Firstly, unlike traditional bottom-up classifiers, we use a pre-trained generative model to perform top-down reasoning and infer the label information. The proposed generative classifier has shown to be promising in handling input distribution shifts. Secondly, we focus on improving the network robustness and propose an extension to adversarial training by considering the transformation invariance. Proposed method improves the robustness over state-of-the-art methods by 2.5% on MNIST and 3.7% on CIFAR-10. Thirdly, we focus on designing networks that generalize well at predicting physics response. Our physics prior knowledge is used to guide the designing of the network architecture, which enables efficient learning and inference. Proposed network is able to generalize well even when it is trained with a single image pair.
The robustness of a neural network is defined as the stability of the network output under small input perturbations. It has been shown that neural networks are very sensitive to input perturbations, and the prediction from convolutional neural networks can be totally different for input images that are visually indistinguishable to human eyes. Based on such property, hackers can reversely engineer the input to trick machine learning systems in targeted ways. These adversarial attacks have shown to be surprisingly effective, which has raised serious concerns over safety-critical applications like autonomous driving. In the meantime, many established defense mechanisms have shown to be vulnerable under more advanced attacks proposed later, and how to improve the robustness of neural networks is still an open question.
The generalizability of neural networks refers to the ability of networks to perform well on unseen data rather than just the data that they were trained on. Neural networks often fail to carry out reliable generalizations when the testing data is of different distribution compared with the training one, which will make autonomous driving systems risky under new environment. The generalizability of neural networks can also be limited whenever there is a scarcity of training data, while it can be expensive to acquire large datasets either experimentally or numerically for engineering applications, such as material and chemical design.
In this dissertation, we are thus motivated to improve the robustness and generalizability of neural networks. Firstly, unlike traditional bottom-up classifiers, we use a pre-trained generative model to perform top-down reasoning and infer the label information. The proposed generative classifier has shown to be promising in handling input distribution shifts. Secondly, we focus on improving the network robustness and propose an extension to adversarial training by considering the transformation invariance. Proposed method improves the robustness over state-of-the-art methods by 2.5% on MNIST and 3.7% on CIFAR-10. Thirdly, we focus on designing networks that generalize well at predicting physics response. Our physics prior knowledge is used to guide the designing of the network architecture, which enables efficient learning and inference. Proposed network is able to generalize well even when it is trained with a single image pair.
ContributorsYao, Houpu (Author) / Ren, Yi (Thesis advisor) / Liu, Yongming (Committee member) / Li, Baoxin (Committee member) / Yang, Yezhou (Committee member) / Marvi, Hamidreza (Committee member) / Arizona State University (Publisher)
Created2019