Read each of our poster sessions abstracts below. Interested in contacting the students / teams? No problem. That is available as well. For your convenience, the poster sessions are organized in the following groups:
Climate/environmental (climate action plan)
Distributed automation and distribution modeling
Markets (business side)
MicrogridsPower systems engineering
Renewables generation and integration
Smart cities (smart homes, electric vehicles, etc.)
Smart inverters and power electronics
Tatjana Dokic, Mladen Kezunovic - Texas A&M University, College Station
Abstract: The predictive method for distribution line vegetation management based on the risk framework is developed. The state of risk is calculated for each line section using variety of factors extracted from network parameters and historical outage data, historical weather and weather forecast, and variety of vegetation indices. The framework implements the spatiotemporal correlation of the collected data. The prediction model used in this study is Gaussian Conditional Random Field, which takes into account the spatial interdependencies between different sections. This enables better prediction accuracy, and a capability to deal with missing data. Based on the calculated risk the dynamic optimal tree trimming schedule is developed which minimizes the overall risk for the system under a given predetermined budget. The method was applied to the real distribution network and utility data. The testing confirmed that the outages occurred in the zones with risk predicted to be greater than 64%. The results obtained on the real utility network show that optimal tree trimming based on the developed risk framework for vegetation management could significantly decrease the overall risk of the system. It also demonstrates that the achieved reduction in risk has a potential of reducing the cost of reactive tree trimming.
Juan Ospina1, Alvi Newaz1, Nikhil Gupta1, M. Omar Faruque1, Rick Meeker2 - 1) FAMU-FSU College of Engineering, 2) Nhu, Energy, Inc.
Abstract: This research proposes an optimal control solution designed to manage the distribution grid-connected distributed energy resources (DERs) by developing a generalizable and scalable framework capable of controlling multiple DERs based on forecasted values and real-time energy prices. The control scheme is divided into two hierarchical level. The first level is in charge of optimizing the local energy resources and loads. This local optimization by the first level is accomplished by using an optimal real-time control approach based on a sampling-based model predictive control scheme designed to allocate the available energy resources at the local level of the system. The other level will control voltage to maintain stability and globally optimize the DERs with the help of the first level of control. The outcome of the research would be a controller capable of cost optimizing the loads and available energy resources in real-time without violating any of the system constraints.
Prateek Jain, Jacob Mueller, Jonathan Kimball - Missouri University Science & Technology
Abstract: Rising climatic concerns call for unconventional/renewable energy sources which reduce the carbon footprint. Microgrids integrating with a varied variety of renewable energies play a key role in utilizing these energies in a more efficient and environmentally friendly manner. Battery systems effectively help to utilize these energies more efficiently. This research presents a method based on Markov Decision Process (MDP) integrated with load and solar forecasting to derive an optimal charging/discharging action of Battery every 15 minutes throughout the day. This would not only reduce the monthly billing cost by reducing the peak load demand, but also maintain the state of charge of the battery. One month simulation results show that the microgrid with battery bank and MDP algorithm shaves the maximum load demand by 27.7% which leads to a cost saving of 9.94%.
Anne Clarke - Portland State University
Abstract: Today's electrical grid is presented with new challenges due to increasing penetration of residential solar and other distributed generation sources. In order to maintain balance and stability in the grid without building costly large-scale generation plants, utilities are turning to distributed energy resources for use in demand response. Demand response is a cost-efficient way to balance grid load/generation without the need for increase capital investment in traditional generation resources. With improved access and control, electric water heaters could very well become easily-exploitable distributed energy resources for utilities. In order to properly control and use a distributed energy resource, it is important to know how these resources operate and and their patterns of behavior in different environments. This paper presents an electric water heater model intended for analyzing the impact of water heaters as distributed energy resources for demand response.
Crystal Eppinger - Portland State University
Abstract: The Distributed Management System (DMS) is the second tier of a bi-level ad hoc grid management system, the Distributed Energy Resource Aggregation System (DERAS). DERAS utilizes an IoT framework designed to aggregate resources, optimize scheduling and provide utility ancillary services. The multi-agent platform provides the security of a distributed network and the coordination of a central control paradigm. The DMS facilitates peer-to-peer interaction between assets, forming a distributed network that can facilitate grid recovery and the formation of power hubs following a potential large-scale grid disturbance. It communicates with neighboring devices to schedule regional participation, effectively minimizing the data transfer between each tier. Optimization algorithms ensure an emphasis on grid-welfare over device autonomy. This will maintain a focus on grid stability and reliability, preventing non-optimal behavior, such as device fighting, a characteristic of autonomous systems. DERAS was built to demonstrate the value of coordinating and controlling DER. A clear understanding of networked DER reliability will affect future utility resource portfolio purchases. We hypothesize that large-scale DER aggregation, coordination and control can be achieved using a secure IoT platform. Such a system would provide ancillary services that contribute to a low-carbon, resilient and reliable grid with expanded opportunities for utility-customer interactions.
