Research on flood prediction based on optimized AI model VMD_BiLSTM : A case study in BeiJing flood, China

Accurate water level prediction plays a key role in flood prevention and disaster relief. Many scholars have applied machine learning to the field of water level prediction, but the rapid fluctuation of water levels poses a huge challenge to traditional machine learning models. To address this problem, this study proposed an improved artificial intelligence algorithm, the VMD-BiLSTM model. The model decomposes complex time series into relatively stable patterns, improving the prediction accuracy, especially when dealing with irregular data. The simulation experiment demonstrated the model's prediction performance. During testing, the R² value of the LSTM model was 0.584 and the RMSE value was 0.021m, while the R² value of the BiLSTM model increased to 0.773 and the RMSE decreased to 0.016m. In contrast, the optimized VMD-BiLSTM model performed better, with an R² value of 0.940 and a RMSE value of 0.008m. Compared with the LSTM model, the R² of the VMD-BiLSTM model increased by 61.05% and the RMSE decreased by 61.88%. The outcomes demonstrate that the VMD-BiLSTM model is not only significantly better than the traditional LSTM model in prediction accuracy, but also has better generalization and prediction capabilities. This approach significantly advances flood prediction capabilities, offering practical value for disaster prevention and water resource management applications.