结构振动测试传感器优化布置方法
详细信息 查看官网全文
- 英文论文题名：An optimal approach on sensor placement for structural vibration test
- 论文作者：于海洋
- 英文论文作者：YU Hai-yang ; The Naval Equipment Department
- 年：2016
- 作者机构：海军装备部;
- 论文关键词：振动 ; 测试 ; 传感器布置 ; 条件数
- 英文论文关键词：Vibration ; Test ; Sensor placement ; Condition number
- 会议召开时间：2016-09-17
- 会议录名称：2016年度声学技术学术会议论文集
- 语种：中文
- 分类号：U661.44;TP212
- 学会代码：SXJS
- 会议名称：2016年度声学技术学术会议
- 会议地点：中国湖北武汉
- 主办单位：船舶振动噪声重点实验室、湖北省声学学会
- 学会名称：《声学技术》编辑部
- 页数：6
- 文件大小：632k
- 原文格式：D
- 会议级别：全国
摘要
大型船体结构振动测试的关键问题是传感器数量很多,过多的传感器花费太高但减少传感器数量又可能导致误差明显增大。本文基于结构模态矩阵条件数优选传感器布置位置和数量,并重构整个结构振动,在迭代过程的每个循环中消去对模态矩阵条件数贡献最大的传感器位置。文中分析了每次迭代消除位置数量的影响并做了仿真验证。实验验证了本文方法的准确性,某实验结构26个优化的传感器布置与200个初始传感器布置对比其重构误差仅为2.4d B,另一实验结构27个优化的传感器布置与264个初始传感器布置对比其重构误差仅为3.0d B。本方法或可用于大型结构物的振动测量与实验。
The critical issue of vibration test on hull structures with large size is the numerous amounts of sensors. Too many sensors lead to excessive consumption whereas notable error would come from reducing the number of sensors. An approach is proposed to reconstruct the complete structural vibration, which choose the locations and number of sensors optimally based on the condition number of structural modal matrix. This iterative progress eliminates the sensor location that contributes the most significantly to the condition number of modal matrix in each cycle. The influence of the location number eliminated in each cycle is analysed and emulated in this paper. An experiment validates the precise of this approach. Vibration responses reconstructed from the data of 26 optimal sensors are consistent with the responses measured by 200 initial sensors, the vibration error is only 2.4 d B for one test structure. For another structure, the error is 3.0 d B between the reconstructed data from 27 optimal sensors and the responses measured by 264 initial sensors. This approach may be applied to vibration test and measurement on large structures.
The critical issue of vibration test on hull structures with large size is the numerous amounts of sensors. Too many sensors lead to excessive consumption whereas notable error would come from reducing the number of sensors. An approach is proposed to reconstruct the complete structural vibration, which choose the locations and number of sensors optimally based on the condition number of structural modal matrix. This iterative progress eliminates the sensor location that contributes the most significantly to the condition number of modal matrix in each cycle. The influence of the location number eliminated in each cycle is analysed and emulated in this paper. An experiment validates the precise of this approach. Vibration responses reconstructed from the data of 26 optimal sensors are consistent with the responses measured by 200 initial sensors, the vibration error is only 2.4 d B for one test structure. For another structure, the error is 3.0 d B between the reconstructed data from 27 optimal sensors and the responses measured by 264 initial sensors. This approach may be applied to vibration test and measurement on large structures.
引文
[1]Kammer DC.Sensor placement for on-orbit modal identification and correlation of large space structures[J].Journal of Guidance,Control,and Dynamics.1991,14(2):251-259.
[2]Heo G,Wang ML,Satpathi D.Optimal transducer placement for health monitoring of long span bridge[J].Soil Dynamics and Earthquake Engineering.1997,16(7-8):495-502.
[3]Meo M,Zumpano G.On the optimal sensor placement techniques for a bridge structure[J].Engineering Structures.2005,27(10):1488-1497.
[4]Allemang RJ,Brown DL.A correlation coefficient for modal vector analysis[A].Proceedings of the 1st International Modal Analysis Conference[C];Orlando,Florida 1982.p.110-116.
[5]Dochain D,Tali-Maamar N,Babary JP.On modelling,monitoring and control of fixed bed bioreactors[J].Computers Chem.Engng.1997,21(11):1222-1266.
[6]Li YY,Yam LH.Sensitivity analyses of sensor locations for vibration control and damage detection of thin-plate systems[J].Journal of Sound and Vibration.2001,240(4):623-636.
[7]Choi HG,Thite AN,Thompson DJ.Methods for selecting sensor locations for improving indirect force determination[A].Proceedings of the Institute of Acoustics[C].2004,26(2):265-276.
[2]Heo G,Wang ML,Satpathi D.Optimal transducer placement for health monitoring of long span bridge[J].Soil Dynamics and Earthquake Engineering.1997,16(7-8):495-502.
[3]Meo M,Zumpano G.On the optimal sensor placement techniques for a bridge structure[J].Engineering Structures.2005,27(10):1488-1497.
[4]Allemang RJ,Brown DL.A correlation coefficient for modal vector analysis[A].Proceedings of the 1st International Modal Analysis Conference[C];Orlando,Florida 1982.p.110-116.
[5]Dochain D,Tali-Maamar N,Babary JP.On modelling,monitoring and control of fixed bed bioreactors[J].Computers Chem.Engng.1997,21(11):1222-1266.
[6]Li YY,Yam LH.Sensitivity analyses of sensor locations for vibration control and damage detection of thin-plate systems[J].Journal of Sound and Vibration.2001,240(4):623-636.
[7]Choi HG,Thite AN,Thompson DJ.Methods for selecting sensor locations for improving indirect force determination[A].Proceedings of the Institute of Acoustics[C].2004,26(2):265-276.