A Novel Constitutive Parameters Identification Procedure for Hyperelastic Skeletal Muscles Using Two-Way Neural Networks

2021 ◽  
pp. 2150060
Author(s):  
Yang Li ◽  
Jianbing Sang ◽  
Xinyu Wei ◽  
Zijian Wan ◽  
G. R. Liu
Keyword(s):  
Neural Networks ◽  
Skeletal Muscles ◽  
Work Load ◽  
Hyperelastic Material ◽  
Load Prediction ◽  
Material Parameters ◽  
Penalty Factor ◽  
Comparison Results ◽  
Predicted Values ◽  
Constitutive Parameters

Muscle soreness can occur after working beyond the habitual load, especially for people engaged in high-intensity work load. Prediction of hyperelastic material parameters is essentially an inverse process, which possesses challenges. This work presents a novel procedure that combines nonlinear finite element method (FEM), two-way neural networks (NNs) together with experiments, to predict the hyperelastic material parameters of skeletal muscles. FEM models are first established to simulate nonlinear deformation of skeletal muscles subject to compressions. A dataset of nonlinear relationship between nominal stress and principal stretch of skeletal muscles is created using our FEM models. The dataset is then used to establish two-way NNs, in which a forward NN is trained and it is in turn used to train the inverse NN. The inverse NN is used to predict the hyperelastic material parameters of skeletal muscles. Finally, experiments are carried out using fresh skeletal muscles to validate the predictions in great detail. In order to examine the accuracy of the two-way NNs predicted values against the experimental ones, a decision coefficient [Formula: see text] with penalty factor is introduced to evaluate the performance. Studies have also been conducted to compare the present two-way NNs approach with the other existing methods, including the directly (one-way) inverse problem NN, and improved niche genetic algorithm (INGA). The comparison results show that two-way NNs model is an accurate approach to identify the hyperelastic parameters of skeletal muscles. The present two-way NNs method can be further expanded to the predictions of constitutive parameters of other type of nonlinear materials.

scholarly journals Estimating Johnson-Cook Material Parameters using Neural Networks

2021 ◽  
Vol 53 ◽  
pp. 680-689
Author(s):  
Nesar Ahmed Titu ◽  
Matt Baucum ◽  
Timothy No ◽  
Mitchell Trotsky ◽  
Jaydeep Karandikar ◽  
...  
Keyword(s):  
Neural Networks ◽  
Material Parameters

scholarly journals Biaxial estimation of biomechanical constitutive parameters of passive porcine sclera soft tissue

2021 ◽  
Author(s):  
Zwelihle Ndlovu ◽  
Dawood Desai ◽  
Thanyani Pandelani ◽  
Harry Ngwangwa ◽  
Fulufhelo Nemavhola
Keyword(s):  
Finite Element ◽  
Soft Tissue ◽  
Finite Element Model ◽  
Test Data ◽  
Element Model ◽  
Finite Element Models ◽  
Anisotropic Model ◽  
Material Models ◽  
Material Parameters ◽  
Constitutive Parameters

This study assesses the modelling capabilities of four constitutive hyperplastic material models to fit the experimental data of the porcine sclera soft tissue. It further estimates the material parameters and discusses their applicability to a finite element model by examining the statistical dispersion measured through the standard deviation. Fifteen sclera tissues were harvested from porcine’ slaughtered at an abattoir and were subjected to equi-biaxial testing. The results show that all the four material models yielded very good correlations at correlations above 96 %. The polynomial (anisotropic) model gave the best correlation of 98 %. However, the estimated material parameters varied widely from one test to another such that there would be needed to normalise the test data to avoid long optimisation processes after applying the average material parameters to finite element models. However, for application of the estimated material parameters to finite element models, there would be needed to consider normalising the test data to reduce the search region for the optimisation algorithms. Although the polynomial (anisotropic) model yielded the best correlation, it was found that the Choi-Vito had the least variation in the estimated material parameters thereby making it an easier option for application of its material parameters to a finite element model and also requiring minimum effort in the optimisation procedure. For the porcine sclera tissue, it was found that the anisotropy more influenced by the fiber-related properties than the background material matrix related properties.

