Dynamic Stress Compensation Method for Large Deformation Flexible Body Based on Hybrid Neural Network Model

Abstract

The invention discloses a dynamic stress compensation method for a large deformation flexible body based on a hybrid neural network model, and belongs to the fields of parachute measurement technology and computer machine learning. This method can be applied to the dynamic stress compensation of flexible bodies with large deformation, especially the dynamic stress compensation in the working process of parachutes, hovercraft and other equipment. The hybrid neural network model includes a knowledge-based model and a backpropagation neural network model. The knowledge-based model represents the main characteristics of the large deformation flexible body mechanical model, and the backpropagation neural network represents the relationship between the large deformation flexible body mechanical model and the knowledge-based model. difference characteristics. The invention uses a high-precision universal testing machine to obtain sample data, and uses a hybrid neural network model to construct a dynamic stress compensation model. The compensation result has high precision and low computational complexity, and can be effectively applied to the dynamic stress compensation of large-deformation flexible bodies. It provides a new way of thinking about the dynamic stress compensation problem in this environment.

Description

technical field [0001] The invention belongs to the fields of parachute measurement technology and computer machine learning, and in particular relates to a dynamic stress compensation method for a large-deformation flexible body based on a hybrid neural network model (Hybrid Neural Network Model, HNNM). Background technique [0002] At present, the CFD (Computational Fluid Dynamics)-MSD (Mass Spring Damper) coupling model established on the basis of parachute measurement technology, aerodynamics and structural mechanics uses the basic knowledge of computational fluid dynamics and structural dynamics to perform numerical solutions and simulate The structure and flow field changes of large deformation flexible body during working process. This model can better reflect the nature of the aeroelastic force on the flexible body with large deformation. However, the uncertainty of the air flow rate, elastic coefficient and damping coefficient of the large deformation flexible body...

AI Technical Summary

Problems solved by technology

However, the hybrid neural network model in the prior art mainly compensates nonlinear stress and strain in the application scenarios of small deformation, large displacement and large rotation, and there is no related description of dynamic stress compensation applied to large deformation flexible bodies.

Images