Tylor Slay - Portland State University
Abstract: With the advent of Internet of Things (IoT) platforms that can facilitate secure distributed coordination and control of large numbers of devices, DER asset aggregation may be designed to provide sophisticated utility services, such as frequency regulation, economic arbitrage and inc/dec ramp rate control. The potential for providing services such as these will facilitate large-scale adoption of renewable energy resources, whose non-dispatchable and stochastic characteristics limit penetration within the current technological con-text of the electric power system.
Neil Cammardella1, Joel Mathias1, Robert Moye1, Ana Bušić2, Sean Meyn1 - 1) University of Florida, 2) Inria/ENS
Abstract: Earlier this year, San Diego Gas & Electric installed a 30 MW, 120 MWh lithium-ion battery storage facility - the world's largest. But, recent work has shown that local control can enable flexible loads to provide equivalent battery-like services to the grid. Our analysis shows that it would have been much cheaper to acquire 30 MW, 120 MWh of energy storage by retro-fitting 120,000 water heaters with control hardware and telecommunications. Surprisingly, it would have also been cheaper to buy 120,000 brand new water heaters!
Krishna Sandeep Ayyagari, Nikolaos Gatsis, Ahmad F. Taha - University of Texas, San Antonio
Abstract: Distribution systems are undergoing a transformative change from passive circuits where power flows from upstream to downstream nodes to systems with distributed energy resources accommodating bidirectional power flows. Distributed energy resource typically include photovoltaic (PV) arrays and energy storage units. The time-varying renewable energy generation and user real and reactive power consumption are sources of uncertainty and can result in serious under- or over-voltage conditions. This work is concerned with optimally dispatching reactive power from inverters of PV units, which can be crucial for supporting safe nodal voltages. In this context, a stochastic optimal power flow problem is formulated, where the reactive power control actions are expressed as linear functions of the uncertain quantities. Voltage security specifications are included as probabilistic are included as probabilistic constraints. The objective is to minimize the thermal losses on the distribution lines by optimizing the linear policy that maps the uncertainty to control actions. The advantage is that when the uncertainty (that is, solar irradiance and power consumption) is observed, the reactive power dispatch can be easily computed. Numerical tests are conducted on an IEEE test distribution network to validate the satisfaction of the probabilistic voltage specifications.
Tumininu Lawanson, Roozbeh Karandeh, Dr. Valentina Cecchi, Dr. Zachary Wartell, Isaac Cho - University of North Carolina at Charlotte
Abstract: The deployment of smart meters and sensors in the electric power distribution system has led to a big data problem for electric utilities. A notable challenge for electric utilities is gleaning actionable intelligence from these accumulated data to improve operations at the distribution system level. Due to the volume, variance and velocity of distribution system data, advanced analytics is required to reveal correlations and patterns hidden in these vast data. In this work, an interactive visual analytics interface that enables distribution system operators make more informed decisions from distribution system data is presented. This interface utilizes datasets such as advanced metering infrastructure (AMI) and geographical information systems (GIS), and presents opportunities to improve the situational awareness of the distribution system in areas such as asset awareness, outage prediction, and grid optimization. Leveraging data from a real-world distribution network, this work presents the use of the visual analytics interface to identify inaccurate meter-to-service transformer associations, and to visualize voltage violations within the distribution network.
Reza Roofegari nejad, Amir Golshani, Dr. Wei Sun - University of Central Florida
Abstract: Power systems are constantly being affected by natural or man-made threats. The self-healing smart grid technologies enable system operators to quickly and automatically recover transmission and distribution systems from outages and blackouts. However, the current restoration studies solve the transmission and distribution system restoration problems separately without considering their impacts on each other, which may prolong the restoration process. On the other hand, with the emerging active distribution networks, distributed generators can actively take part in restoration procedure. Therefore, the interaction between the restoration of two interconnected systems should be taken into account. In this study, an integrated model of transmission and distribution systems restoration is proposed by exchanging data of power and voltage between two separate optimization problems. The effectiveness of the method is demonstrated by IEEE test networks. Simulation results demonstrate that the integrated restoration strategy with optimal scheduling of distributed generation resources can expedite the restoration process.
Ke Xu, Yuan Liao - University of Kentucky
Abstract: In power distribution systems, overcurrent relays (OC) are widely used for protection of the systems. The problem of overcurrent relay coordination in protection systems consists of selecting appropriate settings, satisfying the requirements of sensitivity, selectivity, reliability, and speed . Existing optimum coordination methods are mainly simulated under the conditions of fixed presence of DG and load in the distribution system, which is the static conditions for the distribution system. When there is a variation of load or distributed generations in the distribution system, the above optimum coordination methods are not good enough to find the most desirable coordination of overcurrent relays under prevailing operating conditions. This research proposes a method for online adaptive coordination of overcurrent relays considering significant load variation and renewable generation variation for systems with increasing integration of renewable energy sources.
Everett Hall, Thomas Ortmeyer - Clarkson University
Abstract: The project involves the creation of a model of the response of a Lithium-Ion (Li-Ion) battery to a solar photovoltaic (PV) supplied power system across a range of seasonal and weather variations. The model considers stand-alone operation of the array. The model was built in MATLAB Simulink and calculates the charge on the battery, cell voltage, and the cycle number of the battery from the instantaneous current, ambient temperature, and initial health of the battery cells. Results are presented that show the impacts of seasonal and weather variations on capability to serve the load and battery life. The developed model is especially important for stand-alone solar applications, as the batteries may be used in these system for many years.
