EVALUATION AND PREDICTION OF SURFACE WATER POLLUTION IN CHINA BASED ON IMPROVED ENTROPY-WEIGHTED TOPSIS ALGORITHM AND METABOLISM GM (1,1) GREY MODEL

Surface water pollution evaluation and prediction is a complex dynamic decision-making problem involving balancing multiple conflicting internal factors, which poses a significant challenge to traditional management techniques. This paper combines improved entropy-weighted TOPSIS algorithm with metabolism GM (1,1) grey model to propose a new interdisciplinary research method that effectively overcomes the limitations of traditional methods. The effectiveness of the proposed method is verified through evaluation and prediction using surface water quality data from 20 regions in China. To further improve the accuracy of evaluation and prediction processes, a correlation analysis and standard normal distribution test were conducted on the water quality data in advance. Simultaneously, traditional, new information, and metabolism GM (1,1) grey models were constructed, and the metabolism GM (1,1) with the smallest sum of squared errors was selected as the final prediction model. The research indicates that on one hand, Nanjin Pass in Yichang City, Hubei Province, received the highest evaluation score of 0.0574, signifying the best water quality, while Heishan Tou in Hulun Buir City, Inner Mongolia, obtained the lowest evaluation score of 0.0323, indicating the worst. On the other hand, the percentage of qualified water in March, April and May 2024 was 80.402%, 81.205%, and 81.938% respectively. The research methodology in this paper can serve as a theoretical tool for planners and decision-makers to effectively monitor and manage surface water pollution.