초록 ▼

A system and method is provided for detecting anomalies in turbine engines emanating from the main shaft and/or main shaft bearings. The anomaly detection system includes a sensor data processor and a matrix analysis mechanism. The sensor data processor receives engine sensor data, including main e

A system and method is provided for detecting anomalies in turbine engines emanating from the main shaft and/or main shaft bearings. The anomaly detection system includes a sensor data processor and a matrix analysis mechanism. The sensor data processor receives engine sensor data, including main engine speed data during spin down, and formats the engine sensor data into an appropriate matrix. The matrix analysis mechanism receives the sensor data matrix and performs a singular value analysis on the sensor data matrix to detect potential anomalies in the turbine engine main shaft and/or bearings. The output of the matrix analysis mechanism is passed to a diagnostic system where further evaluation of the anomaly detection determination can occur.

대표청구항 ▼

The invention claimed is: 1. An anomaly detection system for detecting anomalies in a plurality of turbine engines, the anomaly detection system comprising: a sensor data processor, the sensor data processor adapted to receive engine sensor data from the plurality of turbine engines and format the

The invention claimed is: 1. An anomaly detection system for detecting anomalies in a plurality of turbine engines, the anomaly detection system comprising: a sensor data processor, the sensor data processor adapted to receive engine sensor data from the plurality of turbine engines and format the engine sensor data into a sensor data matrix, where the sensor data matrix of engine sensor data comprises a multi-dimensional array with rows and columns; and a matrix analysis mechanism, the matrix analysis mechanism adapted to perform a singular value analysis on the sensor data matrix to compare the sensor data from plurality of turbine engines and detect potential anomalies in the plurality of turbine engines, and wherein the anomaly detection system is further adapted to generate a notification of detected potential anomalies in the plurality of turbine engines. 2. The system of claim 1 wherein the sensor data processor formats the sensor data into the sensor data matrix by placing sensor data from each of the plurality of turbine engines into a corresponding row in the sensor data matrix. 3. The system of claim 1 wherein the sensor data includes data from multiple spin down occurrences taken after fuel flow has been shut off, and wherein the sensor data processor formats the sensor data into the sensor data matrix by placing sensor data from each of the multiple spin down occurrences into a corresponding row in the sensor data matrix. 4. The system of claim 1 wherein the sensor data comprises main shaft speed data. 5. The system of claim 1 wherein the sensor data comprises main shaft speed data taken during turbine engine spin-down, wherein engine spin-down occurs for each turbine engine after fuel flow has been shut off. 6. The system of claim 1 wherein the matrix analysis mechanism is adapted to perform a singular value analysis on the sensor data matrix to detect potential anomalies in the plurality of turbine engines by calculating a singular value from the sensor data and comparing the singular value to a threshold value. 7. The system of claim 1 wherein the matrix analysis mechanism is adapted to perform a singular value analysis on the sensor data matrix to detect potential anomalies in the plurality of turbine engines by calculating a covariance matrix from the sensor data matrix and by calculating at least a second singular value from the covariance matrix and comparing the second singular value to a threshold value. 8. The system of claim 5 wherein the main shaft speed data taken during turbine engine spin-down comprises data collected from the plurality of turbine engines between two defined main shaft speeds after the fuel flow has been shut off. 9. The system of claim 6 wherein the matrix analysis mechanism calculates the singular value using a QR decomposition for symmetric matrices. 10. The system of claim 7 wherein the notification of detected potential anomalies is made after a predetermined number of successive second singular values exceed the threshold value. 11. A method of detecting anomalies in a plurality of turbine engines, the method comprising the steps of: a) receiving sensor data from the plurality of turbine engines; b) formatting the sensor data into a sensor data matrix, where the sensor data matrix of sensor data comprises a multi-dimensional array with rows and columns; c) performing a singular value analysis on the sensor data matrix to compare the sensor data from the plurality of turbine engines and detect potential anomalies in the plurality of turbine engines; and d) generating a notification of detected potential anomalies in the plurality of turbine engines. 12. The method of claim 11 wherein the step of formatting the sensor data into the sensor data matrix comprises placing the sensor data from each of the plurality of turbine engines into a corresponding row in the sensor data matrix. 