Matching Items (28)
Filtering by
- All Subjects: engineering
- All Subjects: Electrical Engineering
![151779-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-09/151779-Thumbnail%20Image.png?versionId=vFugX3IIK3IQ4mfjdfPrqrv2kKY1GIV0&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240427/us-west-2/s3/aws4_request&X-Amz-Date=20240427T060247Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=2cec332ff52fb0e9ffbbe89ca1cb5ad3f39fe6de7989fa463a97137e9784aa84&itok=a5hlaMqL)
Description
Photovoltaic (PV) module nameplates typically provide the module's electrical characteristics at standard test conditions (STC). The STC conditions are: irradiance of 1000 W/m2, cell temperature of 25oC and sunlight spectrum at air mass 1.5. However, modules in the field experience a wide range of environmental conditions which affect their electrical characteristics and render the nameplate data insufficient in determining a module's overall, actual field performance. To make sound technical and financial decisions, designers and investors need additional performance data to determine the energy produced by modules operating under various field conditions. The angle of incidence (AOI) of sunlight on PV modules is one of the major parameters which dictate the amount of light reaching the solar cells. The experiment was carried out at the Arizona State University- Photovoltaic Reliability Laboratory (ASU-PRL). The data obtained was processed in accordance with the IEC 61853-2 model to obtain relative optical response of the modules (response which does not include the cosine effect). The results were then compared with theoretical models for air-glass interface and also with the empirical model developed by Sandia National Laboratories. The results showed that all modules with glass as the superstrate had identical optical response and were in agreement with both the IEC 61853-2 model and other theoretical and empirical models. The performance degradation of module over years of exposure in the field is dependent upon factors such as environmental conditions, system configuration, etc. Analyzing the degradation of power and other related performance parameters over time will provide vital information regarding possible degradation rates and mechanisms of the modules. An extensive study was conducted by previous ASU-PRL students on approximately 1700 modules which have over 13 years of hot- dry climatic field condition. An analysis of the results obtained in previous ASU-PRL studies show that the major degradation in crystalline silicon modules having glass/polymer construction is encapsulant discoloration (causing short circuit current drop) and solder bond degradation (causing fill factor drop due to series resistance increase). The power degradation for crystalline silicon modules having glass/glass construction was primarily attributed to encapsulant delamination (causing open-circuit voltage drop).
ContributorsVasantha Janakeeraman, Suryanarayana (Author) / Tamizhmani, Govindasamy (Thesis advisor) / Rogers, Bradley (Committee member) / Macia, Narciso (Committee member) / Arizona State University (Publisher)
Created2013
![151827-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-09/151827-Thumbnail%20Image.png?versionId=EubbuHeLBzkOtAySWH0QOttaWPaZpB3K&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240427/us-west-2/s3/aws4_request&X-Amz-Date=20240427T060142Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=a6a6b7087e34495a4d2e0532038cc49f689161bde8fc5792a2454dbbd40d3d03&itok=zoEQ66yY)
Description
The object of this study was a 26 year old residential Photovoltaic (PV) monocrystalline silicon (c-Si) power plant, called Solar One, built by developer John F. Long in Phoenix, Arizona (a hot-dry field condition). The task for Arizona State University Photovoltaic Reliability Laboratory (ASU-PRL) graduate students was to evaluate the power plant through visual inspection, electrical performance, and infrared thermography. The purpose of this evaluation was to measure and understand the extent of degradation to the system along with the identification of the failure modes in this hot-dry climatic condition. This 4000 module bipolar system was originally installed with a 200 kW DC output of PV array (17 degree fixed tilt) and an AC output of 175 kVA. The system was shown to degrade approximately at a rate of 2.3% per year with no apparent potential induced degradation (PID) effect. The power plant is made of two arrays, the north array and the south array. Due to a limited time frame to execute this large project, this work was performed by two masters students (Jonathan Belmont and Kolapo Olakonu) and the test results are presented in two masters theses. This thesis presents the results obtained on the north array and the other thesis presents the results obtained on the south array. The resulting study showed that PV module design, array configuration, vandalism, installation methods and Arizona environmental conditions have had an effect on this system's longevity and reliability. Ultimately, encapsulation browning, higher series resistance (potentially due to solder bond fatigue) and non-cell interconnect ribbon breakages outside the modules were determined to be the primary causes for the power loss.
