August 26, 2016
April 4, 2016
Rachelle Hollander, Adjo Amekudzi-Kennedy, Sarah Bell, Frazier Benya, Cliff Davidson, Craig Farkos, David Fasenfast, Regina Guyer, Angelique Hjarding, Michael Lizotte, Dianne Quigley, Diana Watts, & Kate Whitefoot
The Integrated Network for Social Sustainability (INSS) is a research-coordination network supported by the National Science Foundation that is currently in its third year of activities. Individual and institutional members, representing a wide range of fields and interests, are devoted to addressing social sustainability as an important, understudied issue under the broader rubric of sustainability and sustainable development. The INSS has developed a number of affinity groups and a set of activities to facilitate its development. An annual conference draws members together to review and report on their efforts. At the first conference, a group interested in developing a research agenda formed. This Community Essay shares its members’ perspectives about priorities for future research and education on social sustainability, highlighting efforts for greater inclusion of marginalized populations in research.
January 28, 2016
Raghuveer Gouribhatla, M.S.
(MS Thesis Adviser: Dr. Srinivas S. Pulugurtha)
This research is aimed at estimating and predicting energy loss due to traffic congestion at link-level in the transportation network. INRIX data, for the year of 2012, for the city of Charlotte, North Carolina was considered for research and analysis. The raw data contains travel time information for every minute of the day during the year. Three different times of the day were considered for research and analysis. They are: morning peak hour (8 am - 9 am), off-peak hour (12 pm - 1 pm) and evening peak hour (5 pm - 6 pm). The variations in energy loss patterns on a weekday and a weekend day during the selected times of the day were examined. Mathematical equations were used to compute delay due to congestion and the energy loss.
Energy loss maps at link-level and Voronoi maps were generated in Geographic Information Systems (GIS) environment to examine spatial variations in energy loss. The Voronoi maps have the potential to predict energy loss for links with missing data. As an example, figures below depict geospatial variations in energy loss due to recurring congestion (effect of traffic) on a weekday (left) and weekend (right) from Corridors such as Brookshire freeway and I-485 in the city of Charlotte were observed to experience the highest energy loss during most of the times considered in this research. The energy losses were higher towards the downtown/uptown area and decreased as the distance from the downtown/uptown area increased. Further, the energy loss due to recurring and non-recurring congestion was observed to be the highest during evening peak hour on a weekday and off-peak hour on a weekend, respectively.
June 16, 2015
Optimal Battery Energy Storage Smart Management System for Multiple Value Streams Using Weather Patterns
Dr. Sukumar Kamalasadan and Dr. Tao Hong
In this project, a fully integrated optimal Battery Energy Storage System (BESS) Controller for multiple utility value streams using weather patterns is developed. Three main storage applications were designed and implemented in a cooperation between the University of North Carolina at Charlotte and Duke Energy. Namely, PV Station Capacity Firming (PVCF), Energy Time Shift Arbitrage (ETS) and Volt/VAR Regulation. Other standard auxiliary services like Frequency Regulation and Spinning Reserves are added in the optimization process.
The proposed PVCF algorithm incorporates a low-cost cloud state prediction scheme for system sate optimization. The process involves two main stages. The first stage involves the analysis of historical PV station output and logged cloud state data for the purpose of identifying optimal algorithm parameters for each predetermined day type which is based on cloud state patterns. An offline dynamic programming optimization technique is applied. The second stage involves the utilization of web-based cloud state predictions to recognize day ahead weather conditions and identify cloud state pattern. Pattern pre-calculated algorithm values (dynamic programming results), including starting State of Charge (SoC), are applied for firming maximization. The implemented ETS algorithm utilizes historical feeder data and ambient temperature correlations to predict peak feeder load magnitude and time. Optimization of BESS SoC is performed throughout the firming period based on predicted time of feeder peak load and forecasted cloud sates. The volt/VAR regulation algorithm utilizes the Storage Management System (SMS) reactive power capabilities to optimize voltage at the point of common coupling based on feeder voltage regulators’ tap change operation minimization. Other standard algorithms like frequency regulation, spinning reserves are added to provide additional value. These optimization algorithms are implemented and evaluated at a Duke Energy substation with 1 MW of PV array, 1.25 MVA energy storage converter with 250 kW / 4-hour battery system.
Quantification of individual as well as integrated value of the implemented algorithms is sought , focusing on the revenue generation, ROI, and energy storage life-time optimization results.