Taking a certain coastal area of Jiangsu province as the research background, this study scientifically predicts the runoff on the medium and long-term time scale according to the changes of various climate factors such as atmospheric circulation, sea surface temperature, and solar activity in the first half of the year. A lag correlation is established between various related climate factors and the monthly runoff process in the research area for the previous 1–6 months. Selecting advantageous factors and constructing a significant factor set. Using the improved BP (Back-Propagation) artificial neural network model and combining it with the sensitivity analysis method, a specific number of 8-factor combinations are selected from the set of significant factors for medium and long-term runoff prediction. After that, the prediction results are compared with the forecasting effects of two multi-factor combination runoff simulation schemes formed by stepwise regression and Spearman rank correlation methods. The study concluded that the multi-factor combination simulation effect formed through sensitivity analysis was the best. The 20% standard forecast qualification rate of the three schemes is not significantly different. The Mean Absolute Relative Error of the multi-factor combination training and validation periods simulated through sensitivity analysis is the smallest among the three schemes, which are 36.61% and 38.01%, respectively. The Nash Efficiency Coefficient in the validation period is 0.45, which is far better than other schemes and has better generalization ability. The Standard Deviation of Relative Error in the training and validation periods is much smaller than other schemes, and the dispersion of relative errors is the smallest.

Due to the poor quality of current water consumption data and the frequent absence of industry information, accurately calculating the water consumption for different industries is difficult. Therefore, we propose a BERT-GCN model based on a strong link graph for classification within the water industry. First, we constructed a co-word relation graph based on the typical industry characteristics keywords extracted by the TF-IDF and extracted co-word relation features using a graph convolutional network (GCN). Then the web crawler was utilized to collect the main business data of the enterprise as additional information, and the semantic features were extracted from the supplementary information by the pre-trained language model BERT. Finally, we connected the semantic features with the co-word relation features to obtain the enhanced feature vector of the enterprise for the classification of the enterprise through the full connection layer. The experimental results in Xiuzhou District and Zhuji City show that compared with TextCNN, BERT-FC, TextGCN and Word2Vec-GCN models, the BERT-GCN has the best performance in classification evaluation indicators of precision, recall and F1-score. The relevant research provides technical and theoretical guidance for the government to carry out dynamic, rapid and accurate management of the water conservancy industry.