ContributorsBelmont, Jonathan (Author) / Tamizhmani, Govindasamy (Thesis advisor) / Henderson, Mark (Committee member) / Rogers, Bradley (Committee member) / Arizona State University (Publisher)
Created2013
![151340-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-08/151340-Thumbnail%20Image.png?versionId=PTeUsfMGut6aw3dgu.6ROQwN1WQCpQJR&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240427/us-west-2/s3/aws4_request&X-Amz-Date=20240427T060142Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=929b47d740de106f3f48971e9e66488a8b65236afb08d0e813fc8e2eb0611e20&itok=R5xnwHMX)
Description
Potential induced degradation (PID) due to high system voltages is one of the major degradation mechanisms in photovoltaic (PV) modules, adversely affecting their performance due to the combined effects of the following factors: system voltage, superstrate/glass surface conductivity, encapsulant conductivity, silicon nitride anti-reflection coating property and interface property (glass/encapsulant; encapsulant/cell; encapsulant/backsheet). Previous studies carried out at ASU's Photovoltaic Reliability Laboratory (ASU-PRL) showed that only negative voltage bias (positive grounded systems) adversely affects the performance of commonly available crystalline silicon modules. In previous studies, the surface conductivity of the glass surface was obtained using either conductive carbon layer extending from the glass surface to the frame or humidity inside an environmental chamber. This thesis investigates the influence of glass surface conductivity disruption on PV modules. In this study, conductive carbon was applied only on the module's glass surface without extending to the frame and the surface conductivity was disrupted (no carbon layer) at 2cm distance from the periphery of frame inner edges. This study was carried out under dry heat at two different temperatures (60 °C and 85 °C) and three different negative bias voltages (-300V, -400V, and -600V). To replicate closeness to the field conditions, half of the selected modules were pre-stressed under damp heat for 1000 hours (DH 1000) and the remaining half under 200 hours of thermal cycling (TC 200). When the surface continuity was disrupted by maintaining a 2 cm gap from the frame to the edge of the conductive layer, as demonstrated in this study, the degradation was found to be absent or negligibly small even after 35 hours of negative bias at elevated temperatures. This preliminary study appears to indicate that the modules could become immune to PID losses if the continuity of the glass surface conductivity is disrupted at the inside boundary of the frame. The surface conductivity of the glass, due to water layer formation in a humid condition, close to the frame could be disrupted just by applying a water repelling (hydrophobic) but high transmittance surface coating (such as Teflon) or modifying the frame/glass edges with water repellent properties.
ContributorsTatapudi, Sai Ravi Vasista (Author) / Tamizhmani, Govindasamy (Thesis advisor) / Srinivasan, Devarajan (Committee member) / Rogers, Bradley (Committee member) / Arizona State University (Publisher)
Created2012
![151426-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-08/151426-Thumbnail%20Image.png?versionId=FJp04WNXe1HXNPkZQamM30Vi0EumIFh6&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240427/us-west-2/s3/aws4_request&X-Amz-Date=20240427T060254Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=0bd9391fe8d52d99c18e492415fe2bd57b8f88606cd9ced28c523924b4ffac54&itok=v4v3MwVG)
Description
While the piezoelectric effect has been around for some time, it has only recently caught interest as a potential sustainable energy harvesting device. Piezoelectric energy harvesting has been developed for shoes and panels, but has yet to be integrated into a marketable bicycle tire. For this thesis, the development and feasibility of a piezoelectric tire was done. This includes the development of a circuit that incorporates piezoceramic elements, energy harvesting circuitry, and an energy storage device. A single phase circuit was designed using an ac-dc diode rectifier. An electrolytic capacitor was used as the energy storage device. A financial feasibility was also done to determine targets for manufacturing cost and sales price. These models take into account market trends for high performance tires, economies of scale, and the possibility of government subsidies. This research will help understand the potential for the marketability of a piezoelectric energy harvesting tire that can create electricity for remote use. This study found that there are many obstacles that must be addressed before a piezoelectric tire can be marketed to the general public. The power output of this device is miniscule compared to an alkaline battery. In order for this device to approach the power output of an alkaline battery the weight of the device would also become an issue. Additionally this device is very costly compared to the average bicycle tire. Lastly, this device is extreme fragile and easily broken. In order for this device to become marketable the issues of power output, cost, weight, and durability must all be successfully overcome.
