Sparse Total Focus Imaging of Rail Transit Rail Defects Based on Grey Wolf Algorithm

Abstract

[Objective] Using the ultrasonic phased array technology can realize high-precision imaging detection of rail transit rail defects. However, the widely used FMC-TFM (full matrix capture-total focusing method) has the disadvantages of long computation time and low real-time performance. To reduce imaging calculation time, the sparse matrix is used to replace the full matrix for ultrasonic imaging. But the traditional intelligent optimization algorithm faces the trouble of slow convergence and is easy to fall into the local optimum when solving the sparse array design problem. Therefore, the sparse total focusing imaging method of rail defects based on the grey wolf algorithm is proposed to improve convergence performance and global search ability. [Method] The sparse array performance under optimization algorithm is analyzed. The ultrasonic phased array instrument is used to collect ultrasonic signals on the rail samples. Total focusing imaging is realized through the sparse matrix to analyze the imaging quality and time. [Result & Conclusion] The sparse array obtained by optimization algorithm has high sidelobe inhibition, and the PSL (peak sidelobe level) can reach -12.83 dB. At the PSL threshold of -6 dB, the width of the main lobe of the sparse array is comparable to that of the main lobe of 2.8° of the full array. When the sparsity rate is 75%, the quality of the rail imaging performance indicators is close to full array, and the imaging time is shortened by 56.35%.

Keywords

• urban rail transit
• rail defect
• sparse total focus imaging method
• grey wolf algorithm