Prediction of the failure of a human supraspinatus tendon using a NARX network

2021 ◽  
Author(s):  
Fulufhelo Nemavhola ◽  
Harry Ngwangwa

The modelling of tendon behaviour during failure stages is nonlinear and heavily random. However, the understanding of its behavior during such stages, and development of models that can give an accurate prediction of its behavior during failure can provide a means for developing effective tendon therapies. This study is aimed at demonstrating the capability of an artificial neural network in the modelling of failure phases in tendons. A nonlinear autoregressive with exogenous inputs network is applied to three different tensile test data of the human supraspinatus tendons. Owing to data scarcity, the network was trained using two different test data which were randomly sampled and divided into 50%, 25% and 25% proportions for training, validation and preliminary testing. The third test data were used for the final testing phase. The procedure was cyclically performed for each of the results that have been presented in this study. The neural network predictions are presented as curves fitted over actual test results with corresponding error plots. The results indicate that the network is able to accurately predict the failure behaviour of these tendons with correlations of above 99 % for all tests. This is an excellent and very promising result in the light of the difficulties that most deterministic mechanistic models encounter in the modelling of soft tissue failure behaviour. With further development of this technique, sports and exercise physicians would enhance knowledge in mechanisms of tendon failure and be able to devise more injury preventive strategies.

2020 ◽  
Vol 2020 ◽  
pp. 1-14
Author(s):  
Wanli Li ◽  
Mingjian Chen ◽  
Chao Zhang ◽  
Lundong Zhang ◽  
Rui Chen

A navigation grade Strapdown Inertial Navigation System (SINS) combined with a Doppler Velocity Log (DVL) is widely used for autonomous navigation of underwater vehicles. Whether the DVL is able to provide continuous velocity measurements is of crucial importance to the integrated navigation precision. Considering that the DVL may fail during the missions, a novel neural network-based SINS/DVL integrated navigation approach is proposed. The nonlinear autoregressive exogenous (NARX) neural network, which is able to provide reliable predictions, is employed. While the DVL is available, the neural network is trained by the body frame velocity and its increment from the SINS and the DVL measurements. Once the DVL fails, the well trained network is able to forecast the velocity which can be used for the subsequent navigation. From the experimental results, it is clearly shown that the neural network is able to provide reliable velocity predictions for about 200 s–300 s during DVL malfunction and hence maintain the short-term accuracy of the integrated navigation.


Author(s):  
Jae Eun Yoon ◽  
Jong Joon Lee ◽  
Tong Seop Kim ◽  
Jeong Lak Sohn

This study aims to simulate performance deterioration of a microturbine and apply artificial neural network to its performance diagnosis. As it is hard to obtain test data with degraded component performance, the degraded engine data have been acquired through simulation. Artificial neural network is adopted as the diagnosis tool. First, the microturbine has been tested to get reference operation data, assumed to be degradation free. Then, a simulation program was set up to regenerate the performance test data. Deterioration of each component (compressor, turbine and recuperator) was modeled by changes in the component characteristic parameters such as compressor and turbine efficiency, their flow capacities and recuperator effectiveness and pressure drop. Single and double faults (deterioration of single and two components) were simulated to generate fault data. The neural network was trained with majority of the data sets. Then, the remaining data sets were used to check the predictability of the neural network. Given measurable performance parameters (power, temperatures, pressures) as inputs to the neural network, characteristic parameters of each component were predicted as outputs and compared with original data. The neural network produced sufficiently accurate prediction. Reducing the number of input data decreased prediction accuracy. However, excluding up to a couple of input data still produced acceptable accuracy.


2020 ◽  
Vol 2020 ◽  
pp. 1-6 ◽  
Author(s):  
Ghassane Benrhmach ◽  
Khalil Namir ◽  
Abdelwahed Namir ◽  
Jamal Bouyaghroumni

Time series analysis and prediction are major scientific challenges that find their applications in fields as diverse as finance, biology, economics, meteorology, and so on. Obtaining the method with the least prediction error is one of the difficult problems of financial market and investment analysts. State space modelling is an efficient and flexible method for statistical inference of a broad class of time series and other data. The neural network is an important tool for analyzing time series especially when it is nonlinear and nonstationary. Essential tools for the study of Box-Jenkins methodology, neural networks, and extended Kalman filter were put together. We examine the use of the nonlinear autoregressive neural network method as a prediction technique for financial time series and the application of the extended Kalman filter algorithm to improve the accuracy of the model. As application on a real example, we are analyzing the time series of the daily price of steel over a 790-day period for establishing the superiority of this method over other existing methods. The simulation results using MATLAB and R software show that the model is capable of producing a reasonable accuracy.


