REU Projects

REU students have the opportunity to perform research with UT faculty while utilizing CURENT’s world-class resources. Specific research projects CURENT REU students have done can be viewed below:

2016 REU Projects

Adam Foshie:

“Data Acquisition and Analysis for Probabilistic Residential Demand Response Modeling”

When working on many applications concerning the power grid, creating simulation tools to model the kind of data that engineers will be working with is very important. To create these tools correctly so that they are accurate and effective, one must gather, study, and understand a large sample of real data relevant to the situation. One way to do this is to create a well written survey that polls a large reliable pool of participants on information relevant to the cause. The objective of this research is to create a survey that will bring in the appropriate information needed to run an effective simulation tool to show a probabilistic model of power demand in residential loads. Creating questions that gather this data effectively will give insight on the usefulness of Demand Response programs in the residential sector. Furthermore, the survey also aims to gather the participants’ motives as to why they would be interested in taking part in types of Demand Response programs. This will give an understanding of how much economic and environmental factors play into a participant’s willingness to save energy. The work that will be done includes researching heavy power consumers in a residential setting, finding what changes in a residence results in the most effective DR strategy, and coordinating with other students/faculty to suggest and mold the right questions needed to yield the data that we will need.

REU Poster     REU Paper

Andrew Wittenburg:

“Energy Disaggregation Using Convolutional Sparse Coding”

Research has shown that consumers given an itemized report of their power usage reduce their future usage by a significant amount. However, the cost of infrastructure necessary to measure plug level power usage in every home and office would be prohibitive. With the advent of smart meters, power companies can now monitor the aggregate power of the house nearly once per minute. Through the use of signal processing and pattern recognition, this total power can be disaggregated to determine the usage of individual devices and even components within these. This is known as non-intrusive load monitoring. Current techniques include nonnegative matrix factorization and neural networks. In addition to consumer feedback, this process can provide useful information for appliance designers and can determine consumer trends.

REU Poster

Carley Washington:

“Power Flow with HVDC lines”

HVDC systems can be a realistic economic and technical alternative to AC technology in some cases. It is the solution for power transmission over very long distances, via cables, between networks that operate at different frequencies, and for additional power without increasing the short circuit ratio of the AC system. The objective of this project is to study the power flow in AC systems with HVDC lines. This includes researching advanced circuit analysis and power flow formulations, learning power system simulation tools such as PSS/E and PSAT, and building steady-state power system models. An IEEE 14-bus system will be used to construct a HVDC incorporated system having different control methods to study the impact of HVDC lines on power flow.

REU Poster     REU Paper     REU Presentation

Clarence Jackson II:

“Data Streaming Client Development in SIMULINK for LTB Real-time Application”

In order to predict grid behavior, it is useful to have a real time simulator able to model areas of the grid. The Large Scale Testbed is the software tool used simulate large regions of the power grid, where the ePHASORsim-based real-time platform is capable of simulating power grid in SIMULINK. The mechanism the different modules use to communicate is the DiME streaming server, and it is crucial to stream data from SIMULINK so that the real-time simulation can be analyzed and visualized. The objective of this research is to develop a MATLAB s-function that will allow for SIMULINK model communication via the dime server. With this the communication capabilities of the large scale testbed will be extended to allow for SIMULINK model communication.

REU Poster

John  Curtin:

“Fault Simulation for Hardware Emulation”

My personal project is to accurately simulate the effects of a shunted impedance across the lines of a polyphase transmission line network. Specifically, the simulation will show the effects of this impedance on the transmission lines of a power grid. A common example of this phenomenon is a tree branch falling on a set of power transmission lines. Ideally, this simulation will be able to be implemented on the Hardware Test Bed (HTB) to increase its flexibility. However, implementation on the HTB will be done only if time permits, either in the form of a hardware module or modification of the existing hardware code.

