System and method for optimal sensor placement

1 . A computing device, the computing device comprising:

a memory storing computer readable instructions; and

processing circuitry configured to execute the computer readable instructions to cause the computing device to,

receive a plurality of fault conditions and an initial sensor suite associated with at least one object to be simulated, the initial sensor suite including a plurality of candidate physical sensors for the at least one object,

perform failure mode analysis of the initial sensor suite, the performing the failure mode analysis including generating at least one dependency-matrix (D-matrix) based on the plurality of fault conditions and the plurality of candidate physical sensors,

generate a recommended sensor suite associated with the at least one object based on results of the failure mode analysis, the recommended sensor suite including at least one recommended sensor, the at least one recommended sensor being a subset of the plurality of candidate physical sensors,

receive a simulation file corresponding to the at least one object to be simulated, the simulation file including baseline condition data of the at least one object to be simulated and at least one fault condition data of the at least one object to be simulated,

modify the simulation file to include the at least one recommended sensor, and

display the at least one object to be simulated and the at least one recommended sensor on a display screen using the modified simulation file, the at least one recommended sensor displayed in a recommended location relative to the at least one object, the recommended location determined during the failure mode analysis.

2 . The computing device of claim 1 , wherein the at least one object is at least one of:

a power plant, a turbine, a power grid, a jet engine, a factory environment, an environmental system, a HVAC system, a computational system, or any combinations thereof.

3 . The computing device of claim 1 , wherein the processing circuitry is further configured to execute the computer readable instructions to cause the computing device to:

determine at least one virtual candidate sensor based on at least two candidate physical sensors of the plurality of candidate physical sensors; and

add the at least one virtual candidate sensor to the initial sensor suite and the at least one D-matrix.

4 . The computing device of claim 3 , wherein the processing circuitry is further configured to execute the computer readable instructions to cause the computing device to:

receive information, the information including at least one of user preference information, system design preference information, sensor preference information, sensor capability information, sensor group information, sensor life-cycle cost information, fault criticality information, fault failure rate information, or any combinations thereof;

determine sensor-fault condition correlation scores for each candidate physical sensor and each virtual candidate sensor included in the at least one D-matrix based on the received information; and

select the at least one recommended sensor from the plurality of candidate physical sensors and the at least one virtual candidate sensor based on the determined sensor-fault condition correlation scores.

5 . The computing device of claim 3 , wherein the processing circuitry is further configured to execute the computer readable instructions to cause the computing device to perform the failure mode analysis of the initial sensor suite by:

selecting a plurality of test sensor suites from the plurality of candidate physical sensors and the at least one virtual candidate sensor, each of the test sensor suites being a different combination of sensors from the plurality of candidate physical sensors and the at least one virtual candidate sensor;

calculating a fault detection score for each of the plurality of test sensor suites for each of the plurality of fault conditions;

determining an overall fault detection coverage score of each of the plurality of fault conditions based on the calculated fault detection scores for each of the plurality of test sensor suites associated with the respective fault condition and a desired threshold fault detection coverage value associated with the respective fault condition; and

selecting the at least one recommended sensor from the plurality of test sensor suites based on the determined overall fault detection coverage score of each of the plurality of fault conditions.

6 . The computing device of claim 3 , wherein the processing circuitry is further configured to execute the computer readable instructions to cause the computing device to perform the failure mode analysis of the initial sensor suite by:

selecting a plurality of test sensor suites from the plurality of candidate physical sensors and the at least one virtual candidate sensor, each of the test sensor suites being a different combination of sensors from the plurality of candidate physical sensors and the at least one virtual candidate sensor;

calculating a fault isolation coverage score for each of the plurality of test sensor suites for each of the plurality of fault conditions; and

selecting the at least one recommended sensor from the plurality of test sensor suites based on the calculated fault isolation coverage score of each of the plurality of fault conditions and a desired threshold fault isolation coverage value.

7 . The computing device of claim 3 , wherein the processing circuitry is further configured to execute the computer readable instructions to cause the computing device to perform the failure mode analysis of the initial sensor suite by:

for each candidate physical sensor and each virtual candidate sensor, simulating life-cycle costs using the respective candidate sensor for at least one physical location corresponding to the at least one object associated with each of the plurality of fault conditions based on initial sensor cost of the respective candidate sensor, maintenance cost of the respective candidate sensor, and savings from avoidance of the respective fault condition;

determining sensor-fault condition correlation scores for the plurality of candidate sensors for the at least one physical location based on the simulated life-cycle costs; and

selecting the at least one recommended sensor from the plurality of candidate physical sensors and the at least one virtual candidate sensor based on the determined sensor-fault condition correlation scores.

