A fuzzy-logic based expert system for diagnosis and control of an integrated wastewater treatment

Abstract

A supervisory expert system based on fuzzy logic rules was developed for diagnosis and control of a lab- scale plant comprising anaerobic/anoxic/aerobic modules for combined high rate biological N and C removal. The design and implementation of a computational environment in LabVIEW for data acquisition, plant operation and distributed equipment control is described. The Fuzzy Logic toolbox for MATLAB was also used for the development of the fuzzy logic rule based system. The fuzzy rules were generated from quantitative and qualitative information, to identify the status of the plant operation and to decide the best commands to be sent to the final control elements to recover the stable operation in case of disturbances of the processes. A step increase in ammonia concentration from 20 to 60 mg N/L was applied during a trial period of 73 hours. Recycle flow rate from the aerobic to the anoxic module and by-pass flow rate from the influent directly to the anoxic reactor were the output of the fuzzy system that were automatically changed (from 34 to 111 L/day and from 8 to 13 L/day, respectively), when new plant conditions were recognized by the expert system. Denitrification efficiency higher than 85% was achieved 30 hours after the disturbance and 15 hours after the system response at an HRT as low as 1.5 hours. Nitrification efficiency gradually increased from 12 to 50% at an HRT of 3 hours. The system proved to properly react in order to set adequate operating conditions that timely led to recover efficient N and C removal rates.