ContributorsMalotte, Christopher (Author) / Madakannan, Arunachalanadar (Thesis advisor) / Srinivasan, Devarajan (Committee member) / Rogers, Bradley (Committee member) / Arizona State University (Publisher)
Created2012
![152336-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-08/152336-Thumbnail%20Image.png?versionId=UpMJVYtfmhhxxV_wbY3pQPCcrcBvJSm4&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240427/us-west-2/s3/aws4_request&X-Amz-Date=20240427T060232Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=7970a291dc908ffe04bea2e2fd5c9069fd3c955ecbd3763a1b48c94aa1dcf10b&itok=Z14eizji)
Description
Flow measurement has always been one of the most critical processes in many industrial and clinical applications. The dynamic behavior of flow helps to define the state of a process. An industrial example would be that in an aircraft, where the rate of airflow passing the aircraft is used to determine the speed of the plane. A clinical example would be that the flow of a patient's breath which could help determine the state of the patient's lungs. This project is focused on the flow-meter that are used for airflow measurement in human lungs. In order to do these measurements, resistive-type flow-meters are commonly used in respiratory measurement systems. This method consists of passing the respiratory flow through a fluid resistive component, while measuring the resulting pressure drop, which is linearly related to volumetric flow rate. These types of flow-meters typically have a low frequency response but are adequate for most applications, including spirometry and respiration monitoring. In the case of lung parameter estimation methods, such as the Quick Obstruction Method, it becomes important to have a higher frequency response in the flow-meter so that the high frequency components in the flow are measurable. The following three types of flow-meters were: a. Capillary type b. Screen Pneumotach type c. Square Edge orifice type To measure the frequency response, a sinusoidal flow is generated with a small speaker and passed through the flow-meter that is connected to a large, rigid container. True flow is proportional to the derivative of the pressure inside the container. True flow is then compared with the measured flow, which is proportional to the pressure drop across the flow-meter. In order to do the characterization, two LabVIEW data acquisition programs have been developed, one for transducer calibration, and another one that records flow and pressure data for frequency response testing of the flow-meter. In addition, a model that explains the behavior exhibited by the flow-meter has been proposed and simulated. This model contains a fluid resistor and inductor in series. The final step in this project was to approximate the frequency response data to the developed model expressed as a transfer function.
ContributorsHu, Jianchen (Author) / Macia, Narciso (Thesis advisor) / Pollat, Scott (Committee member) / Rogers, Bradley (Committee member) / Arizona State University (Publisher)
Created2013
![152931-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-08/152931-Thumbnail%20Image.png?versionId=xqVIXvtMV32UQaI1HbXmgTD2oFGTUsGf&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240427/us-west-2/s3/aws4_request&X-Amz-Date=20240427T060254Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=c85d2dd9ea5d682faf3fa1dc0123a680e13e7902055295d3f94bf3830888eb11&itok=2ia3S-pn)
Description
Today's energy market is facing large-scale changes that will affect all market players. Near the top of that list is the rapid deployment of residential solar photovoltaic (PV) systems. Yet that growing trend will be influenced multiple competing interests between various stakeholders, namely the utility, consumers and technology provides. This study provides a series of analyses--utility-side, consumer-side, and combined analyses--to understand and evaluate the effect of increases in residential solar PV market
penetration. Three urban regions have been selected as study locations--Chicago, Phoenix, Seattle--with simulated load data and solar insolation data at each locality. Various time-of-use pricing schedules are investigated, and the effect of net metering is evaluated to determine the optimal capacity of solar PV and battery storage in a typical residential home. The net residential load profile is scaled to assess system-wide technical and economic figures of merit for the utility with an emphasis on intraday load profiles, ramp rates and electricity sales with increasing solar PV penetration. The combined analysis evaluates the least-cost solar PV system for the consumer and models the associated system-wide effects on the electric grid. Utility revenue was found to drop by 1.2% for every percent PV penetration increase, net metering on a monthly or annual basis improved the cost-effectiveness of solar PV but not battery storage, the removal of net metering policy and usage of an improved the cost-effectiveness of battery storage and increases in solar PV penetration reduced the system load factor. As expected, Phoenix had the most favorable economic scenario for residential solar PV, primarily due to high solar insolation. The study location--solar insolation and load profile--was also found to affect the time of
year at which the largest net negative system load was realized.
