I-Corps: Artificial intelligence platform to provide customized power management solutions

California State L A University Auxiliary Services Inc., Los Angeles CA

Investigators

Abstract

The broader impact/commercial potential of this I-Corps project is the development of an Artificial Intelligence (AI) power management system that helps businesses and residential communities reduce their electricity bills. In addition, the proposed technology may reduce the fossil fuel usage required for producing electricity and, ultimately, reduce global warming. The proposed technology may encourage more electric vehicle chargers and renewable energy generators in the customers' internal power network due to more economical operations. In addition, the technology also may alleviate the peak power burden on the utility and generators, which improves the reserve of electrical energy during the hours of a day that it is needed most. The current focus is on large companies, however, the proposed technology may be applied directly to medium and small size businesses and residential buildings as electric vehicle chargers and renewable generators are expected to be widely used. This I-Corps project is based on the development of a software package and real-time Artificial Intelligence (AI) power management system that reduces electricity usage. The proposed software has features such as comprehensive objective functions, including complex Time-Of-Use rate plans and degradation of components, and is intertwined through an iterative process with the proposed AI power management algorithm. Data is collected from the client and fed into the sizing software to determine renewable energy sources (e.g., photovoltaic panel) and battery sizes. After sizing is completed, the AI management system controls the battery's inverter in real-time. The proposed technology finds the most cost-effective strategy using artificial intelligence and optimization techniques, and the real-time decision with the savings and best resiliency outcome is released to the battery's inverter. A comprehensive sensitivity analysis to optimize the parameters of the AI management system has been completed. Preliminary results show that the proposed technology may reduce electricity bills for large-sized businesses up to 15% more than standard peak-shaving energy management methods while also increasing "resiliency under outage.” 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.