Study On Water Quality Evaluation And Prediction Of Artificial Landscape Water

With the intensification of human activities,the problem of water pollution is becoming more and more serious.Artificial landscape water is an important part of the water area,which is prone to eutrophication.The serious eutrophication will destroy the ecological balance of water quality.Therefore,the water quality of artificial landscape water is worth deepening.Research.This paper does the following work around water quality assessment and water quality prediction.This paper selects the artificial landscape lakes such as Jinghu Lake of Guangxi University as the research area,introduces the ecological situation of the area,and adopts scientific methods to set the sampling interval and sampling points.Then,18 indicators were selected for reviewing the literature for water quality research,and sampling methods for each indicator were introduced.Water quality evaluation.The degree of eutrophication can reflect the quality of artificial landscape water,and eutrophication is usually accompanied by the proliferation of algae,which can be characterized by the content of chlorophyll A.In order to effectively evaluate the water quality,the chlorophyll A content was used as the constraint factor,and the 18 primary selection indicators were initially screened by rough set theory,and 10 indicators were selected as pre-evaluation.Then the principal components analysis method is used to study the 10 indicators.The SPSS software is used to extract 2 principal components according to the cumulative contribution rate.Then,the load factor and the component score matrix are arranged to screen and obtain the water temperature,pH,total phosphorus and total nitrogen,chlorophyll A and light were used to establish a water quality evaluation index system.In order to obtain the weight of each evaluation index,the AHP questionnaire was invited to invite 30 water quality related professional experts in Guangxi to evaluate the scores.The group decision analytic hierarchy process was combined with YAAHP software to calculate the weight of each index,and the final water quality evaluation index system was obtained.And further build a comprehensive evaluation model of water quality.Water quality forecast.In order to assess the water quality in the future,water quality prediction studies are needed.First,the collected water quality data is analyzed,and the complete sequence of the corresponding data is obtained for further study.According to the water quality evaluation index system,the six input quantities of the prediction model are determined.In order to reduce the prediction error,a better prediction model is selected.This paper compares the BP neural network based on genetic algorithm optimization,the gray theory based on RBF neural network,and the particle swarm optimization algorithm.(PSO)Optimized the average squared error and the average percentage error of the predicted values of three combined models,such as support vector machine(SVM),and the prediction model based on PSO-SVM is more accurate.The established PSO-SVM prediction model was applied to the sampling test of the other three artificial landscape lakes of Guangxi University.Two methods were used to evaluate the water quality of the predicted results,and the predicted water quality grade results showed that the predicted water quality was basically in a moderate state of trophication.