A Data-Driven Approach to Interactive Visualization of Power Grids

Driven by emerging industry standards, electric utilities and grid coordination organizations are eager to seek advanced tools to assist grid operators to perform mission-critical tasks and enable them to make quick and accurate decisions. The emerging field of visual analytics holds tremendous promise for improving the business practices in todays electric power industry. However, the conducted investigation has revealed that the existing commercial power grid visualization tools restrict the visualization process to follow a limited number of pre-defined patterns created by human designers, thus hindering users ability to discover. In the final analysis of the 2003 East Coast Blackout, lack of situational awareness was cited as one of the contributing factors which aggravated the severity of the event. This project proposes a data-driven approach to interactive visualization of power grids. The proposed approach relies on developing powerful data manipulation algorithms to create visualizations based on the characteristics of empirically or mathematically derived data. The resulting visual presentation emphasizes what the data is rather than how the data should be presented in a pre-designed manner, thus fostering insight and enabling knowledge discovery. The proposed data-driven visualization approach, as demonstrated in the successfully completed Phase II project, has proven to be promising for building the next-generation power grid visualization tools. Application of this approach has resulted in a state-of-the-art model exploratory tool currently licensed by many utility organizations and vendors in the power industry. The supported use cases range from exploratory data analysis to visual data mining. Users have been proposing integrating the developed data-driven visualization solutions with their legacy utility information systems to provide added values. The proposed sequential Phase IIB of this project will focus on extending the pioneered data-driven visualization techniques to enhance situational awareness in a power grid control room. The goal is to assist system operators to perceptually monitor a large number of events and present the analytical information in a way that reduces cognitive demands on operators. The proposed research will be executed in collaboration with two industrial partners. The emphasis will be put on providing utilities with a rapid and cost-effective integration solution without requiring an overhaul of the existing utility information infrastructure. It is expected that the research to be performed will result in a commercialgrade real-time visualization tool that can be integrated with the existing utility information infrastructures, providing much-needed situational awareness and real-time decision support.