基于Harris-SUSAN算法的发动机叶片裂纹检测系统设计

Design of Engine Blade Crack Detection System Based on Harris-SUSAN Algorithm

摘要:

目前设计的发动机叶片裂纹检测系统存在裂纹感应电压能力差的问题，导致缺陷信号峰值检测结果误差较大，为此提出基于Harris-SUSAN算法设计一种新的发动机叶片裂纹检测系统。利用传感器中高灵敏度的二轴霍尼韦尔各向异性磁阻(AMR)元件，分析水平与垂直方向范围内的磁场分布与变化情况，设计传感器，选择旋转电磁激励台实现激励台的旋转磁化。引用Harris-SUSAN算法设定检测程序，计算兴趣值确定最优点，比较磁场强度，检测发动机叶片裂纹。实验结果表明，基于Harris-SUSAN算法的发动机叶片裂纹检测系统能够有效提高裂纹感应电压能力，增强缺陷信号峰值检测结果检测准确率。

Abstract:

At present, there is a problem that the ability of crack induced voltage is poor in the design of engine blade crack detection system, which leads to large error of defect signal peak detection results. Therefore, a new engine blade crack detection system based on Harris SUSAN algorithm is proposed. In this paper, the high sensitive two-axis Honeywell anisotropic magnetoresistance (AMR) element is used to analyze the distribution and variation of magnetic field in the horizontal and vertical directions. The sensor is designed and the rotating electromagnetic excitation platform is selected to realize the rotating magnetization of the excitation platform. The Harris SUSAN algorithm is used to set up the detection program, calculate the value of interest to determine the best advantage, and compare the magnetic field strength to detect the engine blade crack. The experimental results show that the crack detection system based on Harris SUSAN algorithm can effectively improve the ability of crack induced voltage and enhance the detection accuracy of defect signal peak detection results.