PROJECT PROFILE: University of Southern California (ENERGISE)

You are here

Title: DEEP SOLAR: Data Driven Modeling and Analytics for Enhanced System Layer Implementation
Funding Opportunity: ENERGISE
SunShot Subprogram: Systems Integration
Location: Los Angeles, CA
Amount Awarded: $1,886,975
Awardee Cost Share: $537,027

This project uses data to develop novel representations of distributed energy resource owners’ interactions via data-driven models along with stochastic reserve optimizations that enable net-load balancing in near real-time. The project develops a transformational distributed grid control architecture as part of an enhanced system layer at the distribution network level. It optimizes the coordinated usage of a large number of variable and distributed resources, decentralized energy storage, and load to ensure real-time, system-wide, net-load management and automated adaptation to real-time variability in a cost-effective, secure, and reliable manner.


The research team will improve predictive analytics through real time visualization, accurate modeling and prediction of individual load curves and generation patterns for day-ahead, hour-ahead, and 5-minute forecast horizons. They are also targeting stochastic analytics by increasing grid edge functionality through the optimization of operational and emerging distribution markets with greater than 99% observability and a solution time of less than 60 seconds. Finally, they will improve dynamic scenario analysis using a cloud-enabled software platform, which will be scalable to 1 million active nodes.


The enhanced system layer under development consists of a software toolkit for real-time net-load management, distributed, and adaptive control of distributed energy resources that is open-source and portable across utilities with interoperable middleware. This solution will, for the first time, show that a distribution network can achieve a 100% or greater level of integration of renewable energy sources onto the grid.