Management and prediction of river flood utilizing optimization approach of artificial intelligence evolutionary...

Area of Science:

• Environmental Science
• Geospatial Analysis
• Artificial Intelligence

Background:

• Flooding poses significant global risks, necessitating effective flood susceptibility mapping (FSM), especially in urban environments.
• Traditional flood prediction models struggle with data complexity and computational demands.
• Artificial intelligence, specifically evolutionary algorithms, offers potential for more robust and adaptable flood prediction.

Purpose of the Study:

• To evaluate artificial neural network (ANN) algorithms integrated with evolutionary computation for FSM.
• To enhance flood prediction accuracy and optimize flood management strategies using novel AI approaches.
• To assess the performance of Black Hole Algorithm (BHA), Future Search Algorithm (FSA), Heap-based Optimization (HBO), and Multiverse Optimization (MVO) for FSM.

Main Methods:

• Utilized ANN models trained on eight key flood-influencing geographical factors: elevation, rainfall, slope, NDVI, aspect, geology, land use, and river data.
• Employed evolutionary algorithms (BHA, FSA, HBO, MVO) to optimize ANN performance for flood prediction.
• Validated models using historical flood data from the Fars region, employing metrics such as Mean Square Error (MSE), Mean Absolute Error (MAE), and Receiver Operating Characteristic (ROC) curves.

Main Results:

• Hybridized models (BHA-MLP, FSA-MLP, MVO-MLP, HBO-MLP) demonstrated substantial improvements in prediction accuracy.
• Significant increases in accuracy indices and Area Under the Curve (AUC) values were observed across the tested evolutionary algorithms.
• The study confirmed the efficacy of evolutionary AI in enhancing FSM and flood management.

Conclusions:

• Hybridized evolutionary AI models provide an effective and cost-efficient solution for urban flood vulnerability mapping.
• These advanced models offer valuable insights for improving flood preparedness and emergency response planning.
• The integration of evolutionary algorithms represents a significant advancement in AI-driven geospatial risk assessment.