{"title":"[Comparative Study of Water Quality Prediction Methods Based on Different Artificial Neural Network].","authors":"Ming-Jun Xiao, Yi-Chun Zhu, Wen-Yuan Gao, Yu Zeng, Hao Li, Shuo-Fu Chen, Ping Liu, Hong-Li Huang","doi":"10.13227/j.hjkx.202310074","DOIUrl":null,"url":null,"abstract":"<p><p>The prediction of future data using existing data is an effective tool for regional planning and watershed management. The back propagation neural network (BPNN) and convolutional neural network (CNN) were used to construct a prediction model based on the water quality index of Hengyang in Xiangjiang River Basin from April to May 2022 and the results of permanganate index prediction by different models were compared. The prediction results displayed by BPNN could predict the water quality; however, overfitting occurred during the prediction. BPNN modified by particle swarm optimization (PSO) could avoid overfitting, which improved the parameter selection method of the BPNN mode. The CNN model had a better prediction effect, which had a more complex structure and a more scientific fitting method to avoid the model falling into the local extreme value during the fitting process and improve the accuracy of the model prediction results. The evaluation parameters including root-mean-square error (RMSE), coefficient of determination (<i>R</i><sup>2</sup>), and mean absolute error (MAE) were used to predict the accuracy of the network. Compared with that of the traditional BPNN model, PSO-BPNN reduced the RESM of the test set from 0.278 2 mg·L<sup>-1</sup> to 0.210 9 mg·L<sup>-1</sup>, reduced the MAE of the test set from 0.222 3 mg·L<sup>-1</sup> to 0.153 7 mg·L<sup>-1</sup> and increased the <i>R</i><sup>2</sup> of the test set from 0.864 0 to 0.921 8, which indicated that PSO-BPNN had more stable fitting ability. RMSE, MAE, and <i>R</i><sup>2</sup> of the test set in the CNN model were 0.122 0 mg·L<sup>-1</sup>, 0.092 7 mg·L<sup>-1</sup>, and 0.970 5, respectively, which showed that CNN had a better fitting and prediction effect than that of BPNN.</p>","PeriodicalId":35937,"journal":{"name":"Huanjing Kexue/Environmental Science","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Huanjing Kexue/Environmental Science","FirstCategoryId":"1087","ListUrlMain":"https://doi.org/10.13227/j.hjkx.202310074","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"Environmental Science","Score":null,"Total":0}
引用次数: 0
Abstract
The prediction of future data using existing data is an effective tool for regional planning and watershed management. The back propagation neural network (BPNN) and convolutional neural network (CNN) were used to construct a prediction model based on the water quality index of Hengyang in Xiangjiang River Basin from April to May 2022 and the results of permanganate index prediction by different models were compared. The prediction results displayed by BPNN could predict the water quality; however, overfitting occurred during the prediction. BPNN modified by particle swarm optimization (PSO) could avoid overfitting, which improved the parameter selection method of the BPNN mode. The CNN model had a better prediction effect, which had a more complex structure and a more scientific fitting method to avoid the model falling into the local extreme value during the fitting process and improve the accuracy of the model prediction results. The evaluation parameters including root-mean-square error (RMSE), coefficient of determination (R2), and mean absolute error (MAE) were used to predict the accuracy of the network. Compared with that of the traditional BPNN model, PSO-BPNN reduced the RESM of the test set from 0.278 2 mg·L-1 to 0.210 9 mg·L-1, reduced the MAE of the test set from 0.222 3 mg·L-1 to 0.153 7 mg·L-1 and increased the R2 of the test set from 0.864 0 to 0.921 8, which indicated that PSO-BPNN had more stable fitting ability. RMSE, MAE, and R2 of the test set in the CNN model were 0.122 0 mg·L-1, 0.092 7 mg·L-1, and 0.970 5, respectively, which showed that CNN had a better fitting and prediction effect than that of BPNN.