摘要
热轧板厂采用轧制力测量系统实时测量轧制压力和空载辊缝,以间接测量和控制带钢轧制厚度。轧制力传感器本身的结构特点和其工作的恶劣环境决定了它是整个控制系统中最容易发生故障的环节之一。为了避免传感器故障造成巨大经济损失,本文针对所研发的轧制力传感器进行了传感器故障分析、诊断及隔离方法的研究,提出了一种基于卡尔曼滤波和多重假设检验的故障诊断、隔离方法,仿真证明该方法可以及时、有效地检测到传感器故障并对输出数据进行重构。
Rolling force measuring system is used in hot rolling plant to indirectly measure and control the thickness of hot-rolled plate. The force sensor is one of the most fragile ceils in the whole control sys- tem because of its special structure and the bad environment. In the motive of avoiding huge economic losses caused by sensor trouble, the paper focuses on sensor fault detection and isolation（FDI） , and proposes a FDI algorithm method based on Kalman Filter and Multiple-failure-hypothesis. And simulations validate the effectiveness and instantaneity of sensor fault detection and configuration of output data of the proposed FDI algorithm.
出处
《冶金丛刊》
2012年第6期1-4,共4页
Metallurgical Collections
基金
广东省省部产学研结合引导项目2011B090400264
