基于观测图像识别的天文望远镜智能故障辅助诊断系统  

The Intelligent Fault Auxiliary Diagnosis System of Astronomical Telescope Based on Observation Image Recognition

作　　者：张豫龙 王建峰[1] 李陶然[1] 葛亮[1] 兀颖 赵勇[1,2] 姜晓军 Zhang Yulong;Wang Jianfeng;Li Taoran;Ge Liang;Wu Ying;Zhao Yong;Jiang Xiaojun(Key Laboratory of Optical Astronomy,National Astronomical Observatories,Chinese Academy of Sciences,Beijing 100101,China;School of Astronomy and Space Science,University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院光学天文重点实验室(国家天文台),北京100101 [2]中国科学院大学天文与空间科学学院,北京100049

出　　处：《天文研究与技术》2020年第3期392-398,共7页Astronomical Research & Technology

基　　金：国家自然科学基金(U1831209)资助。

摘　　要：设计并构建了基于观测图像识别的望远镜智能故障辅助诊断系统,通过收集望远镜在观测时的故障图像与传感器信息来综合判断望远镜在观测时产生的故障,并给出相应的解决建议。利用卷积神经网络方法进行图像智能识别,并使用故障树分析法寻找望远镜系统的薄弱点。经过仿真与实测检验,证明了系统能够有效地发现判断故障并给出建议,提高了望远镜的可靠性,为故障诊断技术智能化做好了铺垫。We designed and built an intelligent auxiliary system for fault diagnosis of telescope based on observation image recognition.The system collects the measured image and sensor information of the telescope to comprehensively judge the faults during observation,and gives corresponding solutions.We use the convolution neural network method to intelligently identify images and use fault tree analysis to find the weak points of the telescope system.The simulation and actual tests prove that the system can effectively find,judge faults and give suggestions.It improves the reliability of the telescope and accumulates experience for the intelligentization of fault diagnosis technology.

关 键 词：故障诊断 故障树 图像识别 天文望远镜 

分 类 号：TH751[机械工程—仪器科学与技术]