IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
UP-0827244
(2007-07-10)
|
등록번호 |
US-7596470
(2009-10-12)
|
발명자
/ 주소 |
|
출원인 / 주소 |
- Advanced Structure Monitoring, Inc.
|
대리인 / 주소 |
Patent Office of Dr. Chung Sik Park
|
인용정보 |
피인용 횟수 :
15 인용 특허 :
32 |
초록
▼
Systems, methods and recordable media for prognosticating a structural condition. A method for prognosticating a structural condition by use of a network having a plurality of diagnostic network patches (DNP), each of the patches being able to operate as at least one of a transmitter patch and a sen
Systems, methods and recordable media for prognosticating a structural condition. A method for prognosticating a structural condition by use of a network having a plurality of diagnostic network patches (DNP), each of the patches being able to operate as at least one of a transmitter patch and a sensor patch, includes the steps of: (a) causing at least one of the patches to emit a transmitter signal and a portion of the other patches to receive a set of sensor signals at a point in time; (b) repeating the step (a) to receive multiple sets of sensor signals at multiple points in time, respectively; and (c) prognosticating, based on the multiple sets of signals, a structural condition of a host structure of the network at a target point in time.
대표청구항
▼
What is claimed is: 1. A computer-implemented method for prognosticating a structural condition by use of a network having a plurality of diagnostic network patches (DNP), each of the patches being able to operate as at least one of an actuator patch and a receiver patch, at least one of the patche
What is claimed is: 1. A computer-implemented method for prognosticating a structural condition by use of a network having a plurality of diagnostic network patches (DNP), each of the patches being able to operate as at least one of an actuator patch and a receiver patch, at least one of the patches being able to operate as both an actuator patch and a receiver patch, the method comprising: (a) causing, by use of a computer processor, the at least one of the patches to emit a transmitter signal and a portion of the other patches to receive a set of sensor signals at a point in time, each said sensor signal being a portion of the transmitter signal emitted by the at least one of the patches and propagated through the structure; (b) repeating the step (a) to receive multiple sets of sensor signals at multiple points in time, respectively; and (c) prognosticating, based on the multiple sets of signals, a structural condition of a host structure of the network at a target point in time. 2. The method at claim 1, wherein the step of prognosticating a structural condition includes: building a dynamic system model of the network, the dynamic system model having at least one system parameter; determining the system parameter for each of the multiple sets of sensor signals; estimating the system parameter at the target point in time based on the determined system parameters for the multiple sets of sensor signals; and predicting the structural condition by use of the estimated system parameter. 3. The method of claim 2, further comprising: generating a set of sensor signals by use of the estimated system parameter. 4. The method of claim 3, further comprising: adding the generated set of sensor signals to the multiple sets of signals thereby to update the multiple sets of signals; and predicting, based on the updated multiple sets of signals, the structural condition at a target point in time. 5. The method of claim 2, wherein the dynamic system model is a state space model or an autoregressive moving average with exogenous input (ARMAX) model. 6. The method of claim 2, wherein the step of determining the system parameter is performed by use of a subspace system identification method. 7. The method of claim 2, wherein the step of estimating the system parameter includes: training a neural network with the determined system parameters; and causing the trained neural network to generate the estimated system parameter. 8. The method of claim 7, wherein the neural network is a recurrent neural network or a feedforward neural network. 9. The method of claim 1, wherein the transmitter signal includes at least one of ultrasonic wave, Lamb wave, vibrational wave, and acoustic wave. 10. A computer-implemented method for prognosticating a structural condition by use of a network having a plurality of sensors, the method comprising: preparing a network that includes a plurality of paths and nodes, each of the sensors corresponding to one of the nodes, each path connecting two of the nodes of the network to thereby form an edge of the network, at least one of the sensors being adapted to emit a transmitter signal that propagates through a host structure of the network, the sensors being adapted to receive the transmitter signal emitted by the at least one of the sensors; causing the sensors to obtain a set of sensor signals generated by an external load applied to the host structure at a point in time; analyzing the set of sensor signals by use of the network to characterize the external load and a variation in a structural condition of the host structure; repeating the steps of causing the sensors to obtain a set of sensor signals and analyzing the set of sensor signals to thereby obtain multiple sets of sensor signals from the sensors and variations in the structural condition of the host structure; and prognosticating, based on the multiple sets of sensor signals and the variations in the structural condition, a structural condition of the host structure upon application of an external load to the host structure at a future point in time. 11. The method of claim 10, wherein the step of prognosticating a structural condition includes: training a neural network with the obtained multiple sets of sensor signals; and causing the trained neural network to generate a set of sensor signals at the future paint in time. 12. The method of claim 11, wherein the neural network is a recurrent neural network or a feedforward neural network. 13. The method of claim 12, further comprising: building a dynamic system model of the network, the dynamic system model having at least one system parameter. 14. The method of claim 13, wherein the dynamic system model is a state space model or an autoregressive moving average with exogenous input (ARMAX) model. 15. The method of claim 13, wherein the step of prognosticating a structural condition Includes: predicting a structural condition of the host structure of the network by use of the system parameter. 16. The method of claim 15, wherein the structural condition is affected by an impact on the host structure. 