高速动车组横梁应变场光纤测量与反演研究

Optical Fiber-Based Strain Field Measurement and Reconstruction for High-Speed EMU Beam

针对高速动车组车底底架横梁的应变场反演问题,基于模态叠加法和传感器优化配置理论,提出一种根据部分测点的实测单向应变反演结构空间应变场的方法。以结构的应变模态为输入,以模态坐标独立性为目标,基于信息熵理论建立了传感器优化配置的数学模型,再根据模态坐标和振型采用正则化方法反演结构应变场。仿真计算结果表明,无噪声干扰下,反演的精度较高,最大均方误差为9.87×10~(-12),并且正则化方法可以有效抑制噪声对结果的影响。搭建了基于光纤光栅传感系统的测试试验台,基于实测应变数据和模态结果反演应变场,结果表明,反演后的应变幅值相对误差带为1.21%~10.79%,反演数据和实测数据的均方误差带为0.80~3.69。该方法能够准确识别结构的模态坐标,为获取高速列车关键结构的应变场分布提供参考。

This paper addresses the issue of strain field reconstruction for the underframe beam of high-speed EMUs by proposing a method that reconstructs the structural spatial strain field according from the measured unidirectional strain at specific measuring points,utilizing the modal superposition method and sensor optimal configuration theory.With the strain mode of the structure as the input and the independence of modal coordinates as the optimization goal,a mathematical model for optimal sensor configuration is established based on information entropy theory,and then the structural strain field is reconstructed using a regularization method in accordance with modal coordinates and vibration modes.Noise-free simulation results indicate high reconstruction accuracy,with the maximum mean square error being 9.87×10-12.The regularization method can effectively suppress the impact of noise on the results.A test bench based on Fiber Bragg grating sensing system has been constructed,and the strain field is reconstructed based on the measured strain data and modal results.The results show that the relative error band for the reconstructed strain amplitude is 1.21%~10.79%,and the mean square error band for the reconstructed and the measured data is 0.80~3.69.The method proposed in this paper can accurately identify the modal coordinates of the structure,offering a novel approach for acquiring the strain field distribution of critical structure in high-speed train.