8 . The computing device of claim 3 , wherein the processing circuitry is further configured to execute the computer readable instructions to cause the computing device to:

for each candidate physical sensor and each virtual candidate sensor,

generate balanced baseline conditions simulation data for baseline conditions for the respective sensor,

generate balanced fault condition simulation data for each fault condition for the respective sensor, and

determine sensor-fault condition correlation score for the respective sensor based on the generated balanced baseline conditions simulation data and the generated balanced fault condition simulation data; and

select the at least one recommended sensor from the plurality of candidate physical sensors and the at least one virtual candidate sensor based on the determined sensor-fault condition correlation scores.

9 . A method of operating a computing device, the method comprising:

receiving a plurality of fault conditions and an initial sensor suite associated with at least one object to be simulated, the initial sensor suite including a plurality of candidate physical sensors for the at least one object;

performing failure mode analysis of the initial sensor suite, the performing the failure mode analysis including generating at least one dependency-matrix (D-matrix) based on the plurality of fault conditions and the plurality of candidate physical sensors;

generating a recommended sensor suite associated with the at least one object based on results of the failure mode analysis, the recommended sensor suite including at least one recommended sensor, the at least one recommended sensor being a subset of the plurality of candidate physical sensors;

receiving a simulation file corresponding to the at least one object to be simulated, the simulation file including baseline condition data of the at least one object to be simulated and at least one fault condition data of the at least one object to be simulated;

modifying the simulation file to include the at least one recommended sensor; and

displaying the at least one object to be simulated and the at least one recommended sensor on a display screen using the modified simulation file, the at least one recommended sensor displayed in a recommended location relative to the at least one object, the recommended location determined during the failure mode analysis.

10 . The method of claim 9 , wherein

the at least one object is at least one of,

a power plant, a turbine, a power grid, a jet engine, a factory environment, an environmental system, a HVAC system, a computational system, or any combinations thereof.

11 . The method of claim 9 , the method further comprising:

determining at least one virtual candidate sensor based on at least two candidate physical sensors of the plurality of candidate physical sensors; and

adding the at least one virtual candidate sensor to the initial sensor suite and the at least one D-matrix.

12 . The method of claim 11 , the method further comprising:

receiving information, the information including at least one of user preference information, system design preference information, sensor preference information, sensor capability information, sensor group information, sensor life-cycle cost information, fault criticality information, fault failure rate information, or any combinations thereof;

determining sensor-fault condition correlation scores for each candidate physical sensor and each virtual candidate sensor included in the at least one D-matrix based on the received information; and

selecting the at least one recommended sensor from the plurality of candidate physical sensors and the at least one virtual candidate sensor based on the determined sensor-fault condition correlation scores.

13 . The method of claim 12 , the method further comprising:

selecting a plurality of test sensor suites from the plurality of candidate physical sensors and the at least one virtual candidate sensor, each of the test sensor suites being a different combination of sensors from the plurality of candidate physical sensors and the at least one virtual candidate sensor;

calculating a fault detection score for each of the plurality of test sensor suites for each of the plurality of fault conditions;

determining an overall fault detection coverage score of each of the plurality of fault conditions based on the calculated fault detection scores for each of the plurality of test sensor suites associated with the respective fault condition and a desired threshold fault detection coverage value associated with the respective fault condition; and

selecting the at least one recommended sensor from the plurality of test sensor suites based on the determined overall fault detection coverage score of each of the plurality of fault conditions.

14 . The method of claim 12 , the method further comprising:

selecting a plurality of test sensor suites from the plurality of candidate physical sensors and the at least one virtual candidate sensor, each of the test sensor suites being a different combination of sensors from the plurality of candidate physical sensors and the at least one virtual candidate sensor;

calculating a fault isolation coverage score for each of the plurality of test sensor suites for each of the plurality of fault conditions; and

selecting the at least one recommended sensor from the plurality of test sensor suites based on the calculated fault isolation coverage score of each of the plurality of fault conditions and a desired threshold fault isolation coverage value.

15 . The method of claim 12 , the method further comprising:

for each candidate physical sensor and each virtual candidate sensor, simulating life-cycle costs using the respective candidate sensor for at least one physical location corresponding to the at least one object associated with each of the plurality of fault conditions based on initial sensor cost of the respective candidate sensor, maintenance cost of the respective candidate sensor, and savings from avoidance of the respective fault condition;

determining the sensor-fault condition correlation scores for the plurality of candidate sensors for the at least one physical location based on the simulated life-cycle costs; and

selecting the at least one recommended sensor from the plurality of candidate physical sensors and the at least one virtual candidate sensor based on the determined sensor-fault condition correlation scores.