REU Poster     REU Paper     REU Presentation

Jessica Wert:

“Analysis & Visualization of Frequency in Power Systems”

In order for the power grid to function properly, the frequency must be kept within an acceptable band. This is achieved by maintaining the balance between the load and generation in the system. With increased penetration of renewable energy into the grid, the frequency of the system suffers greatly as there is a relative reduction of inertia and the dynamic performance of the system is affected. Fast acting energy storage systems provide a solution to improve the dynamic performance of the system as they can provide energy at a faster rate when demanded. The distribution of inertia will determine the dynamic performance of the system and will depend greatly on the penetration of renewables into the grid. This project aims to study the distribution of inertia in the system by means of creating a visualization tool which would help identify the center of inertia and other points which would be useful to determine the location for the deployment of fast acting energy storage devices with the increased utilization of renewable energy.

REU Poster     REU Paper     REU Presentation

Kyle Goodrick:

“Automatic Device Characterization”

In almost all power electronic applications the most important element of the circuit is the switching device. The characteristics of the device have a huge impact on the circuit and control most aspects of the circuits performance. There are currently tens of thousands of devices being sold on the market and more are being released every day. As the characteristics of these devices is so critical to circuit performance, knowing the characteristics of each device is incredibly important; however, with the large number of devices currently on the market testing by hand is a very slow and inefficient process. Automated device characterization, partnered with a database of device characteristics allows us to rapidly test and compare a number of devices to find the best fit for a particular design. The goal of this research is to create a standardized system that allows us to quickly test a large number of devices.

REU Poster

Eric Abreut:

“Time Series Based Semi-Analytical Solution of Power Systems and its Application in Direct Methods”

A power system consisting of any number of generators is considered stable if it can reach an equilibrium point after a disturbance. Each real power system can tolerate a certain amount of disturbance while it can lose its stability under some large disturbances. In order to access the system stability and produce a faster response to the problem, there must be something that monitors the system ahead of real time. This project focuses on finding a semi-analytical solution to power system differential equations, where the solution is assumed to follow a certain explicit polynomial in time with unknown coefficients. It is expected that those unknown coefficients can be obtained by solving a set of linear equations. Once solved, it will be possible to determine if the power system will lose its stability following a disturbance and allow for appropriate measures to be taken before the system completely loses its stability.

REU Posters     REU Paper     REU Presentation

Juan  Avalos:

“MOSFET Characteristics at Cryogenic Temperatures”

Throughout the power electronics era, Wide Bandgap semiconductors have become superior to regular silicon semiconductors.  Their switching characteristics allows for smaller sizes and greater efficiency with the capability to tolerate higher voltage. These improvements lead to the focus on the development of superconducting power electronics. Cryogenic operation, around 77 K, of power semiconductor devices can result in improved switching speed and lower on-state voltage than at room temperature.  The switching characteristics of the power MOSFETs are obtained using the cryogenic chamber and the Double Pulse Test (DPT).  The DPT is simply manually triggering two pulse to the gate input.  The falling edge of the first pulse and rising edge of the second pulse corresponds to the turn off and turn on switching transients respectively. This results in great potential for cryogenic power conversion where size and weight are large considerations.  Ultimately, the goals for this research are to construct a smaller and more efficient DC-AC inverter for Boeing&NASA aircraft.

REU Poster     REU Paper     REU Presentation

Kimberly Glasser:

“Smart Home Energy Management System”

Energy management is a significant focus in modern technological research.  Household energy usage can be decreased by allowing the customer to control certain appliances with the Home Management System (HMS) testbed.  My project goal is to continue research from previous semesters, by improving the ability of the current available Home Management System (HMS) testbed.  Previously, the Home Management System (HMS) testbed allowed certain appliances to be always turned on, always turned off or to be turned on and off depending on the price per kilowatt hour ($/kWh).  If the price is above the user defined price, then the appliance is turned off.  Once the price falls below the user defined price the appliance is turned back on.  This allows the customer to choose the level of participation in the energy management program without an abundance of additional effort.  Specifically, my goal is to allow a small water heater to not only be turned on and off, but also on low, medium and high settings.  This will allow the customer more customization options when using the Home Management System (HMS) testbed.

