摘要
电子装备状态监测技术是装备健康管理的关键技术之一,为了对电子装备的健康状态进行有效监测,首先对模型性能评价指标进行了分析;然后对传统SVDD模型进行了研究,针对该模型只对目标类样本建模而导致分类准确率较低的问题,提出了一种基于最大分类间隔的SVDD监测模型;该模型在保证最小化包裹目标类样本数据超球体的同时,使得目标类样本和非目标类样本之间的类间间隔最大,提高了模型的泛化能力;最后以某型装备滤波电路为例进行了仿真分析,分析结果表明,该模型无论是在精度、召回率还是F值上均要优于传统SVDD模型。
Electronic equipment condition monitoring is the key technology of equipment health management, In order to monitor the health status of electronic equipment, the paper analyzing the model evaluating indicator firstly, then studying the normal SVDD model, ai ming at the problem of this model modeling with only target samples while leading to low classification accuracy, bringing the new SVDD model based on maximal classification margin. This model ensure the hypersphere which include the target samples minimum meanwhile mak- ing the classification margin between target samples and nontarget samples maximum, so as to improve the model generalization ability. Fi- nally, taking a filter circuit of an equipment as an example to simulate, the results show that the performance of this model is better than nor- mal SVDD.
出处
《计算机测量与控制》
CSCD
北大核心
2012年第9期2335-2337,共3页
Computer Measurement &Control
