机构地区: 华中理工大学
出 处: 《华中理工大学学报》 1993年第1期57-62,共6页
摘 要: 根据设备监测与诊断系统诊断对象的复杂特点,提出了一种基于发生率的方法对传感器信号可靠性进行确认.通过在线采集一定时间内的数据,在规定的假设空间内,从证据中推导证据.对传感器信号的可靠性确认通过失效和设计检测处理,在事件发生率计算基础上,给出了规定假设空间内传感器信号可靠性的确认概率分配计算方法. The affirmation of the signal reliability in a knowledge-based machine monitoring and fault diagnosing system is of great importance. Since uncertainties in a knowledge base exist in most practical cases, how to deal with uncertainties including spurious sensor signals to guarantee accurate diagnosis is a crucial issue. A method based on a statistical apprach for sensor signal reliability on a machine monitoring and fault diagnosing ststem is proposed. With the on-line data collected within a given time and in a specified hypothetic space,evi- dence is derived from the evidence availale. Based on the rate of occurrence calculation, a method of calculating the updated probability distribution of the affirmation of signal reliability over specified hypothetic space is presented.
领 域: [机械工程]