2014 ◽  
Vol 989-994 ◽  
pp. 544-547
Author(s):  
Quan Li ◽  
Wen Jun Liu ◽  
Ren Ju Cheng ◽  
Cheng Li ◽  
Shan Jiang ◽  
...  

In this paper, the Back-Propagation neural network (BP network) and the establishment of the AZ61 magnesium alloy high temperature constitutive model and test data obtained for training the neural network, after training the neural network to become a knowledge-based constitutive model formed AZ61 magnesium alloy flow stress and dynamic recrystallization of the neural network model tested by the neural network model with traditional regression methods predict contrast, results showed that the higher the accuracy of the neural network model for dealing with a large number of test data to establish the constitutive relations of materials with high stress and promotion of value.


2020 ◽  
Vol 1 (1) ◽  
pp. 1-11
Author(s):  
Muhammad Restu Alviando ◽  
Muhammad Ezar Al Rivan ◽  
Yoannita Yoannita

American Sign Language (ASL) is a sign language in the world. This study uses the neural network method as a classification and the scale invariant feature transform (SIFT) as feature extraction. Training data and test data for ASL images were extracted using the SIFT feature, then ANN training was conducted using 17 training functions with 2 hidden layers. There are architecture used [250-5-10-24], [250-5-15-24] and [250-15-15-24] so there are 3 different ANN architectures. Each architecture is performed 3 times so that there are 9 experiments (3 x 3 trials run the program). Determination of the number of neurons concluded by the training function is selected by the best test results on the test data. Based on the training function and the extraction of SIFT features as input values ​​in the neural network it can be concluded that from 17 training functions, trainb with neuron architecture [250-5-10-24] becomes the best training function producing an accuracy value of 95%, precision of 15 % and recall 5%.  


2018 ◽  
Vol 6 (2) ◽  
pp. 64-72
Author(s):  
Yulisman Yulisman

Abstract The application of E-Learning types that correspond to the characteristics of the E-Learning system can be determined by the user by filling out a questionnaire about the characteristics of the E-Learning System. Many types of E-Learning that have been used by the institution. To determine the type of E-Learning that is appropriate to the characteristics of the E-Learning system can be done by the user by filling out the assessment questionnaire about the characteristics of the E-Learning System. The data processing results of the questionnaire to the application of the Neural Network with the Perceptron method. Neural network is an information processing system that has characteristics similar to a network of nerve biology. Perceptron is a simple network that is usually used to classify a particular type of pattern that is often known as a linear separation. From the calculations that researchers do manually or testing of the training data and test data in accordance with the training measures Perceptron defined in equation f (net) = t (yi), and error = 0 to determine kesusksesannya, then from training data and test data produces the correct target and in accordance with the expected results, which means that the method Perceptron applied were able to predict the type of E-Learning in accordance with the characteristics of E-Learning system is correct and produce the same target with the data results of the questionnaire from the user. Calculation of percentage of E-Learning used STMIK Hang Tuah Pekanbaru corresponding characteristics of the system of E-Learning and Evaluation Model ISO 9126 elearning.htp.ac.id results obtained with the value -1 = No, the performance of 20%, and a value of 1 = Yes, performance 80% and the error = 0 of 20 training data and test patterns. kuliah.htp.ac.id with a value of -1 = No, the performance of 45%, and a value of 1 = Yes, the performance of 55% and error = 0 of 20 training data and test patterns. edmodo.com/es3jelita with a value of -1 = No, the performance of 80%, and a value of 1 = Yes, the performance of 20% and error = 0 of 20 training data and test patterns. Abstrak E-Learning adalah merupakan salah satu media pembelajaran yang memanfaatkan teknologi informasi yang didukung oleh perangkat elektronik serta jaringan internet. Banyak jenis E-Learning yang telah digunakan oleh lembaga pendidikan. Untuk menentukan jenis E-Learning yang sesuai dengan karakteristik sistem E-Learning bisa dilakukan penilaian oleh user dengan mengisi kuesioner tentang karakteristik sistem E-Learning. Pengolahan data hasil dari kuesioner tersebut bisa dengan penerapan Jaringan syaraf tiruan dengan metode Perceptron. Jaringan Syaraf Tiruan  adalah  sistem  pemrosesan informasi  yang  memiliki karakteristik  mirip  dengan jaringan  syaraf  biologi. Perceptron merupakan jaringan sederhana yang biasanya digunakan untuk mengklasifikasikan suatu pola tipe tertentu yang sering dikenal dengan pemisahan secara linear. Dari perhitungan yang peneliti lakukan secara manual atau pengujian terhadap data latih dan data uji yang sesuai dengan langkah-langkah pelatihan Perceptron yang sudah ditetapkan dalam persamaan f(net)= t(yi), dan error = 0 untuk menentukan kesusksesannya, maka dari data latih dan data uji menghasilkan target yang benar dan sesuai dengan hasil yang diharapkan, yang berarti bahwa metode Perceptron yang diterapkan mampu untuk memprediksi jenis E-Learning yang sesuai dengan karakteristik sistem E-Learning sudah benar dan menghasilkan target yang sama dengan data hasil kuesioner dari user. Perhitungan Persentase E-Learning yang digunakan STMIK Hang Tuah Pekanbaru yang sesuai karakteristik sistem E-Learning Model ISO 9126 dan Evaluasi didapat hasil elearning.htp.ac.id dengan nilai  -1=Tidak, performance 20%, dan nilai 1=Ya, performance 80% dan error =0 dari 20 pola data latih dan uji. kuliah.htp.ac.id dengan nilai  -1=Tidak, performance 45%, dan nilai 1=Ya, performance 55% dan error =0 dari 20 pola data latih dan uji. edmodo.com/es3jelita dengan nilai  -1=Tidak, performance 80%, dan nilai 1=Ya, performance 20% dan error =0 dari 20 pola data latih dan uji.