16 . The method of claim 12 , the method further comprising:

for each candidate physical sensor and each virtual candidate sensor,

generating balanced baseline conditions simulation data for baseline conditions for the respective sensor,

generating balanced fault condition simulation data for each fault condition for the respective sensor, and

determining sensor-fault condition correlation score for the respective sensor based on the generated balanced baseline conditions simulation data and the generated balanced fault condition simulation data; and

selecting the at least one recommended sensor from the plurality of candidate physical sensors and the at least one virtual candidate sensor based on the determined sensor-fault condition correlation scores.

17 . A non-transitory computer readable medium storing computer readable instructions, which when executed by processing circuitry of a computing device, causes the computing device to:

receive a plurality of fault conditions and an initial sensor suite associated with at least one object to be simulated, the initial sensor suite including a plurality of candidate physical sensors for the at least one object;

perform failure mode analysis of the initial sensor suite, the performing the failure mode analysis including generating at least one dependency-matrix (D-matrix) based on the plurality of fault conditions and the plurality of candidate physical sensors;

generate a recommended sensor suite associated with the at least one object based on results of the failure mode analysis, the recommended sensor suite including at least one recommended sensor, the at least one recommended sensor being a subset of the plurality of candidate physical sensors;

receive a simulation file corresponding to the at least one object to be simulated, the simulation file including baseline condition data of the at least one object to be simulated and at least one fault condition data of the at least one object to be simulated;

modify the simulation file to include the at least one recommended sensor; and

display the at least one object to be simulated and the at least one recommended sensor on a display screen using the modified simulation file, the at least one recommended sensor displayed in a recommended location relative to the at least one object, the recommended location determined during the failure mode analysis.

18 . The non-transitory computer readable medium of claim 17 , wherein the computing device is further caused to:

receive information, the information including at least one of user preference information, system design preference information, sensor preference information, sensor capability information, sensor group information, sensor life-cycle cost information, fault criticality information, fault failure rate information, or any combinations thereof;

determine sensor-fault condition correlation scores for each candidate physical sensor included in the at least one D-matrix based on the received information; and

select the at least one recommended sensor from the plurality of candidate physical sensors and at least one virtual candidate sensor based on the determined sensor-fault mode correlation scores.

19 . The non-transitory computer readable medium of claim 17 , wherein the computing device is further caused to perform the failure mode analysis of the initial sensor suite by:

determining at least one virtual candidate sensor based on at least two candidate physical sensors of the plurality of candidate physical sensors;

adding the at least one virtual candidate sensor to the initial sensor suite and the at least one D-matrix;

selecting a plurality of test sensor suites from the plurality of candidate physical sensors and the at least one virtual candidate sensor, each of the test sensor suites being a different combination of sensors from the plurality of candidate physical sensors and the at least one virtual candidate sensor;

calculating a fault detection score for each of the plurality of test sensor suites for each of the plurality of fault conditions;

determining an overall fault detection coverage score of each of the plurality of fault conditions based on the calculated fault detection scores for each of the plurality of test sensor suites associated with the respective fault condition and a desired threshold fault detection coverage value associated with the respective fault condition; and

selecting the at least one recommended sensor from the plurality of test sensor suites based on the determined overall fault detection coverage score of each of the plurality of fault conditions.

20 . The non-transitory computer readable medium of claim 17 , wherein the computing device is further caused to perform the failure mode analysis of the initial sensor suite by:

determining at least one virtual candidate sensor based on at least two candidate physical sensors of the plurality of candidate physical sensors;

adding the at least one virtual candidate sensor to the initial sensor suite and the at least one D-matrix;

for each candidate physical sensor and each virtual candidate sensor, simulating life-cycle costs using the respective candidate sensor for at least one physical location corresponding to the at least one object associated with each of the plurality of fault conditions based on initial sensor cost of the respective candidate sensor, maintenance cost of the respective candidate sensor, and savings from avoidance of the respective fault condition;

determining sensor-fault condition correlation scores for the plurality of candidate sensors for the at least one physical location based on the simulated life-cycle costs; and

selecting the at least one recommended sensor from the plurality of candidate physical sensors and the at least one virtual candidate sensor based on the determined sensor-fault condition correlation scores.