Cesar Magana, Ryan Catalasan, Jane Lee, Andrew Phan, George Lachow - University of California, San Diego
Abstract: In an age of ever increasing amount of renewable energy sources, maintaining grid flexibility is essential for reliability of the electrical grid. The variability in renewables causes uncertainty in managing the demanded load. Energy storage systems can not only make up for the downsides of renewable energy sources, but also provide additional energy during peak hours while serving backups in the event of an emergency. In this paper, the current and forecasted issues associated with renewable energy sources will be presented, while providing a solution with electrical energy storage systems. Next, current energy storage solutions will be analyzed along with modern applications of electrical energy storage. Finally, we analyze various characteristics of the current energy storage solutions, we present use cases of storage solutions to maintain flexibility in the grid for increased penetrations of renewable resources.
Bohan Zhang - University of California, San Diego
Abstract: Brief introduction of microgrid(definition and motivation). Examples of some prominent energy storage technologies on the market suited for microgrids. Elaboration on the benifits of integrating energy storage in microgrids (improved availability, improved system stability, peak-shaving capability). Present challenges in Integrating energy storage(lack of appropriate control and managements). Conclude by summary and talk about future trends
Caroline Kamm - University of Central Florida
Abstract: As microgrid technologies emerge and become implemented in small communities, new market structures must accommodate the fluctuations in supply and demand in regards to both independent prosumers and traditional utilities. This research proposes a bid-shading auction strategy to satisfy demand while incentivizing participants to invest in renewable energy generation through a competitive market structure known as a Peer-to-Peer Energy Network. These results are compared to a utility-dominated dynamic pricing market by running a simulation involving five participating prosumers using weather and energy data from Orlando, Florida. The bid-shading practice was determined to be a viable market solution that kept prices stable within a given range and encouraged renewable energy growth despite theoretical utility price increases.
Gilles Mpembele - Missouri University of Science and Technology
Abstract: The analysis of microgrid stability requires a good modeling of the randomness in the disturbances affecting the system. For a microgrid, disturbances are generated when local power sources and loads are switched on and off. Because of the limited inertia of microgrids, these disturbances have a non-negligible effect on the system. Novel methods include the representation of power systems and microgrids as Stochastic Hybrid Systems. In this frame, continuous variations in the dynamic and algebraics states as well as discrete and randoms effects are conjointly modeled. The present study focuses on a particular type of SHSs called Markov Jump Linear Systems. The purpose is to present a detailed procedure to model a two-inverter based microgrid. A set of matrices are derived and then used to generate the conditional moments of the stochastic model. Methods to evaluate the system stability are also presented. This is done by analysing the system eigenvalues as well as by evaluating the bounds on the stochastic moments. Another aspect of this paper is to identify areas of the microgrid application of the MJLS theory that requires attention and future work.
Andrew Ludtka, Kyle Nadolinski, Kris Hammerick, William Hall, Lacy Taylor - SUNY Buffalo State College
Abstract: This project is the design and analysis of a functional microgrid to evaluate the effect of distributed generation (DG) and explore the concept of transactive energy (TE) using the Smart Grid Lab at SUNY Buffalo State. The testbed is used to examine the advanced techniques needed to manage supply and demand, and assess power quality when DG is introduced into the grid. Physical construction was completed with Lucas-Nuelle® power system laboratory equipment. DG sources consist of a double-fed induction generator emulating a wind turbine and a synchronous generator. An AC power source represents a utility, and supplies power to the microgrid through a model of an underground feeder. A double bus-bar system facilitates the connection of both the distributed assets and utility to a variable load. The system is monitored using an array of smart meters, which transmit data to a Siemens PLC-based SCADA system. Schweitzer Engineering Laboratory (SEL®) relays provide protection and reclosing operations for the feeder line, while a dynamic fault simulation device allows for the testing of relay functions and system stability. The microgrid serves as a means for testing theoretical concepts and demonstrating the operation of modern power systems.
Ali Jafrani, Eric Waters, Tyler Richard, Sajjad Fakour, Jongwon Jun - University of California, San Diego
Abstract: The fundamental differences between AC and DC distribution is one of the primary issues in power generation, storage, and control faced by utility companies as they work to reduce costs and increase efficiency. The development of new technologies, such as cheap solar PV arrays, efficient DC to DC converters, and an increasing desire for personal battery storage, makes DC distributed grids a viable option when compared to conventional AC distribution. The near future of power distribution could involve a combination of both AC and DC as new systems are designed around the increasing amount of DC power generation because it will decrease the steep inverter losses in the current systems
Hawazin Khattab, Aansh Malik, Gulmira Mustapaeva, Jason Liu, Bashar Oro - University of California, San Diego
Abstract: This research will encompass the feasibility of integrating DC into the grid by discussing two propositions: utilization of smart PID controllers and DC based homes and microgrids. The utilization of smart PID controllers would benefit the AC-DC integration by providing stable operation under various loads. Reliability of the DC signal will be discussed in the context of sensitive equipment. Moreover, devices nowadays have some sort of a converter either embedded in the system or in the form of a charger. Most of these converters convert the incoming AC power to DC. In the near future, most homes and the microgrid will use photovoltaic systems to generate DC power which also undergoes similar lossful conversion. DC microgrid is an ideal solution to improve the efficiency of the distribution grid and strengthening the existing AC distribution system.
