A fuzzy rule-based approach for water quality assessment in the distribution network

In this paper, a fuzzy rule-based system with final evidential aggregation is proposed to perform the relative quality assessment of drinking water in the water distribution network (WDN). Partially reliable sensor measurements, incomplete assessments as well as subjective information on water quality parameters (WQP) introduce uncertainty to the water quality assessment process. Historical data recorded in a network are categorized into two groups including microbial and physicochemical parameters. Then, separate rule bases are developed to define microbial and physicochemical aspects of water quality. The distributed assessments of the water quality that result from two rule bases are aggregated using a fuzzy evidential reasoning algorithm. The proposed inference engine provides a decision support tool, which aids the decision makers to come up with management policies based on hundreds of water quality monitoring records. Statistical data on WQPs at fifty-two sampling locations of Quebec City main WDN were used to test the performance of the proposed framework.