REU Poster     REU Paper     REU Presentation

LeAnn Thompson:

“HVDC Dynamic Modeling Using A 9-Bus System”

A direct current (DC) system that operates at high voltage uses are more efficient than an ac system. HVDC allows for easy transfer of power between grids that are operating at different frequencies and that are long distances apart. The intent of this research is to do studies on these HVDC lines and to determine if they improve dynamic performance. In attempting to do so, tools such as PSS/E and a MATLAB based tool, PSAT, will be used to construct a multi-terminal HVDC model on an IEEE 9-bus system. While studying this system, a comparison will be done to see whether using HVDC lines are more efficient as opposed to the lines in the original system.

REU Poster     REU Paper     REU Presentation

Michael Breuhl:

“Wind Speed Modeling on the Seconds Scale”

My project is about studying the impact of wind speed variability on the dynamics of the system. In order to do so, we must create a model for the wind speed associated with seconds. Currently there are models for the hours to months, but they will not work for our purposes. We plan to use autoregressive techniques in order to model the wind behavior. Once we have the model, we can see how the system will react to the transient stability of the system or the impact of power variability on the scale of seconds.

REU Poster

Rushi  Patel:

“Loss Model for Gallium Nitride DC-DC Buck Converter”

In recent years, the use of Gallium Nitride (GaN) converters increased because of its unique physical properties that makes it much more reliable and efficient than silicon converters. The main goal of this project is to derive a loss model that can accurately predict the efficiency of GaN respect to output current of the DC-DC buck converter. This project also investigates the constant variables which help realize the difference between theoretical and experimental efficiency of GaN respect to output current.

REU Poster     REU Paper     REU Presentation

Ryan  Fraser:

“Load Hiding to Preserve Privacy from Smart Meter Measurements”

In order to increase the efficiency, fault detection, and forecasting of the power grid, smart grid technology is rapidly being implemented throughout the power system. However, improved monitoring of individual household power usage has created potential privacy issues. An analysis of a household smart meter’s data is capable of generating detailed depictions of the activities of the homeowners within their own homes. This project seeks to identify a balance between producing useful information to utilities while also providing protections to individual consumers. The result will be a system that masks power consumption data from analysis on an individual household level while simultaneously not affecting aggregate power consumption data.

REU Poster     REU Paper     REU Presentation

Shashwat Sitesh:

“LABVIEW Implementation of Central Controller for Active Power Load Balancing”

In power systems, there are sometimes sudden changes in PV power availability that can place undue loads on backup battery power supplies. In this system, PV and load references are planned every minute and monitored every two seconds and updated when the power needed exceeds the backup battery power. The overarching goal is to ensure that this does not happen. Before the system is implemented, it has to be modeled by local hardware. Currently, the central controller has been implemented in MATLAB. The goal of this research is to implement the central controller in LABVIEW because it has a better hardware interface than MATLAB.

REU Poster     REU Paper     REU Presentation

Tatianne Da Silva:

“Analysis of Variability of Solar Panels in The Distribution System”

In analyzing a distribution feeder, it is possible to see that the voltage varies when compared at different points in the power system. This variation is not ideal, since a constant voltage has to be maintained. With the introduction of solar panels into the power grid, the deviation increases since solar panels inject power into the system. The objective of this research is to observe the voltage in different points in the power system and compare it with and without Static VAR Compensator devices (SVC) in the system, that will try to control the reactive power. The result should show that implementing SVC devices in the distribution system will create a more resilient distribution system to weather conditions.

REU Poster     REU Paper     REU Presentation

2015 REU Projects

Abigail Teron and Allan Bartlett:

“The Comparison of Approximations of Nonlinear Functions Combined with Harmonic Balance Method for Power System Oscillation Frequency Estimation”

Analysis of Power System Oscillation Modes Using Differential Groebner Basis, Function Approximation and the Harmonic Balance Method:Accurately identifying the existing oscillation modes is helpful for analyzing system stability. In this project, the scalability of Differential Groebner Basis will be studied. The Differential Groebner Basis-Harmonic Balance Method (DiffGB-HBM) approach will be applied to a 2-machine system and a 3-machine system to validate its accuracy. Other Function Approximation methods such as Taylor Expansion will also be studied to compare with DiffGB-HBM approach.