Rasheed Abdulkade, Dr. Roy McCann - University of Arkansas
Abstract: The increasing use of distributed energy resources has motivated the development of microgrid structures to improve the reliability of electric power distribution systems. However, as microgrids become more complex, there is an increased occurrence for instabilities being induced through cascaded interconnections of multiple ac-ac converters. It has been found that back-back ac-ac converters have degraded stability margins when connected in a loop configuration compared to a radial configuration. This research investigates a variable structure control strategy; sliding mode control combined with an adaptive control methodology. Thus, the objective is to enhance regulation voltage levels throughout a loop configured microgrid system with respect to uncertainty bounds. A method for analyzing a complex microgrid structure with multiple interconnected converters is presented. The large-scale system is then linearized and a controller is designed using subspace methods. The controller design is then evaluated by application to a previously validated simulation model of a 6 MVA microgrid operated at the National Center for Reliable Electric Power Transmission at the University of Arkansas. Results confirm the robustness of the applied controller for improving the voltage regulation and disturbance rejection in a weak configured microgrid. x
Mehmet Akdogan, Mehdi Abolhassani - University of Houston
Abstract: Active, unbalanced, and harmonic power sharing inaccurately between distributed generators (DGs) and voltage quality problems are critical issues under unbalanced and nonlinear loads in the micro grids. Thus, an advanced power sharing scheme with unbalanced and harmonic loads in islanding micro grid is presented to eradicate power sharing problems and reduce unbalanced voltage and harmonics distortion. The hierarchical control structure comprises primary and secondary levels. For improving voltage quality, a secondary controller is designed to manage compensation of unbalanced and harmonic voltage at point common coupling (PCC). The Primary controller mainly includes power droop for sharing active and reactive power among DGs without communication, voltage and current controllers, and virtual impedance control loop. Virtual impedance loop for positive and negative sequences of fundamental and harmonic components is used to achieve better power sharing of reactive, unbalance and harmonic powers. The proposed approach is able to operate in islanded micro grids for voltage quality enhancement. Effectiveness of the hierarchical power sharing scheme is presented on MATLAB/Simulation and the simulation results of the proposed method are provided to demonstrate that PCC voltage distortion decreased from 9.47% to 4.62% while reactive, unbalanced, and harmonic powers is shared proportionally among DGs after compensation.
Max Liu1, Matthew Backes1, Alfonso Valdes2, Richard Macwan2, Paprapee Buason1 - 1) University of Illinois, Urbana-Champaign, 2) Information Trust Institute
Abstract: During recent years, microgrids are developed to increase resiliency of electric power systems. It is important to realize that; however, frequency of microgrids in an islanded mode is worrisome. As a result, a decentralized secondary frequency control is implemented to enable the frequency stability of distributed energy resources (DERs). We adopt the Alternating Direction Method of Multipliers (ADMM) algorithm, which uses measurements at various points in the systems, for solving the decentralized secondary frequency control updates. Furthermore, we use the Round-Robin technique in addition to an ADMM algorithm to detect malicious DERs. To increase the robustness of the systems, an agreement algorithm evaluates reported measurements with respect to KCL/KVL conditions is proposed for readily indicating the location of the erroneous measurement occurring from cyber attacks. A model simulated in MATLAB Simulink and Opal-RT provides effective results of a secondary frequency control and an agreement algorithm.
Nick Hawkins, Moath Alqatamin, Pablo Rivera, Joseph Latham, Michael McIntyre - University of Louisville
Abstract: The goal of this project is to provide students and faculty with a platform for testing micro-grid control schemes for power systems. The primary objective of this project is to design and construct a residential scale grid-like environment with a programmable load and multi-source capabilities. The ability to integrate renewable energy sources to the grid is challenging, and the proposed micro-grid test bed aims to provide solutions to these challenges. This project is designed to demonstrate how renewable sources can be used not only for supplemental generation, but also as a useful tool in improving common issues that are present in current AC grids. To this effect, various educational demonstrations will be prepared to illustrate the benefits associated with adding alternative sources, specifically solar energy, to the grid. The topics for demonstration include: grid synchronization, maximum power point tracking, power factor correction, unbalanced load compensation, and nonlinear load compensation. The project has four stages of implementation: construction of the test bed hardware, testing and implementing controls systems for the micro-grid, integration of the test bed into the existing distribution system for our DOE Solar Decathlon building, and developing university course materials and educational demonstrations for students at all academic phases.