17. The method of claim 10, wherein the sensor signal includes at least one of ultrasonic wave, Lamb wave, vibrational wave, and acoustic wave. 18. A computer readable medium carrying one or more sequences of instructions for prognosticating a structural condition by use of a network having a plurality of diagnostic network patches (DNP), each of the patches being able to operate as at least one of an actuator patch and a receiver patch, at least one of the patches being able to operate as both an actuator patch and a receiver patch, wherein execution of one or more sequences of instructions by one or more processors causes the one or more processors to perform the steps of: (a) causing, by use of a computer processor, the at least one of the patches to emit a transmitter signal and a portion of the other patches to receive a set of sensor signals at a point in time, each said the sensor signal being a portion of the transmitter signal emitted by the at least one of the patches and propagated through a host structure; (b) repeating the step (a) to receive multiple sets of sensor signals at multiple points in time, respectively; and (c) prognosticating, based on the multiple sets of signals, a structural condition of a host structure of the network at a target point in time. 19. The computer readable medium of claim 18, wherein the step of prognosticating a structural condition includes: building a dynamic system model of the network, the dynamic system model having at least one system parameter; determining the system parameter for each of the multiple sets of sensor signals; estimating the system parameter at the target point in time based on the determined system parameters for the multiple sets of sensor signals; and predicting the structural condition by use of the estimated system parameter. 20. The computer readable medium of claim 18, wherein the one or more sequences of instructions implement a wireless communication method of Wireless Application Protocol (WAP) or Wireless Markup Language (WML) for the Internet Web Access of a WAP-enable cell phone, or HTML enable devices. 21. The computer readable medium of claim 18, wherein execution of one or more sequences of instructions by one or more processors causes the one or more processors to perform the additional step of: converting each of the multiple sets of sensor signals into an eXtensible Markup Language (XML) formatted document. 22. The computer readable medium of claim 18, wherein the transmitter signal includes at least one of ultrasonic wave, Lamb wave, vibrational wave, and acoustic wave. 23. A computer readable medium carrying one or more sequences of instructions for prognosticatlng a structural condition by use of a network having a plurality of sensors, wherein execution of one or more sequences of instructions by one or more processors causes the one or more processors to perform the steps of: preparing a network that includes a plurality of paths and nodes, each of the sensors corresponding to one of the nodes, each path connecting two of the nodes of the network to thereby form an edge of the network, at least one of the sensors being adapted to emit a transmitter signal that propagates through a host structure of the network, the sensors being adapted to receive the transmitter signal emitted by the at least one of the sensors; causing the sensors to obtain a set of sensor signals generated by an external load applied to the host structure at a point in time; analyzing the set of sensor signals by use of the network to characterize the external load and a variation in a structural condition of the host structure; repeating the steps of causing the sensors to obtain a set of sensor signals and analyzing the set of sensor signals to thereby obtain multiple sets of sensor signals from the sensors and variations in the structural condition of the host structure; and prognosticating, based on the multiple sets of sensor signals and the variations in the structural condition, a structural condition of the host structure upon application of an external load to the host structure at a future point in time. 24. The computer readable medium of claim 23, wherein the step of prognosticating a structural condition includes: training a neural network with the obtained multiple sets of sensor signals; and causing the trained neural network to generate a set of sensor signals at the future point in time. 25. The computer readable medium of claim 23, wherein the transmitter signal includes at least one of ultrasonic wave, Lamb wave, vibrational wave, and acoustic wave. 26. A processor for prognosticating a structural condition, the processor being adapted to be included in a system having a network to be coupled to a host structure and having a plurality of diagnostic network patches (DNP), each of the patches being able to operate as at least one of an actuator patch and a receiver patch, the processor being further adapted to: cause a first set of the patches to emit a first set of transmitter signals and a portion of the other patches to receive a first set of sensor signals at a first point in time, each of the first set of sensor signals being a portion of the first set of transmitter signals emitted by the first set of the patches and propagated through the host structure; cause a second set of the patches to emit a second set of transmitter signals and a portion of the other patches to receive a second set of sensor signals at a second point in time, each of the second set of sensor signals being a portion of the second set of transmitter signals emitted by the second set of the patches and propagated through the host structure; generate multiple sets of sensor signals including the first and second sets of sensor signals; and prognosticate, based on the multiple sets of sensor signals, a structural condition of the host structure at a target point in time. 27. The processor of claim 26, wherein the processor is further adapted to: build a dynamic system model of the network, the dynamic system model having at least one system parameter; determine the system parameter for each of the multiple sets of sensor signals; estimate the system parameter at the target point in time based on the system parameters determined for the multiple sets of sensor signals; and predict the structural condition by use of the system parameter estimated at the target point in time. 28. the processor of claim 26, wherein the transmitter signal includes at least one of ultrasonic wave, Lamb wave, vibrational wave, and acoustic wave.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.