Research on Damages Evaluation Method With Multi-Objective Feature Extraction Optimization Scheme for M/OD Impact Risk Assessment

As the number of space debris (also called meteoroid/orbital debris-M/OD) increases in recent years, the hypervelocity-impact (HVI) events of M/OD on spacecrafts have become one of the most main risks threatening human activity in space. For the automatical M/OD risk assessment, some effective nondestructive testing (NDT) methods are critical to realizing the evaluation of the HVI damages. In this paper, a novel HVI damage evaluation method based on the active infrared thermal wave image detection technology with multi-objective feature extraction optimization (MO-FEO) is proposed to achieve the quantitative evaluation of M/OD HVI damages. For the precise selection of representative temperature point in thermal infrared image data, the proposed MO-FEO method has the advantage not only of considering the difference among temperature points in different thermal temperature categories but also considering the correlation among temperature points of each thermal temperature category. The multi-objective feature extraction problem decomposed by Tchebycheff aggregation is used to seek the representative temperature points through an evolution process brought the selection pressure and fitness value. In addition to the MO-FEO frame, the variable step search and classification of temperature points are also implemented in the HVI damage evaluation strategy to improve efficiency. Some experimental results of infrared detection for the real M/OD HVI test articles are proposed to illustrate the effectiveness of the proposed method.