Utilizing deep belief network optimized by balanced Manta ray foraging optimization algorithm for estimating the...

Area of Science:

• Structural Engineering
• Seismic Design
• Computational Mechanics

Background:

• Shear walls are critical for building lateral load resistance, especially seismic forces.
• Accurate shear capacity prediction in concrete shear walls is vital for seismic safety.
• Existing methods may lack precision in complex structural analyses.

Purpose of the Study:

• To develop and validate a novel model for predicting the shear strength of concrete shear walls.
• To enhance seismic safety through accurate shear capacity estimation.
• To investigate the influence of various parameters on shear strength prediction.

Main Methods:

• Utilized a Deep Belief Network (DBN) for predictive modeling.
• Optimized the DBN using the Balanced Manta Ray Foraging Optimization Algorithm (BMRFOA).
• Trained the model on a dataset of 60 laboratory tests, considering parameters like wall thickness, reinforcement ratios, dimensions, and material strengths.

Main Results:

• The DBN-BMRFOA model demonstrated reliable estimation of concrete shear wall shear capacity.
• The gamma (γ) ratio was identified as the most influential parameter for predictive accuracy.
• The model achieved a satisfactory error margin of approximately 7%.

Conclusions:

• The DBN-BMRFOA approach is effective for predicting the shear strength of reinforced concrete shear walls.
• This computational method offers a reliable tool for seismic safety assessments in structural engineering.
• Further research can explore broader applications and parameter variations.