REU Poster    REU Paper   REU Presentation

Shivam Patel:

“Optimization of Power Flow Through Electric Vehicle Charging”

In order for charging companies to make the most profit from electric vehicle charging, costs and profits must be optimized. One way of doing so is by optimizing the power flow. For a charging station, power is either flowing in or out. Thus, power is being purchased or sold. The objective of this research is to optimize the profits made from this process. To do so, the average power demanded on a given day over time intervals is necessary in this research. Then each assumed electric vehicle is assigned a charging level and a state of charge using proper probability distributions. Once finding and calculating all necessary values, the optimization function is solved under proper constraints, allowing the retrieval of an optimized power flow. The result will depict an optimum power flow over a twenty-four hour time span.

REU Poster    REU Paper     REU Presentation

Chris Matthews:

“Modelling the NPCC System with 12.5% Wind Penetration”

Using AGC to Control Grid Frequency:This project focuses on the design of Automatic Generation Control systems to prevent blackouts and maintain grid frequency in the case of unexpected event.

REU Poster    REU Paper    REU Presentation

Pedro Rivera:

“Thermal analysis and PCB Design for a GaN Power Transistor”

Thermal Analysis and PCB Design for a Power Switching Transistor: This investigation sets out to simulate the thermal effects of the GaN system enhancement mode power switching transistors on the PCB and how effective is the heat sink configuration. It also looks to see if a different heat sink configuration works better and how these simulations compare to a real world model.

REU Poster    REU Paper    REU Presentation

Stephen Tang:

“Neural Network Recognition of Frequency Disturbance Recorder Signals”

ANN on FDR Classification: Explores use of artificial neural network to identify source of a signal from a set of candidate frequency disturbance recorders.

 REU Poster    REU Paper    REU Presentation

Mark Nakmali:

“High Efficiency Power Converters with Gallium Nitride Transistors”

Enhancement Mode Gallium Nitride Converters: The dominant power converter technology today utilizes Silicon; however, the use of Gallium Nitride (GaN) converters have been on the rise. This is because the physical properties of GaN make it possible for it to be smaller and more efficient than Silicon, while still operating at the same parameters. As GaN technology becomes more sophisticated and cheaper to make, the availability of it will increase, which will eventually make it a tough competitor for Silicon. The goal of this project is to determine the efficiency of GaN in a step down converter and compare it to the existing data of Silicon converters.

REU Poster    REU Paper    REU Presentation

Hayden Dahmm:

“Optimization of Distributed Storage Locations and Reactive Power Supply for Improved Voltage Variation”

Maeve Lawniczak

“Increasing Inductor Power Density Using Controllable Electropermanent Magnets”

REU Poster    REU Paper

Michael Breuhl

“Flywheels and Power System Stability”

REU Poster    REU Paper    REU Presentation

Parker Diamond & Austin McEver

“Development of a Distributed MATLAB Environment with Real-Time Data Visualization”

REU Poster    REU Paper    REU Presentation

Joseph Estrada

“Load Disaggregation Using Non-negative Matrix Factorization”

REU Poster    REU Presentation

2014 REU Projects

Jared Baxter:

“Renewable Energy and Power Electronics Demonstration”

For my research project, I am making a display using a slot car track that shows how power electronics work, and the impact that they have in electrical devices. The display will also incorporate energy harvesting through PV panels and wind turbines to power the cars. I will have two slot cars racing each other, one will be powered by a set of PV panels and a wind turbine that is run through power electronics. The other car will be run by a set of identical PV panels and a wind turbine except, combining the power sources will not incorporate power electronics. This interactive display will be used by UTK and CURENT to introduce high school students and other visitors to some of the different aspects of electrical engineering. The display can also be used by UTK professors to introduce power electronics and the basic principles of electrical engineering to students in various courses.

