SCI和EI收录∣中国化工学会会刊

›› 2008, Vol. 16 ›› Issue (4): 584-589.

• SYSTEM ENGINEERING • Previous Articles     Next Articles

Quality Based Prioritized Sensor Fault Monitoring Methodology

SONG Kai1, WANG Haiqing2, LI Ping2, FENG Zhigang3   

  1. 1. School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China;
    2. Institute of Industrial Process Control, Zhejiang University, Hangzhou 310027, China;
    3. Hebei Zhongrun Pharmaceutical Co., Ltd., Shijiazhuang 050041, China
  • Received:2007-07-28 Revised:2008-03-25 Online:2008-08-28 Published:2008-08-28
  • Supported by:
    the National Natural Science Foundation of China(20776128);the Natural Science Foundation of Zhejiang Province(Y107032)

Quality Based Prioritized Sensor Fault Monitoring Methodology

宋凯1, 王海清2, 李平2, 冯志刚3   

  1. 1. School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China;
    2. Institute of Industrial Process Control, Zhejiang University, Hangzhou 310027, China;
    3. Hebei Zhongrun Pharmaceutical Co., Ltd., Shijiazhuang 050041, China
  • 通讯作者: WANG Haiqing, E-mail: hqwang@iipc.zju.edu.cn
  • 基金资助:
    the National Natural Science Foundation of China(20776128);the Natural Science Foundation of Zhejiang Province(Y107032)

Abstract: To improve the detection and identification performance of the Statistical Quality Monitoring (SQM) system, a novel quality based Prioritized Sensor-Fault Detection (PSFD) methodology is proposed. Weighted by the Vp (variable importance in projection) index, which indicates the importance of the sensor variables to the quality variables, the new monitoring statistic, Qv, is developed to ensure that the most vital sensor faults be detected successfully. Subsequently, the ratio between the Detectable Minimum Faulty Magnitude (DMFM) of the most important sensor and of the least important sensor is only Vpmin/Vpmax<<1. The Structured Residuals are designed according to the Vp index to identify and then isolate them. The theoretical findings are fully supported by simulation studies performed on the Tennessee Eastman process.

Key words: partial least squares, statistical quality monitoring, Tennessee Eastman process

摘要: To improve the detection and identification performance of the Statistical Quality Monitoring (SQM) system, a novel quality based Prioritized Sensor-Fault Detection (PSFD) methodology is proposed. Weighted by the Vp (variable importance in projection) index, which indicates the importance of the sensor variables to the quality variables, the new monitoring statistic, Qv, is developed to ensure that the most vital sensor faults be detected successfully. Subsequently, the ratio between the Detectable Minimum Faulty Magnitude (DMFM) of the most important sensor and of the least important sensor is only Vpmin/Vpmax<<1. The Structured Residuals are designed according to the Vp index to identify and then isolate them. The theoretical findings are fully supported by simulation studies performed on the Tennessee Eastman process.

关键词: partial least squares, statistical quality monitoring, Tennessee Eastman process