Prognostics monitor for systems that are subject to failure

A method for predicting system failure modes, wherein the system has a plurality of monitor points, each point having an actual parameter value, wherein the system has a plurality of failure modes, the method comprising the steps of:

• providing a range of desired parameter values for each monitor point;
• providing a monitor point comparator function for said monitor points, wherein said monitor point comparator function compares said actual parameter value with said desired parameter value range and generates a parameter deviation signal if said actual parameter value is outside of said desired parameter value range;
• providing a plurality of ambiguity groups, said ambiguity groups having at least one failure mode associated with a monitor point deviation signal;
• obtaining said actual parameter value history over a plurality of time points;
• extrapolating said plurality of time points into the future;
• predicting if said extrapolated time points will generate a deviation signal from said comparator;
• and, if said deviation signal will be generated, generating a predicted deviation signal, predicted time of deviation and outputting said predicted deviation time;
• and
• for each generated predicted parameter deviation signal, searching said plurality of ambiguity groups for at least one ambiguity group that includes said generated predicted parameter deviation signal, and identifying at least one failure mode associated with said predicted generated deviation signal.

A computer program for executing a method for detecting system failure modes, wherein the system has a plurality of monitor points, each point having an actual parameter value, wherein the system has a plurality of failure modes, the method comprising the steps of:

• providing a range of desired parameter values for each monitor point;
• providing a monitor point comparator function for said monitor points, wherein said monitor point comparator function compares said actual parameter value with said desired parameter value range and generates a parameter deviation signal if said actual parameter value is outside of said desired parameter value range;
• providing a plurality of ambiguity groups, said ambiguity groups having at least one failure mode associated with a monitor point deviation signal;
• obtaining said actual parameter value;
• comparing said actual parameter value with said desired parameter value for each monitor point using said monitor point comparator function and generating a parameter deviation signal if said parameter actual value is outside of said desired parameter value range;
• and
• for each generated parameter deviation signal, searching said plurality of ambiguity groups for at least one ambiguity group that includes said generated parameter deviation signal, and identifying at least one failure mode associated with said generated deviation signal.

A computer program for executing a method for predicting system failure modes, wherein the system has a plurality of monitor points, each point having an actual parameter value, wherein the system has a plurality of failure modes, the method comprising the steps of:

• providing a range of desired parameter values for each monitor point;
• providing a monitor point comparator function for said monitor points, wherein said monitor point comparator function compares said actual parameter value with said desired parameter value range and generates a parameter deviation signal if said actual parameter value is outside of said desired parameter value range;
• providing a plurality of ambiguity groups, said ambiguity groups having at least one failure mode associated with a monitor point deviation signal;
• obtaining said actual parameter value history over a plurality of time points;
• extrapolating said plurality of time points into the future;
• predicting if said extrapolated time points will generate a deviation signal from said comparator;
• and, if said deviation signal will be generated, generating a predicted deviation signal, predicted time of deviation and outputting said predicted deviation time;
• and
• for each generated predicted parameter deviation signal, searching said plurality of ambiguity groups for at least one ambiguity group that includes said generated predicted parameter deviation signal, and identifying at least one failure mode associated with said predicted generated deviation signal.

A computer monitoring system for coupling to a system, said computer monitoring system running a program for executing a method for detecting system failure modes, wherein the system has a plurality of monitor points, each point having an actual parameter value, wherein the system has a plurality of failure modes, wherein the computer monitoring system is coupled to said monitor points and able to read said monitor points, the method comprising the steps of:

• providing a range of desired parameter values for each monitor point;
• providing a monitor point comparator function for said monitor points, wherein said monitor point comparator function compares said actual parameter value with said desired parameter value range and generates a parameter deviation signal if said actual parameter value is outside of said desired parameter value range;
• providing a plurality of ambiguity groups, said ambiguity groups having at least one failure mode associated with a monitor point deviation signal;
• obtaining said actual parameter value;
• comparing said actual parameter value with said desired parameter value for each monitor point using said monitor point comparator function and generating a parameter deviation signal if said parameter actual value is outside of said desired parameter value range;
• and
• for each generated parameter deviation signal, searching said plurality of ambiguity groups for at least one ambiguity group that includes said generated parameter deviation signal, and identifying at least one failure mode associated with said generated deviation signal.

A computer monitoring system for coupling to a system, said computer monitoring system running a program for executing a method for predicting system failures, wherein the system has a plurality of monitor points, each point having an actual parameter value, wherein the system has a plurality of failure modes, wherein the computer monitoring system is coupled to said monitor points and able to read said monitor points, the method comprising the steps of:

• providing a range of desired parameter values for each monitor point;
• providing a monitor point comparator function for said monitor points, wherein said monitor point comparator function compares said actual parameter value with said desired parameter value range and generates a parameter deviation signal if said actual parameter value is outside of said desired parameter value range;
• providing a plurality of ambiguity groups, said ambiguity groups having at least one failure mode associated with a monitor point deviation signal;
• obtaining said actual parameter value history over a plurality of time points;
• extrapolating said plurality of time points into the future;
• predicting if said extrapolated time points will generate a deviation signal from said comparator;
• and, if said deviation signal will be generated, generating a predicted deviation signal, predicted time of deviation and outputting said predicted deviation time;
• and
• for each generated predicted parameter deviation signal, searching said plurality of ambiguity groups for at least one ambiguity group that includes said generated predicted parameter deviation signal, and identifying at least one failure mode associated with said predicted generated deviation signal.