Corey Weimann, Erick Bittenbender, Aryana Nakhai, Dr. Robert Kerestes - University of Pittsburgh
Abstract: As renewable energy technologies become cheaper and more accessible to commercial and residential sectors, consumers and utilities alike are seeking ways to further drive down cost and expand markets with these sustainable technologies. This paper seeks to address some of the major challenges facing widespread distributed energy resources (DERs) and where current research efforts are attempting to breakthrough on these challenges. Forecasting wind generation, HVDC transmission, energy storage technology, and microgrid control systems are all discussed with some solutions presented. While this paper does not propose novel solutions, it does survey hurdles facing these generation methods and proposes potential areas of interest for future research.
Lizon Maharjan - University of Texas, Dallas
Abstract: The recent increase in infiltration of distributed resources has challenged the traditional operation of power systems. Simultaneously, devastating effects of recent natural disasters have questioned the resilience of power infrastructure for an electricity dependent community. The presented study combines innovations in power electronics, controls, communication, and data science to provide a wholesome solution for improving grid reliability. Electronics: Multiport Power Electronics Interface (MPEI) facilitates seamless integration of DERs along with load categorization and DC power sharing for redundancy. Controls: Distributed Co-operative Control (DCC) obviates the fault prone star type communication through distributed controls. Communication: Recent developments in communication such as Internet of Things (IOT) devices allow remote placement of computing resources, hence providing significant computing capacity. Data Science: Use of Machine Learning (ML) algorithms enable integration of features from external sources and neighboring units to predict the grid faults more accurately, hence taking responsive actions to eliminate or reduce the effects of such faults. Furthermore, presented model encourages multifaceted design process which performs simultaneous analysis of all the involved systems rather than considering each system independently or considering one or more systems as additional features.
R. Scott Mongrain, Ziwei Yu, Raja Ayyanar – Arizona State University
Abstract: A software application is being developed to coordinate communication between real and simulated inverters. The application is being developed in Visual Basic, providing a GUI for ease of use. Communication is established via TCP/IP with an open-source implementation of the Modbus protocol. The intent is to allow researchers to control multiple inverters, both physical units capable of communicating over Modbus, and inverters simulated in an Opal RT system. Control objectives include step P-Q reference command following, energy storage monitoring and utilization, and some form of rudimentary economic dispatch as it relates to both grid-tied operation and islanded operation. Test systems are designed as part of the ASU LightWorks initiative, and include a physical microgrid setup featuring an inverter and a web-based monitoring and control system, and a simulated microgrid running in the Opal RT featuring configurable numbers of inverters and loads. The software acts as Modbus master for the simulated system and runs as a stand-alone application, but also features the ability to accept a slave role to a larger Modbus network, acting as an intermediary, tying two (or more) disparate systems together. Furthermore, this configuration allows for sandbox testing of simulations of physical devices prior to live control situations.
Olaolu Ajala, Alejandro Dominguez-Garcia, Pete Sauer, Daniel Loberzon – University of Illonis, Urbana-Champaign
Abstract: This work presents a library of two-dimensional state-space models that are significantly more accurate than the so-called classical model of a synchronous machine, and are therefore useful for a broader range of applications. The dynamic states of the models are the power angle, and the angular speed, and the fidelity of the models is verified through a simulation-based comparison of the response of the proposed models with that of a high-order model, and that of the classical model.
Elizabeth Anne Devore, Dr. S. Mark Halpin - Auburn University
Abstract: A reputable tool for measuring and calculating the source of harmonic impedance in power systems has not yet been developed. In order to develop such a system will require, preferably, a non-invasive method to measure and calculate harmonic impedances. Achieving such a system will require reliable computation of the harmonic impedance based on measured phase voltages and currents. In this work, voltage and current measurements are used to calculate harmonic impedance using symmetrical components. These results show that symmetrical components are not a reliable form for calculating harmonic impedance. Further consideration is given to calculations using alpha, beta, and zero components.
Behdad Vatani, Dr. Badrul Chowdhury, Jeremy Lin - University of North Carolina, Charlotte
Abstract: Demand response can not only be used as a power supply resource to produce a negawatt in high power prices or when the reliability of the grid is threatened, but also can be applied to resolve a transmission expansion planning problem instead of implementing a costly and complex transmission upgrade solution. Essentially, the demand response resources (DRRs) can relieve the capacity requirement for a load area, and thereby relieve the import transmission requirement. This work explores the role of DRRs as an alternative solution to the required transmission upgrades in the context of a proposed capacity market modeled based on PJM’s capacity market model. Numerical results show that the DRRs can indeed replace the needed transmission upgrades.
Mustafa Lahmar, Yuan Liao - University of Kentucky
Abstract: Modeling of power systems plays a key role in power system analysis and designing protection and control algorithms and diverse smart grid applications. Among various power system model parameters, line parameters are key parameters, and accurate value of line parameters will ensure the reliability and accuracy of the algorithms and applications that rely on line parameters. This paper is devoted to estimating transmission line positive-sequence parameters from unsynchronized measurements of voltage and current phasors that are obtained at both terminals of the line. The series impedance and shunt admittance line parameters can be linearly estimated using the algorithm developed in this paper. The algorithm is developed based on the distributed parameter line model, and the line is assumed to be transposed. The approach developed in this paper is considered as a linear method, and the linear least square method is used to estimate the positive-sequence line parameters. The method is also applicable to synchronized measurements obtained by Phasor Measurement Unit and is able to deal with potential synchronization errors by explicitly modeling the synchronization angle. Sample results are reported to demonstrate the effectiveness of the proposed method.
