Sensor Enabled Modeling of Future Distribution Systems with Distributed Energy Resources
Although the electrical distribution system has existed for more than a century, its real-time operation is not significantly automated due to lack of need or means for extensive control. Today’s increasing number of distributed energy resources (DERs) and changing customer patterns challenge this operational paradigm. To maintain, and hopefully improve, the distribution systems’ reliability and efficiency while accommodating a high level of DERs, critical distribution system planning, operation, and control tools must be developed.
Project Innovation + Advantages:
Arizona State University will develop learning-ready models and control tools to maintain sensor-rich distribution systems in the presence of high levels of DER and storage. This approach will include topology processing algorithms, load and DER models for system planning and operation, distribution system state estimation, optimal DER operational scheduling algorithms, and system-level DER control strategies that leverage inverter controls’ flexibility. The project will alter distribution system operation from today’s reactive, load-serving, and outage mitigation-focused approach to an active DER, load, and outage-managed, market-ready approach.
The project will transform how America’s power distribution system operates. It will eliminate today’s net demand/supply paradigm in favor of an active, market‑ready, real-time load management system built on distribution topology and distributed sensing and measurement data. This more agile, faster-reacting system will enable greater adoption of distributed energy resources (DER), which in turn enhances system reliability.
Integrating more DERs in a distribution grid, as opposed to centralized generation, will make the local grid more sustainable and resilient. This will increase the nation’s energy independence and security.
In addition to enabling more renewable DERs, conducting joint data-driven inference on load modeling, voltage monitoring, and control improves energy efficiency. Both of these attributes can help reduce emissions.
The proposed interpretable and adaptive deep learning tools will ensure that the U.S. maintains a technological lead in developing and deploying advanced energy technologies.