摘要
为有效保障键合用封装设备可靠稳定运行,避免故障排查不及时、健康状态难以评估等问题,研究一种键合用封装设备健康状态评估方法。通过提取健康状态特征指标构建层次健康状态评价模型,采用深度学习模型和灰色聚类模型对不同类型指标数据进行处理,利用熵权法确定各指标所占权重,通过灰类加权自下而上对设备的健康状态和健康度进行综合评价,并以实际键合用封装设备为评估对象进行评估分析,分析结果验证了所提方法的合理性、有效性和准确性。
To effectively ensure the reliable and stable operation of bonding packaging equipment and avoid issues such as delayed fault detection and difficulty in assessing the health status,a health status assessment method is proposed.A hierarchical health status evaluation model is constructed by extracting health status feature indicators.Deep learning models and gray clustering models are used to process different types of data.The entropy weight method is used to determine the weight of each parameter.A gray class weighted bottom-up approach is then applied to comprehensively evaluate the health status and level of the equipment.The method is verified to be reasonable,effective and accurate based on the actual bonding packaging equipment.
作者
许红祥
杨帆
尤玉山
邝小乐
XU Hongxiang;YANG Fan;YOU Yushan;KUANG Xiaole(The Eighth Research Academy of CSSC,Nanjing 211153)
出处
《雷达与对抗》
2025年第2期50-57,共8页
Radar & ECM
关键词
键合用封装设备
健康状态
深度学习
灰色聚类
熵权法
bonding packaging equipment
health status
deep learning
gray clustering
entropy weight method