penetration. Three urban regions have been selected as study locations--Chicago, Phoenix, Seattle--with simulated load data and solar insolation data at each locality. Various time-of-use pricing schedules are investigated, and the effect of net metering is evaluated to determine the optimal capacity of solar PV and battery storage in a typical residential home. The net residential load profile is scaled to assess system-wide technical and economic figures of merit for the utility with an emphasis on intraday load profiles, ramp rates and electricity sales with increasing solar PV penetration. The combined analysis evaluates the least-cost solar PV system for the consumer and models the associated system-wide effects on the electric grid. Utility revenue was found to drop by 1.2% for every percent PV penetration increase, net metering on a monthly or annual basis improved the cost-effectiveness of solar PV but not battery storage, the removal of net metering policy and usage of an improved the cost-effectiveness of battery storage and increases in solar PV penetration reduced the system load factor. As expected, Phoenix had the most favorable economic scenario for residential solar PV, primarily due to high solar insolation. The study location--solar insolation and load profile--was also found to affect the time of
year at which the largest net negative system load was realized.
ContributorsArnold, Michael (Author) / Johnson, Nathan G (Thesis advisor) / Rogers, Bradley (Committee member) / Ruddell, Benjamin (Committee member) / Arizona State University (Publisher)
Created2014
![153470-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-09/153470-Thumbnail%20Image.png?versionId=1JEarJMs5qD0cNIQPYpSB_CpMMY6apQc&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240427/us-west-2/s3/aws4_request&X-Amz-Date=20240427T060254Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=9af96835da5ca282d3fdd1cf5a5c1d50922e0dbda79de6a1532213dea91b204e&itok=zrhaYdgJ)
Description
Research was conducted to observe the effect of Number of Transparent Covers and Refractive Index on performance of a domestic Solar Water heating system. The enhancement of efficiency for solar thermal system is an emerging challenge. The knowledge gained from this research will enable to optimize the number of transparent covers and refractive index prior to develop a solar water heater with improved optical efficiency and thermal efficiency for the collector. Numerical simulation is conducted on the performance of the liquid flat plate collector for July 21st and October 21st from 8 am to 4 pm with different refractive index values 1.1, 1.4, 1.7 and different numbers of transparent covers (0-3). In order to accomplish the proposed method the formulation and solutions are executed using simple software MATLAB. The result demonstrates efficiency of flat plate collector increases with the increase of number of covers. The performance of collector decreases when refractive index is higher. The improved useful heat gain is obtained when number of cover used is 3 and refractive index is 1.1.
ContributorsSupriti, Shahina Parvin (Author) / Rogers, Bradley (Thesis advisor) / Madakannan, Arunachalanadar (Committee member) / Rajadas, John (Committee member) / Arizona State University (Publisher)
Created2015
![153107-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-08/153107-Thumbnail%20Image.png?versionId=p02xmi.eEvBXqtTamVL5u6BgDU3Lo89A&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240427/us-west-2/s3/aws4_request&X-Amz-Date=20240427T121405Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=ed0a25b6255d04e0b2a81db2d649ba15f8cb2e417daf26caa1cdb69bed853c6a&itok=sSycd6BZ)
Description
To increase the deployment of photovoltaic (PV) systems, a higher level of performance for PV modules should be sought. Soiling, or dust accumulation on the PV modules, is one of the conditions that negatively affect the performance of the PV modules by reducing the light incident onto the surface of the PV module. This thesis presents two studies that focus on investigating the soiling effect on the performance of the PV modules installed in Metro Phoenix area.
The first study was conducted to investigate the optimum cleaning frequency for cleaning PV modules installed in Mesa, AZ. By monitoring the soiling loss of PV modules mounted on a mock rooftop at ASU-PRL, a detailed soiling modeling was obtained. Same setup was also used for other soiling-related investigations like studying the effect of soiling density on angle of incidence (AOI) dependence, the climatological relevance (CR) to soiling, and spatial variation of the soiling loss. During the first dry season (May to June), the daily soiling rate was found as -0.061% for 20o tilted modules. Based on the obtained soiling rate, cleaning PV modules, when the soiling is just due to dust on 20o tilted residential arrays, was found economically not justifiable.
