Methods and systems for analyzing engine unbalance conditions
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G01M-001/38
G01M-001/00
출원번호
US-0750294
(2003-12-30)
등록번호
US-7321809
(2008-01-22)
발명자
/ 주소
Vian,John L.
Travis,Matt H.
출원인 / 주소
The Boeing Company
대리인 / 주소
Richardson, P.S.,Robert R.
인용정보
피인용 횟수 :
5인용 특허 :
12
초록▼
Methods and systems for analyzing engine unbalance conditions are disclosed. In one embodiment, a method includes receiving vibrational data from a plurality of locations distributed over an engine and a surrounding engine support structure, and inputting the vibrational data into a neural network i
Methods and systems for analyzing engine unbalance conditions are disclosed. In one embodiment, a method includes receiving vibrational data from a plurality of locations distributed over an engine and a surrounding engine support structure, and inputting the vibrational data into a neural network inverse model. The neural network inverse model establishes a relationship between the vibrational data and an unbalance condition of the engine, and outputs diagnostic information indicating the unbalance condition of the engine. In a further embodiment, a method further includes subjecting the vibrational data to a Fast Fourier Transformation to extract a desired once per revolution vibrational data prior to input to the neural network inverse model.
대표청구항▼
What is claimed is: 1. A method of analyzing an engine unbalance condition, the method comprising: receiving vibrational data from a plurality of locations distributed over at least one of an engine and a surrounding engine support structure; inputting the vibrational data into an artificial neural
What is claimed is: 1. A method of analyzing an engine unbalance condition, the method comprising: receiving vibrational data from a plurality of locations distributed over at least one of an engine and a surrounding engine support structure; inputting the vibrational data into an artificial neural network control volume (ANNCV); using a neural network inverse model, establishing a relationship between the vibrational data from the plurality of locations and an unbalance condition of the engine; and outputting diagnostic information from the ANNCV, the diagnostic information indicating the unbalance condition of the engine, the diagnostic information including an unbalance magnitude and an angular location as a function of a rotational frequency of the engine. 2. The method of claim 1, wherein the vibrational data includes at least one measurement of component displacement, component velocity, component acceleration, sound pressure, and acoustic noise. 3. The method of claim 1, wherein inputting the vibrational data into the ANNCV includes inputting the vibrational data in a time domain format into the ANNCV. 4. The method of claim 1, wherein inputting the vibrational data into the ANNCV includes inputting the vibrational data in a complex frequency domain format into the ANNCV. 5. The method of claim 1, further comprising subjecting the vibrational data to a Pre-processing Transformation. 6. The method of claim 5, wherein the Pre-processing Transformation includes a Fourier Transform. 7. The method of claim 5, wherein the Pre-processing Transformation includes Wavelet Transforms. 8. The method of claim 5, wherein subjecting the vibrational data to a Pre-processing Transformation consists of applying a Fourier Transformation that includes framing an FFT block size using the once per revolution signal such that leakage effects are at least partially reduced, which is also known as order tracking. 9. The method of claim 5, wherein inputting the vibrational data into the ANNCV includes inputting the vibrational data in a complex frequency domain format into the ANNCV. 10. The method of claim 1, wherein outputting diagnostic information from the ANNCV includes outputting fan unbalance and angular location data, and low pressure turbine unbalance and angular location data. 11. The method of claim 1, wherein establishing a relationship between the vibrational data from the plurality of locations and an unbalance condition of the engine includes establishing a relationship between the vibrational data from the plurality of locations and an unbalance condition of the engine using at least one of a multilayer perceptron neural network mode, and a support vector machine neural network model. 12. The method of claim 1, further comprising training the neural network inverse model. 13. The method of claim 12, wherein training the neural network inverse model includes adjusting model parameters such that application of a set of inputs and outputs reaches a desired state of definition defined by acceptable error tolerances. 14. The method of claim 12, wherein training the neural network inverse model includes inputting vibrational data to the ANNCV generated by an empirical engine model. 15. The method of claim 12, wherein training the neural network inverse model includes inputting vibrational data to the ANNCV generated using an engine that is subject to residual unbalances and to applied trial weight unbalances. 16. The method of claim 12, wherein training the neural network inverse model includes scaling the vibrational training data prior to inputting into the ANNCV. 