REU Poster

Doug Bouler:

“Amplifier Design and Modeling”

For my REU project I will be researching the application of near-field and far-field RF for future biomedical applications. With the rise of three-dimensional organ printing, devices will need to be designed in order to monitor the function and stability of these new organs. In order for these devices to be of long-term use, surgery must be required or the devices will need some type of extensive power system. The depth of the transplant, the invasiveness involved, and the mobility required all influence the method of supplying power. In the past, radioactive batteries have been used to provide power to biomedical implants, but due to the stigma and stiff regulations that come with radioactive materials, these batteries are rarely used. Lithium-ion batteries, glucose batteries, and other enzymatic batteries are becoming longer-lasting, but still require surgery every eight to ten years in order to replace them. Capacitive charging allows for power to flow very directly across two metal plates, but heat very quickly and the Specific Absorption Rate (SAR) involved is dangerously high. Inductive charging is very efficient given that the coils are ideally aligned with each other. For biomedical applications, this means that the patient must be subjected to wear a battery pack that aligns the coils, restricting mobility, or mustrisk the inefficiency of having a coil further away or not ideally aligned. Far-field RF charging allows patients to move freely while being subjected to minimal excess energy. Efficiency is much less than that of the other forms of wireless power, but greater mobility is allowed and fewer inconveniences, such as moving a large coil or a metal plate, exist. Furthermore, due to the sheer amount of transplants that will be able to occur after FDA approval, the introduction of wireless power reduces the number of surgeries post-transplant that the patient must endure. In short, wireless power increases the longevity of biomedical devices, increases patient mobility, and reduces the number for invasive surgeries.

REU Poster    REU Paper    REU Presentation

Hayden Dahmm:

“An Analysis of Residential Demand Response Design Potential from Consumer Survey Data”

With increased social dependence on electric power, the disparity between consumer demand and electricity supply is a significant issue. This disparity can lead to greater inefficiency and greenhouse gas emissions, and it increases the possibility of system failures. Rising demand is traditionally addressed through Supply Response, in which further supply is added to the market, possibly to the detriment of providers. Improved communication technologies now allow for Demand Response (DR) programs, in which consumers are encouraged or compelled to change electric loads; the goal is typically to shift loads away from times of peak demand to when demand is less, thereby reducing market volatility. Theoretical studies and real world implementations demonstrate the value of DR to smart grid design.  Further research is required to improve program structure.This project considers application of DR to the residential market where the program has not been widely explored. One objective is to estimate the price elasticity of household appliances. With these results, the incentives needed to change consumer behavior will be calculated. Surveys regarding residential power consumption were administered in 2013 and 2014 and the results will be analyzed for demographic patterns.  Using MATLAB, a stochastic model representing varying electricity demand to 100,000 households will be created. This simulation will model the real world residential market and different incentive based demand response programs will be implemented. Additionally, this project will look to segment residential customers based on their consumption habits, and responsiveness to different incentives. The results of the study could help in creating customized demand response programs that maximize consumer involvement and improve system reliability

REU Poster    REU Paper    REU Presentation

Joe  Allen:

“Finding the Hidden Scenes Behind Android Applications”

The research in this project  will assist in finding the hidden scenes behind description files for Android applications.  The research is broken into two parts, getting familiar with state-of-the-art web crawling techniques and classifying the data that was collected.A web crawler is a program that browses the internet in a logical and automatic manner. Web crawlers can be used for  several applications, such as web scraping and web indexing.  The web crawler used in this project will be mainly focused on web scraping. Mainly, scraping application descriptions from the Android Play Store.  Since our web crawler will be focused on solely crawling one page the politeness policies, selection policies, and revisitation policies  will need to be highly specialized.  After the data is  collected it will be stored in a SQL database to be analyzed. When the collection of data is completed, a statistical model to discover the abstract topics that occur in the descriptions will be created.  This is known as a topic model in machine learning and natural language processing. Specifically, this project will focus on the Author-Topic Model, which extends the Latent Dirichlet Allocation (LDA) to include author information.