13. The method of claim 11 wherein the sensor data includes sensor data from multiple spin down occurrences taken after fuel flow has been shut off, and wherein the step of formatting the sensor data into the sensor data matrix comprises placing sensor data from each of the multiple spin down occurrences into a corresponding row in the sensor data matrix. 14. The method of claim 11 wherein the sensor data comprises main shaft speed data. 15. The method of claim 11 wherein the sensor data comprises main shaft speed data taken during turbine engine spin-down, wherein engine spin-down occurs for each turbine engine after fuel flow has been shut off. 16. The method of claim 11 wherein the step of performing a singular value analysis on the sensor data matrix to compare the sensor data from the plurality of turbine engines and detect potential anomalies in the plurality of turbine engines comprises calculating a singular value from the sensor data and comparing the singular value to a threshold value. 17. The method of claim 11 wherein the step of performing a singular value analysis on the sensor data matrix to compare the sensor data from the plurality of turbine engines and detect potential anomalies in the plurality of turbine engines comprises calculating a covariance matrix from the sensor data matrix and calculating at least a second singular value from the covariance matrix and comparing the second singular value to a threshold value. 18. The method of claim 15 wherein the main shaft speed data taken turbine engine spin-down comprises data collected from the plurality of turbine engines between two defined main shaft speeds after the fuel flow has been shut off. 19. The method of claim 16 wherein the step of calculating a singular value from the sensor data comprises using a QR decomposition for symmetric matrices. 20. The method of claim 17 wherein the step of generating a notification of detected potential anomalies in the plurality of turbine engines comprises generating notification after a predetermined number of successive second singular values exceed the threshold value. 21. A program product comprising: a) an anomaly detection program, the anomaly detection program including: a sensor data processor, the sensor data processor adapted to receive engine sensor data from a plurality of turbine engines and format the engine sensor data into a sensor data matrix, where the sensor data matrix of engine sensor data comprises a multi-dimensional array with rows and columns; and a matrix analysis mechanism, the matrix analysis mechanism adapted to perform a singular value analysis on the sensor data matrix to compare the sensor data from plurality of turbine engines and detect potential anomalies in the plurality of turbine engines, and wherein the anomaly detection program is further adapted to generate a notification of detected potential anomalies in the plurality of turbine engines; and b) computer-readable signal bearing media bearing said anomaly detection program. 22. The program product of claim 21 wherein the sensor data processor formats the sensor data into the sensor data matrix by placing sensor data from each of the plurality of turbine engines into a corresponding row in the sensor data matrix. 23. The program product of claim 21 wherein the sensor data includes data from multiple spin down occurrences taken after fuel flow has been shut off, and wherein the sensor data processor formats the sensor data into the sensor data matrix by placing sensor data from each of the multiple spin down occurrences into a corresponding row in the sensor data matrix. 24. The program product of claim 21 wherein the sensor data comprises main shaft speed data. 25. The program product of claim 21 wherein the sensor data comprises main shaft speed data taken during turbine engine spin-down, wherein engine spin-down occurs for each turbine engine after fuel flow has been shut off. 26. The program product of claim 21 wherein the matrix analysis mechanism is adapted to perform a singular value analysis on the sensor data matrix to detect potential anomalies in the plurality of turbine engines by calculating a singular value from the sensor data and comparing the singular value to a threshold value. 27. The program product of claim 21 wherein the matrix analysis mechanism is adapted to perform a singular value analysis on the sensor data matrix to detect potential anomalies in the plurality of turbine engines by calculating a covariance matrix from the sensor data matrix and by calculating at least a second singular value from the covariance matrix and comparing the second singular value to a threshold value. 28. The program product of claim 25 wherein the main shaft speed data taken during turbine engine spin-down comprises data collected from the plurality of turbine engines between two defined main shaft speeds after the fuel flow has been shut off. 29. The program product of claim 26 wherein the matrix analysis mechanism calculates the singular value using a QR decomposition for symmetric matrices. 30. The program product of claim 27 wherein the notification of detected potential anomalies is made after a predetermined number of successive second singular values exceed the threshold value.