Department of Defense (DoD) missions rely on the uninterrupted operation of installation critical infrastructure such as heating and cooling, water and wastewater, electrical power, and alarm systems. Failure of this infrastructure may lead to mission failure. Most DoD critical infrastructure is monitored and controlled though a computer system known sometimes as Supervisory Control and Data Acquisition (SCADA), but this data is often stove piped and used only to drive a user interface with alarms based on simple heuristics. The wealth of information generated by SCADA systems aren’t being fully utilized for helpful services such as anomaly detection and preventive maintenance purposes. This is a missed opportunity for the DoD given the recent progress in data science, including artificial intelligence and machine learning (AI/ML). 

The objective of this project is to increase the resilience of DoD energy infrastructure through the advancement of real-time data analytics. This will be achieved by generating data sets, test harnesses, and baseline algorithms to encourage industry, academia, and government to develop next generation data analytics processes.

There are three primary capabilities that will be developed through this program. The first will be canonical data set collected from real-world SCADA systems that will enable the development of infrastructure data analytics by anyone in the community (academia, industry, or government). No such data set of DoD infrastructure currently exists due to collection challenges, which in turn limits the development of advanced algorithms that typically require large quantities for training and verification. The second capability will be a software test harness allowing data analytics solutions to be tested in a methodical, objective, and repeatable fashion. This type of test and evaluation process and associated test harness for infrastructure data analytics does not currently exist within the DoD. Finally, this project will develop baseline data analytics algorithms leveraging modern data science techniques such as AI/ML to push beyond the current heuristics-based approaches used for monitoring infrastructure.

The data analytics solutions that result from the Platform for Energy Resilience Data Analytics framework will have broad applicability to nearly all DoD installations as solutions will be sought that are generalizable to many different vendors and systems. Findings will allow DoD to objectively compare and evaluate different approaches which in turn can inform acquisition efforts. Finally, this project can help alleviate existing workforce challenges within the DoD as novel AI/ML algorithms can assist new personnel as they develop their system level understanding of the infrastructure they are tasked to maintain. (Anticipated Project Completion - 2028)

DISTRIBUTION STATEMENT A. Approved for public release. Distribution is unlimited.

This material is based upon work supported by the Under Secretary of War for Research and Engineering under Air Force Contract No. FA8702-15-D-0001 or FA8702-25-D-B002. Any opinions, findings, conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Under Secretary of War for Research and Engineering.