Generic Monitoring System in the Biological Wastewater Treatment Process

In this paper a new monitoring algorithm utilizing a change in time-series distribution of process data is presented since the distribution reflects the corresponding operating condition. In order to quantitatively evaluate the difference between two data sets, a modified dissimilarity index is defined. It represents the degree of dissimilarity between data-sets. In training step the confidence interval of each eigenvalue is obtained from the data taken in normal operation. Then, current operating condition is monitored by checking whether dissimilarity index abruptly changes and whether each eigenvalue is contained within its confidence interval. This approach is used to identify various internal and external disturbances in the data from the simulated activated sludge process. Simulation results have clearly shown that the detection performance of the proposed method can detect the various faults and disturbances, and can automatically discriminate between serious and minor anomalies of faults. That is, it can detect not only the disturbances, but isolate the sources of them. These results confirm that the proposed method is a proper monitoring technique for the wastewater treatment process which has nonstationary property and various disturbances.

[1]  John F. MacGregor,et al.  Process monitoring and diagnosis by multiblock PLS methods , 1994 .

[2]  Richard D. Braatz,et al.  Fault detection in industrial processes using canonical variate analysis and dynamic principal component analysis , 2000 .

[3]  S. Qin.  Recursive PLS algorithms for adaptive data modeling , 1998 .

[4]  Bhupinder S. Dayal,et al.  Recursive exponentially weighted PLS and its applications to adaptive control and prediction , 1997 .

[5]  Theodora Kourti,et al.  Process analysis, monitoring and diagnosis, using multivariate projection methods , 1995 .

[6]  Gang Chen,et al.  Predictive on-line monitoring of continuous processes , 1998 .

[7]  Manabu Kano,et al.  Comparison of statistical process monitoring methods: application to the Eastman challenge problem , 2000 .

[8]  Manabu Kano,et al.  Dissimilarity of Process Data for Statistical Process Monitoring , 2000 .

[9]  Manabu Kano,et al.  Comparison of multivariate statistical process monitoring methods with applications to the Eastman challenge problem , 2002 .

[10]  Ali Cinar,et al.  Multivariate statistical methods for monitoring continuous processes: assessment of discrimination power of disturbance models and diagnosis of multiple disturbances , 1995 .

[11]  Leo H. Chiang,et al.  Fault diagnosis in chemical processes using Fisher discriminant analysis, discriminant partial least squares, and principal component analysis , 2000 .

[12]  I. Takács.  A dynamic model of the clarification-thickening process , 1991 .

[13]  B. Kowalski,et al.  Partial least-squares regression: a tutorial , 1986 .

[14]  J. Macgregor,et al.  Monitoring batch processes using multiway principal component analysis , 1994 .

[15]  G. Van Dongen,et al.  Multivariate time series analysis for design and operation of a biological wastewater treatment plant , 1998 .

[16]  Barry M. Wise,et al.  The process chemometrics approach to process monitoring and fault detection , 1995 .

[17]  W. Gujer,et al.  A general model for single-sludge wastewater treatment systems , 1987 .

[18]  Christos Georgakis,et al.  Disturbance detection and isolation by dynamic principal component analysis , 1995 .

[19]  Milos Krofta,et al.  An attempt to understand dissolved air flotation using multivariate data analysis , 1995 .

[20]  B. Bakshi.  Multiscale PCA with application to multivariate statistical process monitoring , 1998 .

[21]  Weihua Li,et al.  Recursive PCA for adaptive process monitoring , 1999 .