Ian Dudley, Brian Eggers, Derek Clark, Tim Eckhart, Dan Richer - Washington State University
Abstract: This project aims to research and develop an image acquisition system with the intent to be mounted on the rotor of an operational hydroelectric generator. With such a system, diagnostic analysis of the stator end windings can be achieved without the need to shut down the system. This project is a proof of concept design that is intended to capture clear images at a known point within the generator and then store that image to an on-board memory location for later retrieval. The camera, power supply, and data acquisition unit must be shown capable of surviving the high temperature, and alternating magnetic field ratings. Also, the enclosure must be shown capable of withstanding the high G-forces associated with the generator. This environment was simulated by constructing a magnetic coil able to produce rated field strengths, and preforming FEA analysis of the system.
Muhammad Usama Usman, Md. Omar Faruque - Florida State University
Abstract: Due to the penetration of distributed generation (DG) in the existing grid system, the power flow is no longer unidirectional. High penetrations of PV with highly variable output can pose design and operation challenges for distribution engineers. One of the challenges is the varying fault current levels. In this work, a case study is conducted to analyze the fault current contribution from PV in a distribution system. IEEE-37 bus benchmark test system used as a test case distribution feeder where four PV plants are added. Different PV penetration levels and locations are considered to study the fault current contribution from the PV plants. It is observed that during fault conditions, PV increases the fault current levels at all points on the system. It is also observed that fault current increases with the increase in PV penetration. This case study can be useful for utilities and distribution protection engineers to set up protective devices according to the fault current contribution from PV.
Martin Magno, Hamna Khan, Nicholas Abi-Samra - University of California San Diego
Abstract: Widespread integration of photovoltaic (PV) generation is experiencing growth rate due to a combination of government incentives, technology improvements, and reduction in costs. The objective of this research is to investigate the effects of PV load generation at the local distribution circuit and evaluate typical distribution circuit parameters that can potentially affect the voltage quality being delivered. Of focus will be the steady-state voltage regulation and voltage imbalance. Certain characteristics unique to this resource must be considered in the system impact and planning process, namely the level of PV power generation as well as the centralized versus distributed placement of PVs. Simulations conducted on the ETAP software will be used to identify the conditions that lead to poor voltage quality, and to recommend steps towards mitigating the impacts of PV generation for the most susceptible customers.
Oluwaseun Akeyo, Vandana Rallabandi, Dan M. Ionel - University of Kentucky
Abstract: This poster analyzes a method of expanding the capacity of an existing irrigation farm with additional pumps powered by solar PV. The system includes PV arrays and battery energy storage connected to a common dc bus, which energizes an array of variable speed inverter driven pumps. Capacity modulation is achieved by energizing an optimal number of pumps required in order to meet a particular load demand with minimum supply energy. A grid connection to the dc bus of the power electronic system is established via a bidirectional converter, such that active and reactive power demands can be both serviced. The controls and the steady-state and transient performance of the system are implemented and simulated with the PSCAD/EMTDC software.
Hayder Abbood, Andrea Benigni - University of South Carolina
Abstract: A Data-Driven Modeling(DDM) approach is presented for static modeling of commercial Photovoltaic (PV) micro-inverters. No prior knowledge of internal components, structure, and control algorithm is assumed in developing the model. The approach is based on Artificial Neural Network (ANN) and Fast Fourier Transform (FFT). A Power Hardware in the Loop (PHIL) approach is used to generate the data. The obtained results demonstrate the good accuracy of the developed model.
Qin Yan, Bei Zhang, Mladen Kezunovic - Texas A & M University
Abstract: A four-stage intelligent optimization and control algorithm for an EV bidirectional charging station equipped with photovoltaic (PV) generation and fixed battery energy storage, and integrated with a commercial building is proposed. It has been stated in stages: a) Stage I, optimization of day-ahead energy management schedules, b) Stage II, multi-tiered EV charging price update and optimization of discharging participation bonus, c) Stage III, optimization of hour-ahead energy management schedules, and d) Stage IV, real-time control. This optimization model aims at minimizing the overall operational cost for each element in the integrated system while taking customers’ satisfaction (EV owners and building customers) into consideration. Such algorithm provides more resilience for unpredictable conditions, provides more incentives for EV users to participate, and better coordinates the integrated system including the building load to reliably serve the customers while lessening cost. Case studies are implemented and the comparison analysis is performed in terms of the use and benefit of each design feature of the algorithm. The results indicate that the proposed algorithm can reduce the operational cost and at the same time provide higher tolerability towards uncertainties.
Ali Zaighum Jafrani, Michael Chen - University of California, San Diego
Abstract: In this paper, the effect of increased penetration of Electric vehicles on Distribution transformers is discussed and quantitatively evaluated. These effects are evaluated by using a modified tool created by Dr. Nicholas Abi Samra which is based on IEEE C57-91-2011 G Annex. This tool uses VBA Logic to superimpose three different vehicle loading profiles on an existing load line and then calculates iteratively the final temperatures within the transformer and calculates the cumulative aging. In this paper, we make use of three different charging time zones and deduce the impact of each time zone on the aging. The time zone which affects the aging most dramatically is Time Zone 2 which contains of the smallest number of hours and thus leads to superimposition of the Electric Vehicles on top of one another. Furthermore, in the paper we try to work with LA head storm data to figure out the adverse effects of having additional EV penetration on the transformers.