REU Poster    REU Paper    REU Presentation

Philip Wolfe:

“Non-intrusive Energy Disaggregation using Signal Unmixing”

As energy conservation is becoming a more and more important topic in today’s society,focus on offering increased knowledge to consumer can raise consumer awareness to help increase energy conserving behavior. With non-intrusive load monitoring, sparse coding algorithms can be used to disaggregate total power consumption of a house into the individual devices’ consumption. This project uses structured prediction, for discriminatively training sparse coding algorithms specifically to maximize disaggregation performance. A sparse coding algorithm is used to learn a model of each device’s power consumption over a typical week, then these learned models are combined to predict the power consumption of different devices in unknown homes, using only their aggregate signal. Unlike previous studies that focus on classification and use artificial lab environments with a small number of devices of the same set, this project aims to disaggregate and uses readily available data from smart meters sampled at an hourly sampling rate. Focus is placed on the generalization ability of algorithms for devices and homes unknown at training time.

REU Poster    REU Paper    REU Presentation

Anthony Perez:

“Distributed Photovoltaic Generation Emulation in Converter Based Power Grid Emulation System”

I am working on the Hardware Test Bed (HTB) project. HTB is known as a transmission network emulator and it is designed to emulate a large-scale power system using interconnected reprogrammable converters and a reconfigurable structure. The difference between a computer simulation and a simulation in the HTB is that the system will be capable of having real-time experiments; also it is possible to integrate protection, control and real-time communication in the system. The main task of my project is to emulate a component of a microgrid and connect it to a 3-ph inverter that is located in the HTB structure. The project is divided in three parts. The first part is to learn about the structure and configuration of the HTB. The second part is to design the component of a microgrid by using the software Matlab, specifically the SimPower System section. The third part will be to emulate the system in the HTB’s lab. During the first part of my project I will be looking if there is any difference between the integration of distributed PV generation and centralized PV generation into the HTB configuration. If there is a significant difference, then I will be emulating distributed generation of PV into the HBT configuration. If not, then I will check the possibility of emulating a diesel generator or the construction of a microgrid itself using the HTB configuration.

REU Poster    REU Paper    REU Presentation

Mark Nakmali:

“Thevenin Equivalent Estimation for Voltage Instability Prediction”

Keeping a close eye on the parameters of an electrical power system is very important because during times of high stress on the system, the risk of voltage collapse increases. This is when the demand is greater than the system is able to output, causing the system to draw on reactive power until it is depleted, causing the system to fail. Normally by watching the decrease in voltage, one can determine if the system is near voltage collapse due to the shape of power versus voltage curve (PV-curve), but with the addition of capacitor banks used to increase transfer capabilities, the PV-curve has less of a voltage drop causing it to become increasingly harder to calculate the system’s proximity to voltage collapse. In order to solve this problem, a method to estimate the Thevenin Equivalent of the system as accurately as possible must be calculated. The purpose of this research project is to explore two methods of estimating the Thevenin Equivalent. These methods are the Least Square Approach and the Kalman Filter Approach. By using computer software, these mathematical methods that will take data from previous measurements, place them in an equation or algorithm, and output the estimation, afterwhich the quality of the output of each method will be compared.

REU Poster    REU Paper    REU Presentation

Abigail Teron:

“Designing a Model to Obtain Residents’ Response for the Financial Incentives in a Demand Response Program”

As a smart home project; this project is based on the design of a model residents’ response to the financial incentives in demand response. By promoting demand response it can be showed that is an important way to make power system more efficient. The retail by price of the electric power consumption will be more accepted by the costumers if the price is decrease in a minimum, considering decreasing the load. The model will be design for the electric power company utility that involves the generation, transmission and the distribution of energy that is going to be sold. This model will predict residents’ response; it will set out more efficient programs for different areas by generating the amount that utility needs to pay.

REU Poster    REU Paper    REU Presentation

Lauren Atwell:

“Three Phase Induction Motor Dynamic Modeling and Behavior Estimation”

Induction motors are alternating current (AC) motors that converter electrical power into mechanical power through electromagnetic induction. Single phase induction motors are used for small loads such as washing machines, air conditioner compressors, and other small household appliances. The most common three-phase induction motors are squirrel cage motors. These motors are widely used for industrial applications because they are stable, rugged, reliable, and relatively inexpensive.Induction motors are complex electromagnetic devices, and the speed of load of these systems should be designed to adapt to different working environments. A closed-loop control motor drive is one type of popular motor drive technology that allows the system to adapt based on the feedback from the load. The purpose of this project is to build a system in Simulink that will simulate a closed-loop motor drive to accurately control speed of a squirrel cage induction motor.

