Combined bias and outlier identification in dynamic data reconciliation

Measured process data normally contain inaccuracies because the measurements are obtained using imperfect instruments. As well as random errors one can expect systematic bias caused by miscalibrated instruments or outliers caused by process peaks such as sudden power fluctuations. Data reconciliation is the adjustment of a set of process data based on a model of the process so that the derived estimates conform to natural laws. In this paper, techniques for the detection and identification of both systematic bias and outliers in dynamic process data are presented. A novel technique for the detection and identification of systematic bias is formulated and presented. The problem of detection, identification and elimination of outliers is also treated using a modified version of a previously available clustering technique. These techniques are also combined to provide a global dynamic data reconciliation (DDR) strategy. The algorithms presented are tested in isolation and in combination using dynamic simulations of two continuous stirred tank reactors (CSTR).

[1]  Michèle Basseville,et al.  Detecting changes in signals and systems - A survey , 1988, Autom..

[2]  J. Romagnoli,et al.  Simultaneous estimation of biases and leaks in process plants , 1999 .

[3]  Chen Shan Kao,et al.  Gross Error Detection in Serially Correlated Process Data , 1990 .

[4]  Miguel J. Bagajewicz,et al.  Gross error modeling and detection in plant linear dynamic reconciliation , 1998 .

[5]  Manfred Morari,et al.  Optimal operation of integrated processing systems. Part I: Open‐loop on‐line optimizing control , 1981 .

[6]  L. Biegler,et al.  Data reconciliation and gross‐error detection for dynamic systems , 1996 .

[7]  George Stephanopoulos,et al.  Rectification of process measurement data in the presence of gross errors , 1981 .

[8]  Mohamed Darouach,et al.  Data reconciliation in generalized linear dynamic systems , 1991 .

[9]  Thomas F. Edgar,et al.  Bias detection and estimation in dynamic data reconciliation , 1994 .

[10]  Richard S.H. Mah,et al.  Generalized likelihood ratios for gross error identification in dynamic processes , 1988 .

[11]  Ziad Hasan Abu-el-zeet,et al.  Optimisation techniques for advanced process supervision and control , 2000 .

[12]  Chen Shan Kao,et al.  Gross Error Detection in Serially Correlated Process Data. 2. Dynamic Systems , 1992 .

[13]  M. A. Girshick,et al.  A BAYES APPROACH TO A QUALITY CONTROL MODEL , 1952 .

[14]  Victor M. Becerra,et al.  Dynamic data reconciliation for sequential modular simulators , 1998 .

[15]  Miguel J. Bagajewicz,et al.  Integral approach to plant linear dynamic reconciliation , 1997 .

[16]  Ling-Hwei Chen,et al.  A new non-iterative approach for clustering , 1994, Pattern Recognit. Lett..

[17]  Michèle Basseville,et al.  Detection of abrupt changes: theory and application , 1993 .

[18]  C. M. Crowe,et al.  Data reconciliation — Progress and challenges , 1996 .

[19]  L. Biegler,et al.  Decomposition algorithms for on-line estimation with nonlinear DAE models , 1995 .

[20]  Rolf Isermann,et al.  Trends in the Application of Model Based Fault Detection and Diagnosis of Technical Processes , 1996 .

[21]  Richard S. H. Mah,et al.  Reconcillation and Rectification of Process Flow and Inventory Data , 1976 .

[22]  Victor M. Becerra,et al.  Enhancing model predictive control using dynamic data reconciliation , 2002 .

[23]  James F. Davis,et al.  Gross error detection when variance-covariance matrices are unknown , 1993 .

[24]  Richard S. H. Mah,et al.  Generalized likelihood ratio method for gross error identification , 1987 .

[25]  Alan S. Willsky,et al.  A survey of design methods for failure detection in dynamic systems , 1976, Autom..

[26]  R. Serth,et al.  Gross error detection and data reconciliation in steam-metering systems , 1986 .

[27]  P. Frank,et al.  Survey of robust residual generation and evaluation methods in observer-based fault detection systems , 1997 .

[28]  M. Pollak.  Average Run Lengths of an Optimal Method of Detecting a Change in Distribution. , 1987 .

[29]  L. Lasdon,et al.  Efficient data reconciliation and estimation for dynamic processes using nonlinear programming techniques , 1992 .

[30]  Tony Perris.  Process plant performance: measurement and data processing for optimization and retrofits : by Frantisek Madron (Ellis Horwood, 1992, ISBN 0 13 723875 4, 300 pp, £46.00) , 1994 .

[31]  Ajit C. Tamhane,et al.  Detection of gross errors in process data , 1982 .

[32]  James F. Davis,et al.  Unbiased estimation of gross errors in process measurements , 1992 .

[33]  Jose A. Romagnoli,et al.  A strategy for simultaneous dynamic data reconciliation and outlier detection , 1998 .

[34]  C. Edward Bodington,et al.  Planning, Scheduling, and Control Integration in the Process Industries , 1995 .

[35]  Mohamed Darouach,et al.  Fault detection of multiple biases or process leaks in linear steady state systems , 1994 .