Predictive Fault Detection for Unmanned Communications Facilities

Communications systems depend on the health of their equipment and components to perform reliably over extended periods of time. Maintaining the health of equipment in remote, unmanned communications terminals is especially critical, because unscheduled repairs take longer and have more impact on system availability than is the case for manned facilities with short logistics lines. Instrumental Sciences, Inc. (ISI) proposes new predictive fault detection methods, combining a model-based architecture with multivariate signal processing, trend analysis, identification techniques, and situation assessment algorithms. The proposed approach integrates the techniques into a cognitive processing architecture with decision algorithms that provides timely maintenance action recommendations for maintainers to perform before actual failures occur. The predictive fault detection capability enables predictive condition-based maintenance, which minimizes downtime and use resources efficiently, to replace schedule-based preventive maintenance, which is costly and impacts readiness. Primary objectives of the proposed Phase I effort are to (1) demonstrate the feasibility of the proposed predictive fault detection approach for unmanned communications facilities and (2) evaluate the potential system availability improvements and life cycle cost savings of this predictive fault detection capability over conventional schedule based maintenance and other prognostics methods.