• Title, Summary, Keyword: Multiway independent component analysis (MICA)

Search Result 1, Processing Time 0.026 seconds

FAULT DETECTION, MONITORING AND DIAGNOSIS OF SEQUENCING BATCH REACTOR FOR INTEGRATED WASTEWATER TREATMENT MANAGEMENT SYSTEM

  • Yoo, Chang-Kyoo;Vanrolleghem, Peter A.;Lee, In-Beum
    • Environmental Engineering Research
    • /
    • v.11 no.2
    • /
    • pp.63-76
    • /
    • 2006
  • Multivariate analysis and batch monitoring on a pilot-scale sequencing batch reactor (SBR) are described for integrated wastewater treatment management system, where a batchwise multiway independent component analysis method (MICA) are used to extract meaningful hidden information from non-Gaussian wastewater treatment data. Three-way batch data of SBR are unfolded batch-wisely, and then a non-Gaussian multivariate monitoring method is used to capture the non-Gaussian characteristics of normal batches in biological wastewater treatment plant. It is successfully applied to an 80L SBR for biological wastewater treatment, which is characterized by a variety of error sources with non-Gaussian characteristics. The batchwise multivariate monitoring results of a pilot-scale SBR for integrated wastewater treatment management system showed more powerful monitoring performance on a WWTP application than the conventional method since it can extract non-Gaussian source signals which are independent and cross-correlation of variables.