The second study focuses on evaluating the soiling loss in different locations of Metro Phoenix area of Arizona. The main goal behind the second study was to validate the daily soiling rate obtained from the mock rooftop setup in the first part of this thesis. By collaborating with local solar panel cleaning companies, soiling data for six residential systems in 5 different cities in and around Phoenix was collected, processed, and analyzed. The range of daily soiling rate in the Phoenix area was found as -0.057% to -0.085% for 13-28o tilted arrays. The soiling rate found in the first part of the thesis (-0.061%) for 20o tilted array, was validated since it falls within the range obtained from the second part of the thesis.
The first study was conducted to investigate the optimum cleaning frequency for cleaning PV modules installed in Mesa, AZ. By monitoring the soiling loss of PV modules mounted on a mock rooftop at ASU-PRL, a detailed soiling modeling was obtained. Same setup was also used for other soiling-related investigations like studying the effect of soiling density on angle of incidence (AOI) dependence, the climatological relevance (CR) to soiling, and spatial variation of the soiling loss. During the first dry season (May to June), the daily soiling rate was found as -0.061% for 20o tilted modules. Based on the obtained soiling rate, cleaning PV modules, when the soiling is just due to dust on 20o tilted residential arrays, was found economically not justifiable.
The second study focuses on evaluating the soiling loss in different locations of Metro Phoenix area of Arizona. The main goal behind the second study was to validate the daily soiling rate obtained from the mock rooftop setup in the first part of this thesis. By collaborating with local solar panel cleaning companies, soiling data for six residential systems in 5 different cities in and around Phoenix was collected, processed, and analyzed. The range of daily soiling rate in the Phoenix area was found as -0.057% to -0.085% for 13-28o tilted arrays. The soiling rate found in the first part of the thesis (-0.061%) for 20o tilted array, was validated since it falls within the range obtained from the second part of the thesis.
ContributorsNaeem, Mohammad Hussain (Author) / Tamizhmani, Govindasamy (Thesis advisor) / Rogers, Bradley (Committee member) / Srinivasan, Devarajan (Committee member) / Arizona State University (Publisher)
Created2014
![150573-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-08/150573-Thumbnail%20Image.png?versionId=peYZBsKTcj9Mx9OiCxso4WwIVKkN6_KK&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240427/us-west-2/s3/aws4_request&X-Amz-Date=20240427T104131Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=a23798878ba4c95fad326ef8f093b6aac8c85440899bedaefcfb1d5fba0ae5bf&itok=uhjN8hAE)
Description
This report presents the effects and analysis of the effects of Pulsed-Gas Metal Arc Welding's (P-GMAW) on Lean Duplex stainless steel. Although the welding of Duplex and Super Duplex Stainless steels have been well documented in both the laboratory and construction industry, the use of Lean Duplex has not. The purpose for conducting this research is to ensure that the correct Ferrite-Austenite phase balance along with the correct welding procedures are used in the creation of reactor cores for new construction nuclear power generation stations. In this project the effects of Lincoln Electrics ER-2209 GMAW wire are studied. Suggestions and improvements to the welding process are then proposed in order to increase the weldability, strength, gas selection, and ferrite count. The weldability will be measured using X-Ray photography in order to determine if any inclusions, lack of fusion, or voids are found post welding, along with welder feedback. The ferritic point count method in accordance with ASTM A562-08, is employed so that the amount of ferrite and austenite can be calculated in the same manor that is currently being used in industry. These will then be correlated to the tensile strength and impact toughness in the heat-affected zone (HAZ) of the weld based on the ASTM A923 testing method.
