Research On Water Quality Evaluation And Prediction Of Mopanshan Reservoi

The comprehensive establishment of the river chief system and lake chief system in the whole country is a great innovation in water culture at the institutional level,as well as a great innovation in water conservancy in the new era,which greatly enriches the connotation of Chinese water culture.In the first task of river chiefs,the ideas reflected are not only to manage and solve the water quality problems of rivers and lakes,but also to deeply understand the water quality problems of reservoirs and do a good job in the safety of reservoir water quality.The main method to ensure the safety of reservoir water quality is to conduct reasonable water quality evaluation,which is the basis for water environment management and protection.With the increasing strength of the country and the rapid social and economic development,water environment pollution and water shortage have posed a serious threat to the water security and regional economic development of the people,and water environment governance is indispensable in the face of increasingly serious water pollution problems,as well as human survival,social stability and sustainable economic development.Based on the Basic Research Fund of Provincial Universities（2020-KYYWF-1020）"Research on Reservoir Water Quality Evaluation Based on the Background of River Chief System",this paper conducts in-depth research and discussion on the reasonable evaluation of reservoir water quality and scientific water quality prediction with Mopanshan Reservoir as the research object.Using the monitoring data of Mopanshan Reservoir in 2021,the main factors affecting the water quality level of the reservoir were analyzed,and finally COD（chemical oxygen demand）,p H value,DO（dissolved oxygen）,and NH₃-N（ammonia nitrogen）were selected as the evaluation index,and the game theory method was used to combine the subjective weights determined by the AHP method with the objective weights determined by the entropy weight method to optimize,and a TOPSIS evaluation model for water quality of Mopanshan Reservoir based on AHP-entropy weight-game theory combination empowerment was constructed.The results of grey correlation model evaluation were compared.The evaluation results show that the water quality grades of Mopanshan Reservoir based on the TOPSIS model weighted by game theory combination are Class I and II.,and the scores from January to December are greater than 0.0547（Class II reservoir water quality score）.Mopanshan Reservoir had the highest water quality score of 0.0770 in May and the lowest score in August with a score of 0.0549,with a decrease in water quality,but the overall water quality was relatively stable.From the perspective of empowerment results,the polluting factor with greater influence was NH₃-N（ammonia nitrogen）index,and the water quality grades of Mopanshan Reservoir evaluated by the gray correlation model were Class I.,II.,III.,and the scores from January to December were all greater than 0.0415（Class III.reservoir water quality score）.Mopanshan Reservoir had the highest water quality score of 0.0739 in December and the lowest score in August with a score of 0.0512,indicating a decrease in water quality,but the overall water quality was relatively stable and met the needs of water safety.Using the monitoring data of Mopanshan Reservoir from 2019 to 2020,NH₃-N（ammonia nitrogen）and COD（chemical oxygen demand）indexes were predicted based on the gray prediction combination model and BP neural network,respectively.The prediction results of the gray prediction combination model were compared with the water quality monitoring data of Mopanshan Reservoir in 2021,and the prediction results showed that the error range of NH₃-N（ammonia nitrogen）prediction results was10%-45%,and the error range of COD（chemical oxygen demand）prediction results was 6%-19%,and the prediction results were good,and the prediction errors were within the acceptable range.The prediction results of BP neural network were compared with the water quality monitoring data of Mopanshan Reservoir in 2021,and the prediction results showed that the error range of NH₃-N（ammonia nitrogen）prediction results was 27%-38%,and the error range of COD（chemical oxygen demand）prediction results was 7%-17%,and the prediction errors were within the acceptable range.Therefore,the gray combination model and BP neural network model can be used in the prediction research of Mopanshan Reservoir to improve the risk prevention and control ability of Mopanshan Reservoir.