Related Experiment Video
Updated: Oct 4, 2025
09:32
Subject-specific Musculoskeletal Model for Studying Bone Strain During Dynamic Motion
Published on: April 11, 2018
9.8K
Feedforward backpropagation artificial neural networks for predicting mechanical responses in complex nonlinear
Saeed Mouloodi1, Hadi Rahmanpanah1, Soheil Gohari1
1Department of Mechanical Engineering, The University of Melbourne, Melbourne, Australia.
Journal of the Mechanical Behavior of Biomedical Materials
|February 3, 2022
Summary
This study successfully trained artificial neural networks (ANNs) to predict the mechanical behavior of complex materials like long bones. It identified key parameters for optimal network performance and generalization, avoiding overfitting.
Area of Science:
- Materials Science
- Computational Engineering
- Biomechanical Engineering
Background:
- Artificial neural networks (ANNs) are increasingly used for materials modeling, but optimal training methods and influential parameters are not well-defined.
- Long bones, as complex composite materials, present significant challenges for traditional constitutive modeling due to their non-homogeneous and anisotropic properties.
Purpose of the Study:
- To predict the mechanical responses of long bones using feedforward backpropagation ANNs.
- To identify optimal ANN architectures and investigate parameters influencing prediction accuracy and generalization.
- To provide insights into training robust ANNs for complex engineering structures.
Main Methods:
- Utilized experimental recordings to train feedforward backpropagation ANNs for predicting loading, displacement, and strains.
- Investigated the impact of training algorithms, data noise injection, noise levels, and data normalization on ANN performance.
- Evaluated over 60,000 ANNs to determine optimal network structures and assess generalization ability, focusing on preventing overfitting.
Main Results:
- Developed ANNs capable of accurately predicting mechanical responses of long bone materials.
- Identified crucial parameters, including training algorithms and data preprocessing techniques, that significantly enhance ANN prediction accuracy.
- Demonstrated the importance of examining generalization ability to ensure unbiased and reliable ANNs.
Conclusions:
- Successfully applied ANNs to model the complex mechanical behavior of long bones, overcoming limitations of traditional methods.
- Established a framework for training and validating ANNs, offering guidance for researchers in complex materials modeling.
- The extensive validation of numerous ANNs provides a promising foundation for understanding intricate engineering structures with nonlinear properties.
Related Concept Videos
01:21Linear Approximation in Time Domain
135
Nonlinear systems often require sophisticated approaches for accurate modeling and analysis, with state-space representation being particularly effective. This method is especially useful for systems where variables and parameters vary with time or operating conditions, such as in a simple pendulum or a translational mechanical system with nonlinear springs.
For a simple pendulum with a mass evenly distributed along its length and the center of mass located at half the pendulum's length,...
For a simple pendulum with a mass evenly distributed along its length and the center of mass located at half the pendulum's length,...
135
01:40Bone Remodeling
38.7K
Bone remodeling is a continuous and balanced process of bone resorption by osteoclasts and bone formation by osteoblasts. In adults, it helps maintain bone mass and calcium homeostasis. While mechanical stress can stimulate turnover as part of the normal maintenance and reparative process, several hormones also regulate bone remodeling.
38.7K