The ability of active consumers and end devices to buy/sell energy and related services in a dynamic and interactive manner is expected to create a transactive energy (TE) market presenting new opportunities and challenges for distribution system operations. The implications of a market-based approach to operational management need to be better understood at both the consumer and utility levels. Researchers have typically attempted to address important and open operational questions using a simulation-based scenario analysis approach. The spatial and temporal randomness associated with energy generation/consumption as well as the stochastic behavior of consumers in response to pricing, significantly complicates the scale and number of Monte-Carlo simulations required to make meaningful and statistically valid inferences. Any changes to the network can force a re-computation that is cumbersome. While simulations provide useful insights on system operations for a specific set up, the dynamic nature of the TE market with multiple sources of uncertainty begs for the design of new planning tools. In this project, a suite of novel stochastic geometry based analytical approximations and bounds are derived in order to develop a unique operational planning tool that can be used in a sustaining mode (i.e., in conjunction with simulation tools) or as a disruptive standalone alternative depending on the needs of distribution system operator (DSO).
This proposal introduces a new paradigm for operational planning incorporating analytical approximations and stochastic bounds in lieu of cumbersome simulations. By bringing together techniques from stochastic geometry, graph theory, information theory and power systems analysis, the PIs will derive new probabilistic voltage and loss sensitivity metrics that can be computed quickly and scale well with the size of the network. The probabilistic sensitivity analysis relies on both consumer level and network level stochastic abstractions that help quantify the average impact of varying penetration levels of active consumers on grid performance. The results from the probabilistic sensitivity analysis are used to develop the unique concept of a dominant influencer. Using the dominant influencer models, a network vulnerability rank can be determined and used to (1) quantify the susceptibility of the network to various points of cyber/physical attack, and (2) develop a hybrid control framework for DSOs to deploy and manage control assets within a TE system. The stochastic geometry framework also enables fast identification of inactive voltage constraints that can expedite power flow computations
In summary, this project provides tools that will enable distribution systems operators to efficiently plan and operate distribution systems with large penetration of distributed renewable energy resources, electric vehicles, and engaged consumers. These tools will allow true techno-economic analysis of the system while incorporating consumer preferences. Results from this research will be: (1) integrated into several undergraduate and graduate courses, and (2) disseminated via publications in academic conferences and journals. The PIs will educate the public through Open House displays and field trips for K-12 students to encourage students from under-represented groups to pursue STEM careers.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
