Method for automatically identifying resonance
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G01N-029/12
G01N-029/22
G01N-029/44
H02P-023/14
G01N-029/42
출원번호
US-0259286
(2016-09-08)
등록번호
US-10184917
(2019-01-22)
발명자
/ 주소
Tian, Gang
Knaack, Chris
출원인 / 주소
LINESTREAM TECHNOLOGIES
대리인 / 주소
Amin, Turocy & Watson LLP
인용정보
피인용 횟수 :
0인용 특허 :
19
초록▼
A resonance estimation system implements resonance detection methods that can obtain accurate estimates of a motion system's resonance and amplitude without the need for a high-resolution encoder or high-frequency sampling. The system solves for resonance information by removing the slow motion dyna
A resonance estimation system implements resonance detection methods that can obtain accurate estimates of a motion system's resonance and amplitude without the need for a high-resolution encoder or high-frequency sampling. The system solves for resonance information by removing the slow motion dynamics and fast torque control dynamics from the measured speed transfer function in order to obtain resonance transfer function. Smoothing functions are applied to the obtained resonance frequency response data to remove spikes and obtain relatively smooth gain and phase curves for the resonance frequency response. The system then applies a searching algorithm to determine the locations of the phase peaks in the phase curve data, and uses these phase peak locations to locate the gain peaks in the gain curve data, which correspond to the resonance frequencies and amplitudes. This approach allows the gain peaks to be located even when analyzing non-ideal gain curves that are degraded by noise.
대표청구항▼
1. A system, comprising: a memory;a processor configured to execute components stored on the memory, the components comprising: a torque command generator configured to generate torque command data;a noise injection component configured to inject a noise signal into the torque command data to yield
1. A system, comprising: a memory;a processor configured to execute components stored on the memory, the components comprising: a torque command generator configured to generate torque command data;a noise injection component configured to inject a noise signal into the torque command data to yield noise-injected torque command data;a velocity monitoring component configured to obtain speed data representing a speed of a motion system over time in response to a torque command signal generated based on the noise-injected torque command data; anda resonance determination component configured to generate, based on the speed data and the noise-injected torque command data, resonance frequency response data comprising resonance gain frequency response data and resonance phase frequency response data, and to determine a resonance frequency of the motion system based on a determined frequency of a phase peak in the resonance phase frequency response data. 2. The system of claim 1, wherein the resonance determination component is further configured to perform a high-pass filtering and a fast Fourier transform on the noise-injected torque command data and the speed data to yield transformed torque data and transformed speed data, respectively, and to generate the resonance frequency response data based on the transformed torque data and the transformed speed data. 3. The system of claim 1, wherein the resonance determination component is configured to generate frequency response data for the motion system based on the noise-injected torque command data and the speed data, and to remove slow motion dynamics data and fast torque control dynamics data for the motion system from the frequency response data to yield the resonance frequency response data. 4. The system of claim 3, wherein the resonance determination component is further configured to identify noise in the resonance frequency response data based on a comparison of the resonance gain frequency response data and the resonance phase frequency response data, and to remove the noise from the resonance frequency response data prior to determining the resonance frequency. 5. The system of claim 3, wherein the resonance determination component is further configured to estimate an inertia and a viscous friction coefficient of the motion system based on a measured velocity of the motion system in response to a stable torque command signal, and to determine the slow motion dynamics data based on the inertia and the viscous friction coefficient. 6. The system of claim 1, wherein the resonance determination component is configured to identify the phase peak in the resonance phase frequency response data based on an analysis of the resonance phase frequency response data, to identify a gain peak in the resonance gain frequency response data based on the frequency of the phase peak, and to determine the resonance frequency based on a frequency of the gain peak. 7. The system of claim 6, wherein the resonance determination component is further configured to identify a resonance amplitude based on a magnitude of the gain peak. 8. The system of claim 1, further comprising a noise measurement component configured to measure a baseline noise level in a speed signal measured from the motion system in response to another torque command signal that does not include the noise signal, wherein the noise injection component is configured to set a magnitude of the noise signal based on the baseline noise level. 9. A method, comprising: generating, by a system comprising at least one processor, torque command data representing a constant torque command;adding, by the system, noise data to the torque command data to yield noise-injected torque command data;generating, by the system, a torque command signal based on the noise-injected torque command data;measuring, by the system, speed data representing a speed of a motion system over time in response to the torque command signal;generating, by the system based on the speed data and the noise-injected torque command data, resonance frequency response data comprising resonance gain frequency response data and resonance phase frequency response data;identifying, by the system, a phase peak in the resonance phase frequency response data; anddetermining, by the system, a resonance frequency of the motion system based on a determined frequency of the phase peak. 10. The method of claim 9, wherein the generating the resonance frequency response data comprises: performing a first high-pass filtering and a first fast Fourier transform on the noise-injected torque command data to yield transformed torque data;performing a second high-pass filtering and a second fast Fourier transform on the speed data to yield transformed speed data; andgenerating the resonance gain frequency response data and the resonance phase frequency response data based on the transformed torque data and the transformed speed data. 11. The method of claim 9, wherein the generating the resonance frequency response data comprises: generating frequency response data for the motion system based on the noise-injected torque command data and the speed data; andremoving slow motion dynamics data for the motion system and fast torque control dynamics data for the motion system from the frequency response data to yield the resonance frequency response data. 12. The method of claim 11, further comprising: identifying, by the system, noise in the resonance frequency response data based on a comparison of the resonance gain frequency response data and the resonance phase frequency response data; andremoving, by the system, the noise from the resonance frequency response data prior to the identifying of the phase peak. 13. The method of claim 11, wherein the removing the slow motion dynamic data comprises: estimating an inertia and a viscous friction coefficient of the motion system based on a measured velocity of the motion system in response to a stable torque command signal; anddetermining the slow motion dynamic data based on the inertia and the viscous friction coefficient. 14. The method of claim 11, wherein the generating the resonance frequency response data further comprises: performing multiple iterations of the generating of the frequency response data and the removing of the slow motion dynamics data and the fast torque control dynamics data from the frequency response data to yield multiple sets of resonance frequency response data; andgenerating the resonance frequency response data based on an averaging of the multiple sets of the resonance frequency response data. 15. The method of claim 9, wherein the determining the resonance frequency of the motion system comprises: identifying a gain peak in the resonance gain frequency response data based on the determined frequency of the phase peak; anddetermining the resonance frequency based on a frequency of the gain peak. 16. The method of claim 15, further comprising determining, by the system, a resonance amplitude of the motion system based on a magnitude of the gain peak. 17. The method of claim 9, further comprising: measuring, by the system, a baseline noise level in a speed signal measured from the motion system in response to another torque command signal that does not include the noise signal; andsetting, by the system, a magnitude of the noise signal based on the baseline noise level.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (19)
Lyon Richard H. (Belmont MA) Chai Jangbom (Seoul KRX) Lang Jeffrey H. (Sudbury MA) Hagman Wayne H. (Belmont MA) Umans Stephen D. (Belmont MA) Saarela Olli J. (Helsinki FIX), Apparatus and method for non-invasive diagnosis and control of motor operated valve condition.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.