The Prognostics Framework is a tool set used to develop and deploy a comprehensive health monitoring capability for any system. It uses powerful model-based reasoning techniques to continuously assess system condition and to identify existing and impending faults.

Use of the Prognostics Framework institutes an information framework that organizes relevant data related to 1) the condition of the system, 2) the system's ability to perform required functions over specific time intervals, and 3) the need for maintenance actions and repair parts.

On-Line Condition Monitoring 

The Prognostics Framework uses a powerful, dynamic reasoning capability to perform continuous on-line assessment of a system's condition, including system health over future time intervals. This reasoning capability is deployed as an integral part of the system. During system operation, the Prognostics Framework reads operational data, built-in test (BIT) data and sensor data, and uses its reasoning capability to continuously assess the system's health, identifying existing faults and impending failure events.

System Model Development

The Prognostics Framework reasons by correlating operational, sensor, and BIT data to a design-based model of the system. The model is generated using the Prognostics Framework development tool. The underlying design structure can be imported directly from CAD data (EDIF netlists) or by building a model of the system. The developer defines the test and sensor data that will be available during system operation, and defines the relationship between that data and any faults that the data infers, or coverage. The developer also defines algorithms and mathematical processing to be applied to the data to characterize the onset of a failure condition. If a separate (third-party) prediction technique is available, this can be integrated, either directly, or by capturing the result of its processing into the Prognostics Framework. 

To learn more, select from the menu at the left. 

To view a demonstration of the run-time, click here.