基于SVDD的多时段间歇过程故障检测  

Multi-phase batch processes fault detection based on support vector data description

作　　者：王建林[1] 马琳钰 邱科鹏 刘伟旻 赵利强[1] 

机构地区：[1]北京化工大学信息科学与技术学院,北京100029

出　　处：《仪器仪表学报》2017年第11期2752-2761,共10页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(61240047);北京市自然科学基金(4152041)项目资助

摘　　要：支持向量数据描述(SVDD)不要求过程数据满足正态分布,已应用于间歇过程故障检测。现有的SVDD间歇过程故障检测方法采用聚类分析和模型识别划分间歇过程时段,多时段划分的准确性较低,制约了多时段间歇过程故障检测精度的提高。针对上述问题提出了一种基于SVDD的多时段间歇过程故障检测方法,利用SVDD的超球体半径值与支持向量个数的变化划分间歇过程的多时段,并建立了不同时段的SVDD故障检测模型,使用待检测样本点的球心距与对应时段的超球体半径之差检测过程故障,实现了多时段间歇过程的时段划分与过程故障检测。发酵过程仿真实验和实际生产实验结果表明,该方法能够准确地划分间歇过程的多时段,并且能够针对不同时段进行故障检测,具有较高的检测率。Support vector data description （SVDD） has been applied to fault detection of batch processes without any restriction of process data distribution. However, the SVDD based traditional fault detection method of batch processes just simply divides the phase of batch processes using cluster analysis and model identification, and leads to rough phase division results which may compromise the accuracy of fault detection in multi-phase batch processes. To address this issue, a SVDD based fault detection method for multi-phase batch processes is presented. Firstly, the different phases are divided according to the change of hypersphere radius and support vectors of SVDD. Then, the SVDD-based fault detection models of different phases are established. Finally, fault detection is achieved by judging whether the difference between hypersphere radius of corresponding phase and the distance from the sample point to center of corresponding hypersphere exceeds the control limit. The experimental results of the fed-batch penicillin fermentation process show that the presented method can achieve phase division of multi-phase batch processes with a better accuracy, and further realize multi-phase fault detection of batch processes with a higher fault detection rate.

关 键 词：间歇过程 支持向量数据描述 多时段 故障检测 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置] TL165.3[自动化与计算机技术—控制科学与工程]