Search

We present k-ShapeStream, a clustering method for streaming time-series data. In addition to the algorithmic novelty, the method represents a highly practical approach for electric grid data analytics, requiring no model assumptions or ground truth information, running sustainably on ever growing datasets, and providing intuitive and insightful results to grid operators. We demonstrate the effectiveness of k-ShapeStream using several months of real synchrophasor data from an operational distribution network in California. Through two case studies on (i) transformer tap changes; and (ii) voltage sags, we illustrate how k-ShapeStream assists in identifying and analyzing recurring grid events, a critical task for decision making in electric grids.

In this work, we present a Real-Time, Multi-layer cybEr-power TestbEd for the Resiliency analysis (RT-METER) to support power grid operation and planning. Developed cyber-power testbed provides a mechanism for end-to-end validation of advanced tools for cyber-power grid monitoring, control, and planning. By integrating a host of features across three core layers-physical power system, communication network, and monitoring/control center with advanced tools,---the testbed provides a platform to model and simulate rich and diverse cyber-power grid scenarios and generating realistic sensors, system, and network data. These data are subsequently used in validating advanced tools to assist operators during complex and challenging scenarios, which is essential for the successful operation of the future grid. We detail a suite of algorithmic tools validated using the developed testbed and the generated realistic grid data.

In this work, we present an ongoing, three year project funded by the Advanced Research Project Agency (ARPA-E) to develop an infrastructure to accelerate the development and deployment of artificial intelligence solutions for the electric grid. The project addresses the critical issues that we have identified as hampering the deployment of AI on electric grid measurements. These issues have stymied the potential for insight from high resolution grid measurements and generally hindered artificial intelligence innovation in the utility industry. The project consists of three main components. The first is the deployment of a diverse set of grid sensors to capture a variety of grid behaviour, both from the field and in simulation. The second is the deployment of a highly performant, scalable, cloud-based data management and AI platform designed for time series data to enable the easy storage, processing and analysis of grid sensor data. The third is the cultivation of an open research community of experts around the platform and data through useful educational material, code and data sharing, and data science competitions. Overall, the project will accelerate the development of analytics, machine learning, and AI by addressing existing gaps in data, tools, and people, with the aim of improving the electric grid.

Time synchronized measurements of voltage magnitudes or phasors are increasingly common in electrical networks. Voltage measurement statistics are informative of the underlying network structure or topology making them useful for grid monitoring. However, this connection is poorly understood and many proposed voltage analytics are purely heuristic. We use graph theory to establish sound theoretical connections between voltage measurements and the structure of the underlying network. Our results are important for many applications, from topology estimation to missing data recovery. Based on this new theory, we discuss existing analytics, transforming them from heuristic to theoretically justified approaches, and introduce new analytics. We clarify all assumptions made, to indicate when analytics may fail or perform poorly. Our work enables voltage measurement streams to be transformed into physically meaningful, intuitive, visualizable, actionable information through simple algorithms.