17. A computer program product for analyzing an engine unbalance condition, the computer program product comprising: a first computer program portion adapted to receive vibrational data from a plurality of locations distributed over at least one of an engine and surrounding engine support structure; a second computer program portion adapted to input the vibrational data into an artificial neural network control volume (ANNCV); a third computer program portion adapted to establish a relationship between the vibrational data from the plurality of locations and an unbalance condition of the engine using a neural network inverse model; and a fourth computer program portion adapted to output diagnostic information from the neural network inverse model, the diagnostic information indicating the unbalance condition of the engine by indicating a quantity and an angular position of corrective engine balance weights to achieve desirable vibrational characteristics at selected aircraft component and cabin locations. 18. The computer program product of claim 17, wherein the fourth computer program portion is adapted to output a vibrational magnitude as a function of a rotational frequency of the engine. 19. The computer program product of claim 17, wherein the second computer program portion is adapted to input the vibrational data in a time domain format into the neural network inverse model. 20. The computer program product of claim 17, wherein the second computer program portion is adapted to input the vibrational data in a complex frequency domain format into the neural network inverse model. 21. The computer program product of claim 17, wherein at least one of the first, second, and third computer program portions is adapted to subject the vibrational data to a Fourier Transformation. 22. The computer program product of claim 17, wherein at least one of the first, second, and third computer program portions is adapted to extract a desired once per revolution vibrational data for order tracking signal processing purposes. 23. The computer program product of claim 17, wherein at least one of the first, second, and third computer program portions is adapted to subject the vibrational data to a Wavelet Transformation. 24. The computer program product of claim 17, wherein the third computer program portion is adapted to establish a relationship between the vibrational data from the plurality of locations and an unbalance condition of the engine using at least one of a multilayer perceptron neural network model and a support vector machine neural network model. 25. The computer program product of claim 17, wherein the third computer program portion is adapted to establish a relationship between the vibrational data from a plurality of locations within one defined area to that of a plurality of locations within another defined area using at least one of a multilayer perceptron neural network model and a support vector machine neural network model. 26. The computer program product of claim 17, wherein the third computer program portion is adapted to be trained. 27. The computer program product of claim 26, wherein the third computer program portion is adapted to be trained including adjusting model parameters such that application of a set of inputs and outputs reaches a desired state of definition defined by acceptable error tolerances. 28. The computer program product of claim 26, wherein the third computer program portion is adapted to be trained including inputting vibrational data generated using an engine that is subject to at least one of residual unbalances and applied trial weight unbalances. 29. The computer program product of claim 26, wherein the third computer program portion is adapted to be trained including scaling the vibrational training data prior to inputting into the neural network inverse model. 30. A method of analyzing an engine unbalance condition, the method comprising: collecting at least one vibrational data signal from a plurality of sensors distributed over at least one engine and a surrounding engine support structure; inputting at least one signal into an artificial neural network control volume (ANNCV); analyzing the signal to identify an engine unbalance condition; and outputting diagnostic information from the ANNCV, derived from the unbalance condition, the diagnostic information including an unbalance magnitude and an angular location as a function of a rotational frequency of the engine.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (12)
Lo, James T.; Yu, Lei, Active acoustic and structural vibration control without online controller adjustment and path modeling.
Tiernan Timothy A. (Livonia MI) Geddes Earl R. (Livonia MI) Mollon Mark L. (Livonia MI), Integrated active vibration cancellation and machine diagnostic system.
Petsche Thomas (Neshanic Station NJ) Garrett Charles (Plainsboro NJ), Method for monitoring an electric motor and detecting a departure from normal operation.
Bozich Daniel J. (San Diego CA) MacKay H. Bruce (San Diego CA) Eggert Jay A. (San Diego CA) Muenchau Ernest E. (Vista CA), Neurocontrolled adaptive process control system.
Bozich Daniel J. (San Diego CA) MacKay H. Bruce (San Diego CA) Eggert Jay A. (San Diego CA) Muenchau Ernest E. (Vista CA), Neurocontrolled adaptive process control system.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.