Dead Sea Water Levels Analysis Using Artificial Neural Networks and Firefly Algorithm

2022 ◽  
pp. 1118-1129
Author(s):  
Nawaf N. Hamadneh
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Water Level ◽  
Firefly Algorithm ◽  
Sea Water ◽  
Dead Sea ◽  
Water Levels ◽  
Artificial Neural ◽  
The Dead ◽  
Predicted Values

In this study, the performance of adaptive multilayer perceptron neural network (MLPNN) for predicting the Dead Sea water level is discussed. Firefly Algorithm (FFA), as an optimization algorithm is used for training the neural networks. To propose the MLPNN-FFA model, Dead Sea water levels over the period 1810–2005 are applied to train MLPNN. Statistical tests evaluate the accuracy of the hybrid MLPNN-FFA model. The predicted values of the proposed model were compared with the results obtained by another method. The results reveal that the artificial neural network (ANN) models exhibit high accuracy and reliability for the prediction of the Dead Sea water levels. The results also reveal that the Dead Sea water level would be around -450 until 2050.

Simulation of Stock Prediction System using Artificial Neural Networks

2020 ◽  
pp. 511-530
Author(s):  
Omisore Olatunji Mumini ◽  
Fayemiwo Michael Adebisi ◽  
Ofoegbu Osita Edward ◽  
Adeniyi Shukurat Abidemi
Keyword(s):  
Neural Networks ◽  
Test Data ◽  
Stock Prices ◽  
Prediction Accuracy ◽  
Stock Exchange ◽  
Training Data ◽  
Prediction System ◽  
Closing Price ◽  
Non Linear ◽  
Predicted Values

Stock trading, used to predict the direction of future stock prices, is a dynamic business primarily based on human intuition. This involves analyzing some non-linear fundamental and technical stock variables which are recorded periodically. This study presents the development of an ANN-based prediction model for forecasting closing price in the stock markets. The major steps taken are identification of technical variables used for prediction of stock prices, collection and pre-processing of stock data, and formulation of the ANN-based predictive model. Stock data of periods between 2010 and 2014 were collected from the Nigerian Stock Exchange (NSE) and stored in a database. The data collected were classified into training and test data, where the training data was used to learn non-linear patterns that exist in the dataset; and test data was used to validate the prediction accuracy of the model. Evaluation results obtained from WEKA shows that discrepancies between actual and predicted values are insignificant.

scholarly journals An Investigation of Deep Learning Models for EEG-Based Emotion Recognition

2020 ◽  
Vol 14 ◽  
Author(s):  
Yaqing Zhang ◽  
Jinling Chen ◽  
Jen Hong Tan ◽  
Yuxuan Chen ◽  
Yunyi Chen ◽  
...  
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Deep Learning ◽  
Emotion Recognition ◽  
Real Life ◽  
Learning Rate ◽  
Learning Models ◽  
Learning Methods ◽  
Time Frequency ◽  
Comparison Results

Emotion is the human brain reacting to objective things. In real life, human emotions are complex and changeable, so research into emotion recognition is of great significance in real life applications. Recently, many deep learning and machine learning methods have been widely applied in emotion recognition based on EEG signals. However, the traditional machine learning method has a major disadvantage in that the feature extraction process is usually cumbersome, which relies heavily on human experts. Then, end-to-end deep learning methods emerged as an effective method to address this disadvantage with the help of raw signal features and time-frequency spectrums. Here, we investigated the application of several deep learning models to the research field of EEG-based emotion recognition, including deep neural networks (DNN), convolutional neural networks (CNN), long short-term memory (LSTM), and a hybrid model of CNN and LSTM (CNN-LSTM). The experiments were carried on the well-known DEAP dataset. Experimental results show that the CNN and CNN-LSTM models had high classification performance in EEG-based emotion recognition, and their accurate extraction rate of RAW data reached 90.12 and 94.17%, respectively. The performance of the DNN model was not as accurate as other models, but the training speed was fast. The LSTM model was not as stable as the CNN and CNN-LSTM models. Moreover, with the same number of parameters, the training speed of the LSTM was much slower and it was difficult to achieve convergence. Additional parameter comparison experiments with other models, including epoch, learning rate, and dropout probability, were also conducted in the paper. Comparison results prove that the DNN model converged to optimal with fewer epochs and a higher learning rate. In contrast, the CNN model needed more epochs to learn. As for dropout probability, reducing the parameters by ~50% each time was appropriate.

Sign in / Sign up

Export Citation Format

Share Document