【摘要】 针对30 mm口径火炮身管的膛线检测，提出一种基于漏磁检测的膛线磁成像技术。对膛线的漏磁信号进行了仿真，建立磁传感器旋转扫描膛内表面采集漏磁信号模型，通过二维插值的方法确定所有采样点的磁场强度。针对磁传感器信号采集过程中的旋转轴误差，分析了偏角、偏心对膛线磁成像的影响，提出一种基于LMS的误差补偿算法。仿真结果表明，该补偿算法分别将存在1°偏角误差、0.8 mm偏心误差、同时存在0.5°偏角误差和0.5 mm偏心误差三种情况的成像图与理想图像之间的SSIM值从0.25、0.29、0.34显著提升至0.95、0.86、0.88,有效补偿了因旋转轴误差造成的磁成像偏差，为后续膛线缺陷检测、缠角检测提供了一种预处理方法，对提高膛线检测精度具有重要意义。更多还原

【Abstract】 For the detection of rifling in a 30 mm caliber artillery barrel, a magnetic imaging technique based on leakage magnetic detection has been proposed. The leakage magnetic signals from the rifling were simulated, and a model was established for the magnetic sensor to rotate and scan the inner surface of the barrel to collect leakage magnetic signals. A two-dimensional interpolation method was used to determine the magnetic field strength at all sampling points. In addressing the rotational axis errors during the magnetic sensor signal acquisition process, the impacts of angular deviation and eccentricity on magnetic imaging of the rifling were analyzed, and a compensation algorithm based on LMS(Least Mean Squares) was proposed. The simulation results indicate that the compensation algorithm significantly improved the SSIM values between the imaging results with 1° angular deviation, 0.8mm eccentricity, and simultaneously existing 0.5° angular deviation and 0.5mm eccentricity cases and the ideal image, increasing them from 0.51, 0.38, and 0.41 to 0.91, 0.85, and 0.85, respectively. This effectively compensated for the magnetic imaging deviations caused by rotational axis errors and provides a preprocessing method for subsequent defect detection and twist angle measurement of the rifling, which is of great significance for improving the accuracy of rifling detection.更多还原