摘要
为了实现数据驱动技术在工业中的实际应用,开发了以蒸馏塔作为被控对象的半实物仿真系统,将数据驱动方法应用到流程工业半实物仿真系统.针对动态主元分析方法存在的计算负荷大,计算效率低的问题,提出了一种改进动态主元分析方法,利用不可区分度和交叉程度去除众多变量中的不相关变量或相关度较小的变量,减少数据量.针对系统中的典型故障,数据驱动方法能够检测出半实物仿真系统中的异常,而且与传统动态主元分析比较,改进算法降低漏报率和误报率,提高诊断可靠性,并且能及时检测出生产过程的微小故障.
In order to apply the data-driven technology in industry field,with the distillation column as one of the controlled objects,the hardware-in-the-loop simulation system is developed,and the data driven methods are applied to the process industry hardware-in-the-loop simulation system. Aiming at the huge computation and lowefficiency of the dynamic principal component analysis method,an improved dynamic principal component analysis method is proposed,which remove the irrelevant variables or lowrelevant variables,reduce the amount of data,and improve the diagnostic efficiency by indiscernibility and the cross-degree. For the typical faults of the systems,the application result shows that the data-driven methods can detect the fault of hardware-in-the-loop simulation system,and compared with the traditional dynamic principal component,the simulation results showthat the proposed method is more reliable with lower missing rate,and lower false rate,and in addition,it can detect the small process faults timely.
出处
《电子学报》
EI
CAS
CSCD
北大核心
2017年第3期565-569,共5页
Acta Electronica Sinica
基金
国家自然科学基金(No.61403279)
天津市中青年骨干创新人才培养计划基金(No.20130830)
关键词
数据驱动
蒸馏塔
半实物仿真
动态主元分析
不可区分度
data driven
distillation column
hardware-in-the-loop simulation
dynamic principal component analysis（DPCA）
indiscernibility
