Development of a PSO-BP neural network model for evaluating material creep properties using three-point bending specimen with fixed constraints

Puntos clave

• This model effectively predicts material creep properties, showing improved accuracy in evaluations.
• Key evidence from the analysis demonstrates a significant reduction in prediction error up to 25%.
• The approach utilizes a PSO-BP neural network model in a three-point bending test with fixed constraints for robust evaluation.
• The findings highlight the potential for refined material testing protocols in engineering applications.