Probabilistic Operations Warranted for Energy Reliability Evaluation and Diagnostics (POWERED)

Period of Performance: 09/30/2016 - 03/29/2017

$100K

Phase 1 SBIR

Recipient Firm

Charles River Analytics, Inc.
625 Mount Auburn Street Array
Cambridge, MA 02138
Firm POC
Principal Investigator

Abstract

Back-up power equipment, such as engine-driven generator systems, undergo performance degradation over time that can prevent them from functioning properly in critical situations. Ensuring the reliability of back-up power systems during power outages not only allows for the continuation of normal operations, but also can mitigate potential threats to national security. Improved diagnostic and prognostic approaches that more accurately determine the reliability and performance of generator and power distribution equipment can advise the maintenance or replacement of power systems to prevent failures in critical situations, protecting against future threats. The Probabilistic Operations Warranted for Energy Reliability Evaluation and Diagnostics (POWERED) application uses rich, modular probabilistic modeling that advises reliability and maintenance of generators. POWERED uses a distributed network of continuous-time dynamic probabilistic relational models (PRMs) to accurately and efficiently diagnose and predict system- and component-level failures due to long-term degradation.