REU Poster    REU Paper    REU Presentation

Casey O’Leary:

“Dynamic Ringtone Adjustment using Audio Sampling on Android Smartphones”

Though people often have their cell phones with them at all times, there are many situations where it is not appropriate for a cell phone to ring or the ringing of a cell phone may not be noticeable due to the surrounding environment. Through research, we have implemented an application and accompanying modifications to the Android Operating System Kernel to implement a system which intelligently chooses a ringtone for the phone based on the characteristics of the surrounding environment. This is done through the manipulation of both the volume of the ringtone and by changing which ringtone is actually played.

REU Poster    REU Paper    REU Presentation

Runsha Long:

“Minimize Total Power Loss in Distribution Network Reconfiguration Considering PEV Delayed Charging Strategy”

The goal of this research project is to develop an open-source software based smart home with plug and play and grid-to-home communication capability. Smart home systems have experienced an increase in popularity with the new generation of home appliances, entertainment systems, and security systems. Current smart home systems are extremely privatized, such as the Samsung Smart Home which is designed to work exclusively with Samsung products. The intent of this project is to eliminate the restrictions of privatized smart home technology, and rely on an open-source software based smart home with a plug and play feature that allows for a home management system to be extended to any household electronic device. On top of developing an extensible smart home, this project considers the potential of grid-to-home communication, which includes real time pricing (RTP), grid status update, household energy usage data consolidation, etc. Grid-to-home communication enables the interaction between distributor and consumer on a real-time basis, which results in better energy savings, peak load shaving, and allows for statistical estimates of regional power usage. There will be two stages to this research. The first stage will be dedicated to modeling the plug and play capability onto the pre-existing Smart Home Energy Management System (SHEMS) and controlling it with the open-source software, openHAB. Arduino Yun will serve as the plug and play communication node between household electronics and the server running openHAB. The second stage will test SHEMS within a simulated grid environment and incorporate configurable algorithms that optimize the system based on RTP and grid status. The aim of this simulation is to create an effective model from which data could be analyzed to form effective strategies that maximizes the efficiency of power grids.

REU Poster    REU Paper    REU Presentation

Stanly Mathew:

“An Analysis of Residential Demand Response Design Potential from Consumer Survey Data”

With increased social dependence on electric power, the disparity between consumer demand and electricity supply is a significant issue. This disparity can lead to greater inefficiency and greenhouse gas emissions, and it increases the possibility of system failures. Rising demand is traditionally addressed through Supply Response, in which further supply is added to the market, possibly to the detriment of providers. Improved communication technologies now allow for Demand Response (DR) programs, in which consumers are encouraged or compelled to change electric loads; the goal is typically to shift loads away from times of peak demand to when demand is less, thereby reducing market volatility. Theoretical studies and real world implementations demonstrate the value of DR to smart grid design.  Further research is required to improve program structure.This project considers application of DR to the residential market where the program has not been widely explored. One objective is to estimate the price elasticity of household appliances. With these results, the incentives needed to change consumer behavior will be calculated. Surveys regarding residential power consumption were administered in 2013 and 2014 and the results will be analyzed for demographic patterns.  Using MATLAB, a stochastic model representing varying electricity demand to 100,000 households will be created. This simulation will model the real world residential market and different incentive based demand response programs will be implemented. Additionally, this project will look to segment residential customers based on their consumption habits, and responsiveness to different incentives. The results of the study could help in creating customized demand response programs that maximize consumer involvement and improve system reliability

REU Poster    REU Paper    REU Presentation

Chad Harley:

Chad Harley’s current work includes the construction of a multi-FPGA system that utilizes peer-to-peer communication to model power systems as a state space. A state space model is mathematical representation of a system which describes the dynamic behavior of the system through coupled first-order differential equations.
Modeling has become an essential aspect of engineering in that it allows us to observe    and understand the attributes of dynamic systems without the high cost of experimentation, concern for safety, or risk of hardware damage. The developed program will be used for testing and benchmarking control systems and future work will also allow for hardware-in-the-loop (HIL) testing via input/output modules. In order to accomplish the aforementioned tasks LabVIEW is used in conjunction with several pieces of National Instruments hardware. This hardware includes the National Instruments PXIe-8135 chassis and two PXIe-7966R FPGA boards.