Author(s):  
Gerardo Diaz ◽  
Antonio Campo

Turbulent forced convection correlations are available in the literature for gases (Pr ∼ 0.7), but the test data leave a gap in the range of Prandtl (Pr) number between 0.1 and 0.7 occupied by binary gas mixtures. In this paper we develop a turbulent forced convection correlation for the Nusselt (Nu) number of in-tube binary gas mixtures for the ranges of Reynolds (Re) number between 104 and 106 and Prandtl (Pr) number between 0.1 and 0.7. A fully connected back-propagation Artificial Neural Network (ANN) is used to learn the pattern of Nu as a function of Re and Pr. Available test data in the range of 0.001 < Pr < 0.1 and 0.7 < Pr < 1000 are provided to the ANN. The test data are separated in two sets to train and test the neural network. A training set with 80% of the data is used to predict a testing set with the remaining 20% of the data. After the network is trained, we make use of the excellent nonlinear interpolation capabilities of ANNs, to predict values of Nu for the sought range 0.1 < Pr < 0.7. These predictions are later used to generate a correlation that aptly covers the complete range of Prandtl numbers.


Energies ◽  
2021 ◽  
Vol 14 (13) ◽  
pp. 4015
Author(s):  
Adán Alberto Jumilla-Corral ◽  
Héctor Enrique Campbell-Ramírez ◽  
Carlos Perez-Tello ◽  
Zulma Yadira Medrano-Hurtado ◽  
Pedro Mayorga-Ortiz ◽  
...  

This research presents the modeling and prediction of the harmonic behavior of current in an electric power supply grid with the integration of photovoltaic power by inverters using artificial neural networks to determine if the use of the proposed neural network is capable of capturing the harmonic behavior of the photovoltaic energy integrated into the user's electrical grids. The methodology used was based on the use of recurrent artificial neural networks of the nonlinear autoregressive with external input type. Work data were obtained from experimental sources through the use of a test bench, measurement, acquisition, and monitoring equipment. The input–output parameters for the neural network were the current values in the inverter and the supply grid, respectively. The results showed that the neural network can capture the dynamics of the analyzed system. The generated model presented flexibility in data handling, allowing to represent and predict the behavior of the harmonic phenomenon. The obtained algorithm can be transferred to physical or virtual systems for the control or reduction of harmonic distortion.


1994 ◽  
Vol 33 (01) ◽  
pp. 157-160 ◽  
Author(s):  
S. Kruse-Andersen ◽  
J. Kolberg ◽  
E. Jakobsen

Abstract:Continuous recording of intraluminal pressures for extended periods of time is currently regarded as a valuable method for detection of esophageal motor abnormalities. A subsequent automatic analysis of the resulting motility data relies on strict mathematical criteria for recognition of pressure events. Due to great variation in events, this method often fails to detect biologically relevant pressure variations. We have tried to develop a new concept for recognition of pressure events based on a neural network. Pressures were recorded for over 23 hours in 29 normal volunteers by means of a portable data recording system. A number of pressure events and non-events were selected from 9 recordings and used for training the network. The performance of the trained network was then verified on recordings from the remaining 20 volunteers. The accuracy and sensitivity of the two systems were comparable. However, the neural network recognized pressure peaks clearly generated by muscular activity that had escaped detection by the conventional program. In conclusion, we believe that neu-rocomputing has potential advantages for automatic analysis of gastrointestinal motility data.


Sign in / Sign up

Export Citation Format

Share Document