基于模式识别的大功率医用激光控制系统故障研究  

Research on fault of high power medical laser control system based on pattern recognition

作　　者：李波 LI Bo(Children’s Hospital of Nanjing Medical University,Nanjing 210000,China)

机构地区：[1]南京医科大学附属儿童医院,南京210000

出　　处：《激光杂志》2022年第12期185-189,共5页Laser Journal

基　　金：江苏科技厅面上研究项目(No.JS20181229933)。

摘　　要：故障研究是保证大功率医用激光控制系统正常运行的重要技术,为提升大功率医用激光控制系统故障诊断的准确性,提出了基于模式识别的大功率医用激光控制系统故障诊断方法。采集大功率医用激光控制系统状态信号,提取特征,引入模式识别技术设计大功率医用激光控制系统故障诊断的分类器,并进行了大功率医用激光控制系统故障诊断的仿真实验。结果表明,本方法的大功率医用激光控制系统故障诊断正确率超过95%,误诊率控制在5%以内,获得了理想的大功率医用激光控制系统故障诊断结果,可以满足控制系统维护实际要求。Fault research is the most important technology to ensure the normal operation of high-power medical laser control system. In order to improve the fault diagnosis accuracy of high-power medical laser control system, a fault diagnosis method of high-power medical laser control system based on pattern recognition is proposed. The state signals of high-power medical laser control system are collected, the characteristics are extracted, the pattern recognition technology is introduced, the classifier of fault diagnosis of high-power medical laser control system is designed, and the simulation experiment of fault diagnosis of high-power medical laser control system is carried out. The results show that the fault diagnosis accuracy of high-power medical laser control system is more than 95%, and the misdiagnosis rate is controlled within 5%. The ideal fault diagnosis results of high-power medical laser control system can meet the actual requirements of control system maintenance.

关 键 词：模式识别技术 激光控制系统 故障特征 故障误诊率 

分 类 号：TN929[电子电信—通信与信息系统]