ContributorsCarter, Roger (Author) / Rogers, Bradley (Thesis advisor) / Gintz, Jerry (Committee member) / Georgeou, Trian (Committee member) / Arizona State University (Publisher)
Created2012
![153807-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-09/153807-Thumbnail%20Image.png?versionId=D.FqtYxZa9erQrAgrjtGJIKf_HYAVyrW&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240427/us-west-2/s3/aws4_request&X-Amz-Date=20240427T060626Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=832500b72a3e818434096fc1aaa77aeeb509961929382fcb5bc3a72c66c5cc3f&itok=lqXIi6x7)
Description
Brain Computer Interfaces are becoming the next generation controllers not only in the medical devices for disabled individuals but also in the gaming and entertainment industries. In order to build an effective Brain Computer Interface, which accurately translates the user thoughts into machine commands, it is important to have robust and fail proof signal processing and machine learning modules which operate on the raw EEG signals and estimate the current thought of the user.
In this thesis, several techniques used to perform EEG signal pre-processing, feature extraction and signal classification have been discussed, implemented, validated and verified; efficient supervised machine learning models, for the EEG motor imagery signal classification are identified. To further improve the performance of system unsupervised feature learning techniques have been investigated by pre-training the Deep Learning models. Use of pre-training stacked autoencoders have been proposed to solve the problems caused by random initialization of weights in neural networks.
Motor Imagery (imaginary hand and leg movements) signals are acquire using the Emotiv EEG headset. Different kinds of features like mean signal, band powers, RMS of the signal have been extracted and supplied to the machine learning (ML) stage, wherein, several ML techniques like LDA, KNN, SVM, Logistic regression and Neural Networks are applied and validated. During the validation phase the performances of various techniques are compared and some important observations are reported. Further, deep Learning techniques like autoencoding have been used to perform unsupervised feature learning. The reliability of the features is analyzed by performing classification by using the ML techniques mentioned earlier. The performance of the neural networks has been further improved by pre-training the network in an unsupervised fashion using stacked autoencoders and supplying the stacked autoencoders’ network parameters as initial parameters to the neural network. All the findings in this research, during each phase (pre-processing, feature extraction, classification) are directly relevant and can be used by the BCI research community for building motor imagery based BCI applications.
Additionally, this thesis attempts to develop, test, and compare the performance of an alternative method for classifying human driving behavior. This thesis proposes the use of driver affective states to know the driving behavior. The purpose of this part of the thesis was to classify the EEG data collected from several subjects while driving simulated vehicle and compare the classification results with those obtained by classifying the driving behavior using vehicle parameters collected simultaneously from all the subjects. The objective here is to see if the drivers’ mental state is reflected in his driving behavior.
In this thesis, several techniques used to perform EEG signal pre-processing, feature extraction and signal classification have been discussed, implemented, validated and verified; efficient supervised machine learning models, for the EEG motor imagery signal classification are identified. To further improve the performance of system unsupervised feature learning techniques have been investigated by pre-training the Deep Learning models. Use of pre-training stacked autoencoders have been proposed to solve the problems caused by random initialization of weights in neural networks.
Motor Imagery (imaginary hand and leg movements) signals are acquire using the Emotiv EEG headset. Different kinds of features like mean signal, band powers, RMS of the signal have been extracted and supplied to the machine learning (ML) stage, wherein, several ML techniques like LDA, KNN, SVM, Logistic regression and Neural Networks are applied and validated. During the validation phase the performances of various techniques are compared and some important observations are reported. Further, deep Learning techniques like autoencoding have been used to perform unsupervised feature learning. The reliability of the features is analyzed by performing classification by using the ML techniques mentioned earlier. The performance of the neural networks has been further improved by pre-training the network in an unsupervised fashion using stacked autoencoders and supplying the stacked autoencoders’ network parameters as initial parameters to the neural network. All the findings in this research, during each phase (pre-processing, feature extraction, classification) are directly relevant and can be used by the BCI research community for building motor imagery based BCI applications.
Additionally, this thesis attempts to develop, test, and compare the performance of an alternative method for classifying human driving behavior. This thesis proposes the use of driver affective states to know the driving behavior. The purpose of this part of the thesis was to classify the EEG data collected from several subjects while driving simulated vehicle and compare the classification results with those obtained by classifying the driving behavior using vehicle parameters collected simultaneously from all the subjects. The objective here is to see if the drivers’ mental state is reflected in his driving behavior.
ContributorsManchala, Vamsi Krishna (Author) / Redkar, Sangram (Thesis advisor) / Rogers, Bradley (Committee member) / Sugar, Thomas (Committee member) / Arizona State University (Publisher)
Created2015