论文部分内容阅读
近年来,过程工业安全事故频发,这使得加强生产过程安全保障变得迫在眉睫,而对于过程故障的监测、诊断是有效规避故障产生严重后果的一个有效方法。本文提出了一种基于核主成分分析(KPCA)的半定量符号有向图(SDG)故障诊断方法。此方法运用KPCA对过程进行异常检测,当找到异常过程变量后,通过引入相对偏移率和分类诊断对传统SDG进行改进,从而得到故障的完整传播路径,为故障诊断以及后续的故障处理提供了有效的指导。通过在TE过程中的仿真验证,结果表明,本方法诊断效率高,精确度高,为保证生产安全运行,提高产品质量提供了新途径。
In recent years, the frequent occurrence of industrial safety accidents in the process has made it extremely urgent to enhance the safety of the production process. However, the monitoring and diagnosis of process failures is an effective way to effectively avoid the serious consequences of the failure. In this paper, we present a semi-quantitative digraph (SDG) fault diagnosis method based on kernel principal component analysis (KPCA). This method uses KPCA to detect the anomaly of the process. After finding the abnormal process variables, the traditional SDG is improved by introducing the relative rate of deviation and the classification of diagnoses, so as to get the complete propagation path of the fault, which provides the fault diagnosis and subsequent fault handling Effective guidance. Through the simulation in TE process, the results show that the method has high diagnostic efficiency and high accuracy, which provides a new way to ensure the safe operation of production and improve the product quality.