Applying Kohonen self-organizing map as a software sensor to predict biochemical oxygen demand

    Research output: Contribution to journalArticle

    28 Citations (Scopus)

    Abstract

    The 5 days at 20 degrees C biochemical oxygen demand (BOD5) is an important parameter for monitoring organic pollution in water and assessing the biotreatability of wastewater. Moreover, BOD5 is used for wastewater treatment plant discharge consents and other water pollution control purposes. However, the traditional bioassay method for estimating the BOD5 involves the incubation of sample water for 5 days. It follows that BOD5 is not available for real-time decisionmaking and process control purposes. On the other hand, previous efforts to solve this problem by developing more rapid biosensors had limited success. This paper reports on the development of Kohonen self-organizing map (KSOM)-based software sensors for the rapid prediction of BOD5. The findings indicate that the KSOM-based BOD5 estimates were in good agreement with those measured using the conventional bioassay method. This offers significant potential for more timely intervention and cost savings during problem diagnosis in water and wastewater treatment processes.

    Original languageEnglish
    Pages (from-to)32-40
    Number of pages9
    JournalWater environment research
    Volume80
    Issue number1
    DOIs
    Publication statusPublished - Jan 2008

    Keywords

    • Bioassay
    • Biochemical oxygen demand
    • Biosensor
    • Cost saving
    • Decisionmaking
    • Kohonen self-organizing map
    • Neural networks
    • Process control
    • Software sensor
    • Wastewater treatment plant

    Fingerprint Dive into the research topics of 'Applying Kohonen self-organizing map as a software sensor to predict biochemical oxygen demand'. Together they form a unique fingerprint.

  • Cite this