Monthly runoff prediction based on variational modal decomposition combined with the dung beetle optimization...

Area of Science:

• Hydrology
• Water Resource Management
• Artificial Intelligence in Environmental Science

Background:

• Accurate monthly runoff forecasting is crucial for effective water resource management and utilization.
• Existing forecasting models often face challenges in achieving high prediction accuracy.
• Advancements in artificial intelligence and optimization algorithms offer potential for improved hydrological predictions.

Purpose of the Study:

• To develop and validate a novel hybrid model for enhanced monthly runoff forecasting.
• To integrate variational modal decomposition (VMD), dung beetle optimization algorithm (DBO), and gated recurrent unit (GRU) for improved prediction accuracy.
• To provide a reliable alternative for monthly runoff prediction in water resource management.

Main Methods:

• Decomposition of historical runoff data using Variational Modal Decomposition (VMD).
• Optimization of Gated Recurrent Unit (GRU) model parameters via the Dung Beetle Optimization (DBO) algorithm.
• Forecasting using the optimized GRU on decomposed runoff components, followed by consolidation of predictions.

Main Results:

• The proposed VMD-DBO-GRU model demonstrated superior prediction accuracy compared to baseline models (BP, SVM, GRU, VMD-GRU, DBO-GRU, EMD-GRU).
• Validation using extensive monthly runoff data (1980-2020) from the Ansha reservoir confirmed the model's effectiveness.
• The hybrid approach successfully captured complex patterns in monthly runoff data.

Conclusions:

• The VMD-DBO-GRU model offers a significant advancement in monthly runoff forecasting accuracy.
• This hybrid approach provides a robust and accurate tool for water resource management and planning.
• The study highlights the potential of combining advanced decomposition and optimization techniques with deep learning for hydrological forecasting.