Huangjie Gong, Dan M. Ionel - University of Kentucky
Abstract: Net zero energy (NZE) building contribute less overall greenhouse gas to the atmosphere than similar non-NZE buildings. Small residences with solar photovoltaic (PV) generators may be able to realize net zero energy consumption. However, as the electricity usage differs from building to building and depends on the time of the day, it is essential to inspect the real-time energy consumption of each building to achieve internal balance. The physical structure of a building can be modeled in SketchUp, a google 3D modeling software tool, with the help of OpenStudio, which is a collection of software tools used to support whole building energy modeling. The model representing a single building is used in EnergyPlus, which is a whole building energy simulation program to estimate energy consumption. Several buildings are simulated in parallel, and the data is transitioned to MATLAB using an internal toolbox called MLE+ for co-simulation with EnergyPlus. MATLAB works as the information center to collect data from each building. With the information from this data, control techniques to ensure that surplus PV generation supports high demand load can be developed, serving to reduce power demand.
Namwon Kim, Babak Parkhideh - University of North Carolina, Charlotte
Abstract: This paper presents a novel PV-battery power electronics topology with a partially-rated universal optimizer enabling a battery integration into the commercial rooftop PV inverter system. The proposed universal optimizer is a dc-dc converter providing wide PV input voltage range and high system efficiency with a seamless battery integration. The battey unit is used as not only a power source but also a PV voltage supporter. In the proposed topology, a dual active bridge dc-dc converter is applied to configure partially-rated power electronics system with bidirectional power flow, galvanic isolation, a high voltage boosting gain, and high conversion efficiency. The proposed topology is analyzed in different operational modes. The PV generation and the battery charging and discharging operations are verified through the controller hardware-in-the-loop platform.
Pankaj Kumar Bhowmik, Dr. Madhav Manjrekar - University of North Carolina, Charlotte
Abstract: In the present decade, utilities are facing enormous challenges to maintain customer power quality at desired level. The major reasons behind this scenario may be attributed to unprecedented increase in several sectors such as renewable energy resources, sensitive electronic loads, distributed energy resources and electric vehicles. Recent literature survey indicates a lot of good research has been reported on power electronic based power conditioning solutions to mitigate these power quality issues in the utility distribution grid. But the practical field implementation of these solutions still remains challenging since there is lack of commercial availability of power semiconductor devices rated at utility scale voltages. The Center-Point-Clamped topology allows the use of power semiconductor devices rated at half the utility voltages, thereby making the commercial implementation more viable. Therefore, Center-Point-Clamped Power Electronic Transformer is a power conditioning solution which can eliminate grid voltage sag, swell, and improve overall THD of the distribution grid, thereby maintaining the grid power quality within permissible limits. The proposed solution also has certain inherent advantages like simpler and fast control, provide power line protection, and increased power density with reduced power density as compared to traditional power conditioning devices like tap changing transformers.
Seungyong Lee, Roy A. McCann - University of Arkansas
Abstract: In this research, a model predictive current control of a three-phase voltage source inverter is developed with an extended state observer (ESO) for grid integration of renewable energy systems. This overcomes many of the challenges for interfacing variable energy resources into electric power systems. Model predictive control (MPC) exhibits excellent dynamic performance by predicting future behavior with minimization of a cost function. Modeling errors and parameter variations which MPC-based control techniques are unable to handle effectively can be attenuated by a constructed ESO. Since an ESO is not dependent of the mathematical model of a system, it can play a key role to suppress model uncertainties and disturbances. The proposed predictive current controller ensures rapid dynamic response of the inverter. Furthermore, incorporation of the ESO enables robust control performance in terms of uncertainties and disturbances. Simulation results confirm the effectiveness of the developed ESO with the predictive current control and improved dynamic performance for a grid-connected voltage-source inverter.
Yibin Zhang, Vandana Rallabandi, Dan M. Ionel - University of Kentucky
Abstract: Multi-level inverters (MLI) overcomes drawbacks in comventional two-level converter, including high THD, high voltage across power switch, high dv/dt of output voltage and electromagnetic interferences, for medium-voltage PV systems. A neutral point clamped multi-level inverter and a cascaded H-bridge multi-level inverter are modeled and studied in Simplorer.
Nicholas Hawkins, Michael McIntyre, Joseph Latham - University of Louisville
Abstract: The methodology of designing and testing a five-level diode clamped multi-level inverter (DCMLI) for high fidelity voltage output is presented. The proposed system uses a diode clamped circuit topology for the creation of more voltage levels than a typical inverter, and utilizes FPGA hardware to interface with LabView to generate PWM switching. This is applied to an LC-filtered RL load both in open and closed-loop form, using a controller that requires a single voltage measurement. Simulation and experimentation results are given to validate the performance of this system.
