复合材料结构冲击的轻量化两级成像定位方法

A Lightweight Two-Stage Imaging Localization Method for Impact on Composite Structures

  • 摘要: 复合材料在航空航天领域被广泛应用,但是复合材料对冲击的抵抗性弱,对复合材料结构冲击开展定位监测在保障飞行器的服役安全、控制维护成本具有重要意义。针对传统时间反转聚焦成像方法在开展大面积结构冲击定位时使用的传感器数量多导致传感网络的附加重量大、冲击成像计算量大的问题,本文提出了一种复合材料结构冲击的轻量化两级成像定位方法,该方法通过Shannon小波变换提取2个不同频率窄带冲击信号的飞行时间差特征开展冲击半径一级成像,一级成像结果中高像素值的位置即冲击可能发生的位置,在此基础上对高像素值位置开展时间反转聚焦二级成像,即可实现准确的冲击成像定位,同时能大幅降低成像计算量。在900 mm×900 mm尺寸的复合材料板布置了5个压电传感器构成稀疏阵列,对该方法进行了实验验证。结果表明,20个冲击位置的平均定位误差仅1.4 cm,平均成像计算时间与传统时间反转聚焦成像方法相比缩短了96.2%。该方法在大面积复合材料结构实现了准确的冲击定位,同时大幅降低了成像计算量。

     

    Abstract: Composite materials are widely used in the aerospace field; however, their relatively low resistance to impact makes impact localization and monitoring of composite structures critically important for ensuring aircraft operational safety and controlling maintenance costs. To address the limitations of conventional time-reversal focusing imaging methods—namely, the large number of sensors required for large-area impact localization, which leads to excessive additional weight of the sensing network and high computational burden—a lightweight two-stage imaging localization method for impact on composite structures was proposed. In the proposed approach, the time-of-flight difference features of two narrowband impact signals at different frequencies are extracted using the Shannon wavelet transform to perform first-stage imaging for impact radius estimation. Locations with high pixel values in the first-stage imaging results are regarded as potential impact regions. Based on these regions, time-reversal focusing–based second-stage imaging is subsequently conducted to achieve accurate impact localization, while significantly reducing the imaging computational cost. Experimental validation was carried out on a 900 mm × 900 mm composite plate instrumented with five piezoelectric sensors arranged as a sparse array. The results indicate that, for 20 impact locations, the proposed method achieves an average localization error of only 1.4 cm, and the average imaging computation time is reduced by 96.2% compared with the conventional time-reversal focusing imaging method. The proposed method enables accurate impact localization in large-area composite structures with substantially reduced computational cost.

     

/

返回文章
返回