Internet of Things-enabled CBM, Diagnosis, and Prognostics for Navy Equipment
Abstract
Mission readiness and longevity of Navy fleet heavily depends on how its equipment is well maintained. Currently, maintenance still accounts for a major portion of the ownership costs for DoD. Fundamentally, the limitations lie in (i) the lack of effective understanding of the system health status in real time; and (ii) the inability to accurately model the degradation and predict the failure time of each in-field unit. If these limitations can be overcome, cost-effective maintenance strategy can then be planned accordingly, only at the time when maintenance is necessary and before the predicted catastrophic failure occurs. Fortunately, Internet of Things (IoT) technology has enabled a new paradigm in which massive and heterogeneous sensors are deployed in the Navy ships and their associated systems such as aircraft and unmanned vehicles. This provides an unprecedented opportunity for (i) closer monitoring of a unit’s health status; (ii) quicker fault diagnosis; (iii) more accurate prognostics regarding the unit’s remaining lifetime; and (iv) proactive maintenance decisions better aligned to future conditions and performance. This proposal aims to bridge the knowledge gaps from the rich data to smart condition-based monitoring, diagnosis and prognostics for navy equipment driven by the IoT technology. Specifically, four inter-related research tasks will be developed: (i) The PI will propose a novel concept of “health index (HI)” that directly combines multiple and heterogeneous streams of sensor data to accurately characterize and visualize the health condition of a unit in real time. (ii) Then, a fault diagnostic index will be established, which can quickly identify the failure mode of the degraded units during condition monitoring. (iii) Next, the PI will extend the proposed data fusion model to further characterize the effect of multiple environmental conditions on the health status of a unit. (iv) With the constructed HI available, the PI will then investigate a sensoryupdated prognostic method with uncertainty quantification to accurately predict the failure time of each in-field unit. If successful, this project will significantly enrich the methodological base and the toolbox of the ONR logistics by providing a paradigm shift on condition monitoring, prognostics, and maintenance driven by the IoT technology. This transformative research will significantly enhance DoD’s competitiveness, such as reduced operational cost, improved mission reliability and safety for warfighters, and enhanced asset utilization and operational readiness, leading to unprecedented advantages across all levels of military operations and over adversaries.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 10, 2017
- Source ID
- N000141712261
Entities
People
- Kaibo Liu
Organizations
- Office of Naval Research
- United States Navy
- University of Wisconsin System