Thomas V. Cook, Justin Hawks, Christopher Greathouse, Alexander Fabian, Dr. Robert Kerestes - University of Pittsburgh
Abstract: Eaton desired a Variable Three Phase Source for demonstrating their circuit breakers and other metering products at trade shows. The device had to run off of a standard wall outlet and provide safe low-voltage and low-current three phase power. Eaton required the device to be self-contained and take regular single phase 120V AC and produce a 3-phase output with a voltage range of 0-5V and a current range of 0-250mA. The frequency had to be able to adjust between 50Hz and 400Hz and the phase must be entirely adjustable.
Santino Graziani - University of Pittsburgh
Abstract: This research proposes a new power electronic converter topology for pulsed power systems. A flying capacitor modification to the standard flyback converter allows for N number of capacitors on the output stage. In previous topologies this method reduced the stress on the capacitor and semiconductor components by spreading the voltage gain requirement across multiple stages. Reducing the component stress allows for smaller and more efficient devices which leads to increased efficiencies. This effect also reduced the required magnetic component size increasing the realizable power density. Taking the modified flyback converter and stacking it at the output in series leads to higher pulsed power peaks and also faster rise times (dv/dt). The goal is to realize a new pulsed power converter that can achieve faster and higher power pulses without sacrificing efficiency or power density. Initial analysis and device simulations for the modified flyback show a promising conversion ratio with added benefit from the flying capacitors. Moving forward the plan is to benchmark existing pulsed power topologies and also steady state DC-DC step-up topologies and optimize a flying capacitor flyback converter for both cases. These designs will, initially, be realized with detailed simulation models.
Susie Wooyoung Choi, Matt Woongkul Lee, Pablo Castro Palavicino, Bulent Sarlioglu - University of Wisconsin-Madison
Abstract: As dynamic loads and different kinds of distributed energy resources are penetrating more in distributed generation system and microgrid application, power quality issue becomes significant to supply stable and clean power. Multi-functional grid-connected inverters (MFGCIs) are getting attending as power electronics interface between the grid and distributed energy resources not only to convert DC to AC power, but also to improve power quality of the utility grid. However, conventional shunt-configured MFGCIs have limitations on capability of mitigating voltage-based power quality problems, the capacity of inverter, and effect of grid condition. This research proposes new configuration of MFGCIs, named as shunt-series-switched MFGCI (SSS-MFGCI) to solve the challenges that conventional MFGCIs have. The proposed SSS-MFGCI can be connected to the grid in shunt or series according different grid disturbances in different operating modes. Optimal and effective use of MFGCI for power quality improvement is achievable without over-design the inverter. New topology, modeling, analysis, design, control, performance evaluation, and simulated and experimental results of proposed MFGCI are presented.
Carlos Caicedo-Narvaez, Lizon Maharjan - University of Texas, Dallas
Abstract: Exponential growth of data collection technologies as well as cloud computing and analytics, also known as Internet of Things, has enabled industry to pursue real-time condition monitoring of electric machines. Using sensor clusters as aggregators, vital electric machine information such as, frame temperature and vibration is sent to the cloud for processing and storage. Such information can be fused with weather forecasts, grid statistics, and other external data to enhance the quality of information used for condition monitoring. Using this fused data, a platform as a service (PaaS) is employed to perform machine learning algorithms such as multilayer perceptron (MLP) and support vector machine (SVM) in order to extend the life of the electric machine, perform just-in-time maintenance, and increase overall efficiency. Applications of such system can be visualized from small 2kW HVAC motors to Megawatt scale generator units.
Turgay Duman, Hariharan Krishnaswami - University of Texas, San Antonio
Abstract: Modular Multilevel converters (MMC) have many advantages over conventional voltage source converters. Some of those advantages are higher voltage levels, low THD, modularity and lower dv/dt per switch. However, they usually contain more number of switches compared to conventional converters. Therefore offline simulation of MMC with higher number of switches can be time-consuming and can present challenges in simulating even few cycles. Sim power systems (SPS) is the main blockset of Simulink to simulate electric circuits. It is based on state-space equations of power systems. It uses pre-calculation of all possible state-space matrices for all switch possibilities. Therefore the simulation of circuits with significantly higher number of switches will not be easy and fast to simulate. One of the real-time simulators developed by OPAL-RT has introduced ARTEMIS solver, which enables to use State Space Nodal (SSN) solver. SSN of ARTEMIS separates the electric circuit into small groups and finds the associated equations for each one. While Sim Power System is trying to solve the entire inverter circuit, SSN divides this complicated network into small groups and solves for each using SSN. This project demonstrates an application of the SSN solver for a cascaded MMC photovoltaic inverter. The inverter consists of several ports per phase, with each port interfaced with solar PV arrays. Each port level inverter has a multilevel Dual Active Bridge (DAB) converter followed by a Neutral Point Clamped (NPC) inverter. The output from each port is cascaded to produce single-phase output, which are then combined to produce a high power three-phase cascaded inverter. Real-time simulation and modeling of three-port cascaded Multilevel DAB and NPC inverter using ARTEMIS solver is presented and results are shown in this poster.