Application of T-S Fuzzy Neural Network in Water Quality Classification and Evaluation

Water quality classification and evaluation is an important part of water resources protection and development. In order to further improve the accuracy and convenience of water quality evaluation classification, combined with the strong fuzzy information learning ability of T-S fuzzy neural network, the water quality monitoring data of a monitoring station in 2019 was used as the training set and validation set, and the 2020 data was used as the testing set, The T-S fuzzy neural network was used to construct a classifier. In view of the problem of data imbalance, The SMOTE oversampling method was used to generate data, and the model effects obtained by training before and after data balance were compared and analyzed. Finally, the trained model was used to classify and evaluate all the stations in the monitoring station and the results were analyzed. Experimental results showed that T-S fuzzy neural network had good results in water quality classification and evaluation.