Stephan Tang:

“FNET Error Detection with Frequency Based Trigger”

Stephen’s research seeks to devise a method to detect faulty frequency disturbance recorder (FDR) units within the frequency monitoring network (FNET) based upon the real time data that said units report. FDRs are distribution level sensors that measure voltage magnitude, phase angle, and frequency in order to provide controllers with an idea of the state of the power grid. The issue of faulty units is important because the FNET alert system is designed to respond immediately to events that affect the power grid such as generation trips, load shedding, and islanding. The approximate locations of events may also be estimated from FDR data, enabling the system to alert the relevant operators to prevent further damage to the power grid. These triggers are caused by real time processing on the FDR data such as extensive deviation of interconnected units or large enduring changes in frequency. Faulty FDR units may cause the system to incorrectly trigger, making an automated FDR verification system a necessity. Despite the fact that FDR measurements within the same interconnection are generally very synchronized, the problem of finding faulty units is more complicated than extracting the inconsistent units. Events such as islanding also cause FDR units have different measurements, making it ambiguous whether such differences in the data are caused by faulty sensors or not. The required research involves fields such as sensor fault/bad data detection. It will ideally conclude in a simple trigger design capable of distinguishing between the two that is practical to implement in real time.

REU Poster    REU Paper    REU Presentation

2013 REU Projects

Casey O’Leary:

“Network Time Protocol Time Accuracy Compared to GPS Clock on Mobile Platforms for Frequency Measurement of AC Power Source”

Casey’s research looked to determine whether or not a Network Time Protocol clock  synchronization could be used on a mobile platform for frequency measurement of an AC power source by comparing an NTP clock to a GPS clock.

REU Poster    REU Presentation

Cory Raszeja:

“Hardware/Firmware Implementation of a Smart Home Energy Management System”

Cory’s research proposed an improved & redesigned prototype of a previous version of a Smart Home Energy Management System (SHEMS). The updated design implemented a more compact board while including an important failsafe & better overall functionality.

REU Poster    REU Presentation

Geneva Doak:

“An Improved Test Set-up for Cell Impedance Measurement System”

Geneva’s Research proposed an improved test set-up for cell impedance measurement system by exploring two methods of measurement, using a lock-in amplifier and a DAQ card.

REU Poster     REU Presentation

Jae Jang:

“Optimization of Electricity Market:Demand Side Management”

Jae’s research provided an insight of the benefits in Demand Side Management by shifting the load from peak hours to non-peak hours. It also evaluated the different LMP (Locational Marginal Pricing) characteristics for different regions and times.

REU Poster     REU Presentation

Jason Chan:

“Simulation of a Smart Home Energy Management System with Dynamic Price Response”

Jason’s research looked to simulate smart home energy management system with dynamic price response using MATLAB with common household loads such as an Electric Water Heater and the charging station for an Electric Vehicle.

REU Poster     REU Presentation

Leonardo Bernal:

“Wide-area Power System Frequency Visualization and Oscillation Monitoring”

Leonardo’s research geared towards developing a MATLAB visualization program that could identify the center of oscillation (CoO) for power system.

REU Poster     REU Presentation

Matthew Lambert and Alfredo Bermudez:

“Emulator for Energy Management System in order to Test New Control Strategies”

Matthew and Alfredo’s research proposed to develop an emulator for energy management system and test new control strategies for load frequency problems.

REU Poster     REU Presentation

Saajid Haque: 

“Impact of Large-Scale PV Penetration on Power System Voltage and Angle Stability”

The use of a  Finite Impulse Response (FIR) band pass filter as a possible solution to improve the frequency estimation of single- phase power system.

REU Poster     REU Presentation