Evaluation of Water Quality by Means of Probabilistic Multi-Objective Optimization

Abstract

Water quality assessment is a problem in a complex multi-attribute system, and rational approach is still in need for comprehensive evaluation. In this article, a Probabilistic Multi-Objective Optimization (PMOO) evaluation approach for water quality assessment is presented rationally. Firstly, it proposes a quantitative evaluation method of the utility value of each evaluated attribute (indicator), the integration of the utility value with the probabilistic multi-objective optimization is conducted to complete the evaluation and comparison of water quality evaluation comprehensively with a practical example from 7 sampling points of Jinchang city of Gansu province, China, to illuminate the procedure of the approach. Six performance metrics are involved, which include total salinity, total hardness, nitrite, nitrate, phenol, and hexavalent chromium. The evaluated result indicates that the sampling point 1 exhibits the highest total preferable probability and is ranked No. 1 in the comprehensive assessment of quality of water resources by means of PMOO. The proposed approach eliminates the human factors in the previous assessments for comprehensive evaluation with multiple performance metrics.