scholarly journals Comparison of Neural Network Error Measures for Simulation of Slender Marine Structures

2014 ◽  
Vol 2014 ◽  
pp. 1-11 ◽  
Author(s):  
Niels H. Christiansen ◽  
Per Erlend Torbergsen Voie ◽  
Ole Winther ◽  
Jan Høgsberg
Keyword(s):  
Neural Network ◽  
Error Function ◽  
Error Measure ◽  
Error Functions ◽  
Error Measures ◽  
Specific Objective ◽  
The Mean ◽  
Stress Cycles ◽  
Fatigue Life Analysis ◽  
Artificial Neural Network Ann

Training of an artificial neural network (ANN) adjusts the internal weights of the network in order to minimize a predefined error measure. This error measure is given by an error function. Several different error functions are suggested in the literature. However, the far most common measure for regression is the mean square error. This paper looks into the possibility of improving the performance of neural networks by selecting or defining error functions that are tailor-made for a specific objective. A neural network trained to simulate tension forces in an anchor chain on a floating offshore platform is designed and tested. The purpose of setting up the network is to reduce calculation time in a fatigue life analysis. Therefore, the networks trained on different error functions are compared with respect to accuracy of rain flow counts of stress cycles over a number of time series simulations. It is shown that adjusting the error function to perform significantly better on a specific problem is possible. On the other hand. it is also shown that weighted error functions actually can impair the performance of an ANN.

scholarly journals A Novel Neural Network Model for Blood Pressure Estimation Using Photoplethesmography without Electrocardiogram

2018 ◽  
Vol 2018 ◽  
pp. 1-9 ◽  
Author(s):  
Ludi Wang ◽  
Wei Zhou ◽  
Ying Xing ◽  
Xiaoguang Zhou
Keyword(s):  
Neural Network ◽  
Blood Pressure ◽  
Mean Absolute Error ◽  
Wearable Sensors ◽  
Absolute Error ◽  
Noninvasive Estimation ◽  
Treatment Of Hypertension ◽  
The Mean ◽  
Artificial Neural Network Ann ◽  
Blood Pressure Estimation

The prevention, evaluation, and treatment of hypertension have attracted increasing attention in recent years. As photoplethysmography (PPG) technology has been widely applied to wearable sensors, the noninvasive estimation of blood pressure (BP) using the PPG method has received considerable interest. In this paper, a method for estimating systolic and diastolic BP based only on a PPG signal is developed. The multitaper method (MTM) is used for feature extraction, and an artificial neural network (ANN) is used for estimation. Compared with previous approaches, the proposed method obtains better accuracy; the mean absolute error is 4.02 ± 2.79 mmHg for systolic BP and 2.27 ± 1.82 mmHg for diastolic BP.

Best Architecture of an Artificial Neural Network to Model Prefilming Airblast Atomization: Not So Deep Learning

2021 ◽  
Author(s):  
Geoffroy Chaussonnet ◽  
Sebastian Gepperth ◽  
Simon Holz ◽  
Rainer Koch ◽  
Hans-Jörg Bauer
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Feature Space ◽  
Middle Layer ◽  
Ann Model ◽  
Deformation Velocity ◽  
The Mean ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
Domain Of Parameters

Abstract A fully connected Artificial Neural Network (ANN) is used to predict the mean spray characteristics of prefilming airblast atomization. The model is trained from the planar prefilmer experiment from the PhD thesis of Gepperth (2020). The output of the ANN model are the Sauter Mean Diameter, the mean droplet axial velocity, the mean ligament length and the mean ligament deformation velocity. The training database contains 322 different operating points. Two types of model input quantities are investigated and compared. First, nine dimensional parameters are used as inputs for the model. Second, nine non-dimensional groups commonly used for liquid atomization are derived from the first set of inputs. The best architecture is determined after testing over 10000 randomly drawn ANN architectures, with up to 10 layers and up to 128 neurons per layer. The striking results is that for both types of model, the best architectures consist of only 3 hidden layer in the shape of a diabolo. This shape recalls the shape of an autoencoder, where the middle layer would be the feature space of reduced dimensionality. It was found that the model with dimensional input quantities always shows a lower test and validation errors than the one with non-dimensional input quantities. In general, the two types of models provide comparable accuracy, better than typical correlations of SMD and droplet velocity. Finally the extrapolation capability of the models was assessed by a training them on a confined domain of parameters and testing them outside this domain.

Modeling of Phenol Degradation in Spouted Bed Contactor Using Artificial Neural Network (ANN)

2008 ◽  
Vol 3 (2) ◽  
Author(s):  
Madhukar A. Dabhade ◽  
M. B. Saidutta ◽  
D. V. R. Murthy
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Mean Square Error ◽  
Model Prediction ◽  
Phenol Degradation ◽  
Mean Square ◽  
Ann Model ◽  
Spouted Bed ◽  
The Mean ◽  
Artificial Neural Network Ann

Presence of phenol and phenolic compounds in various wastewaters and its harmful effects has led to the use of different treatment methods. Work on biological methods shows the use of different microorganisms and different bioreactors so as to improve the removal efficiency economically. The present work deals with the use of N. hydrocarbonoxydans (NCIM 2386), an actinomycetes, for the degradation of phenol. N. hydrocarbonoxydans was immobilized on GAC and used in a spouted bed contactor for effective contact of microorganisms and the substrate. The contactor performance was studied by varying flow rates, influent concentrations and the solids loading in the contactor. The effect of these variables on phenol degradation was investigated and modeling study was carried out using the artificial neural network (ANN). A feed forward neural network with back propagation was used for the model development. The experiments were planned as per the face centered cube design (FCCD) and used for training of the model, whereas data from four other experimental runs were used for testing and validation of the model. The network was optimized for the number of neurons based on the mean square error. The ANN model with three layers with three input neurons, eight neurons in hidden layers and one output neuron was found to predict effectively the effluent concentration for the given operating conditions in the spouted bed contactor. The mean square error was found to be 9.318e-12 for this ANN model. Also the experimental data was used to develop second order nonlinear empirical model obtained using multiple regression (MR) and the results compared with ANN using correlation coefficient (R2), average absolute error (AAE) and root mean square error (RMSE). Results show that R2, AAE and RMSE values of MR model were 0.9363, 2.085 % and 2.338 % respectively, while in case of ANN model these values were 0.9995, 0.59 % and 1.263 % respectively. This shows that ANN model prediction is better than multiple regression model prediction.

Optimization of the Neural Network by Using the LabVIEW Instrumentation

2012 ◽  
Vol 463-464 ◽  
pp. 1011-1016 ◽  
Author(s):  
Adrian Olaru ◽  
Serban Olaru ◽  
Dan Paune ◽  
Oprean Aurel
Keyword(s):  
Neural Network ◽  
Transfer Functions ◽  
Error Function ◽  
Virtual Instrumentation ◽  
Closed Loops ◽  
Network Parameters ◽  
The Neural Network ◽  
On Line ◽  
The Mean ◽  
Number Of Iterations

The paper shown one assisted method to construct simple and complex neural network and to simulate on-line them. By on-line simulation of some more important neural simple and complex network is possible to know what will be the influences of all network parameters like the input data, weight, biases matrix, sensitive functions, closed loops and delay of time. There are shown some important neurons type, transfer functions, weights and biases of neurons, and some complex layers with different type of neurons. By using the proper virtual LabVIEW instrumentation in on-line using, were established some influences of the network parameters to the number of iterations before canceled the mean square error to the target. Numerical simulation used the proper teaching law and proper virtual instrumentation. In the optimization step of the research on used the minimization of the error function between the output and the target.

IP Core Based Hardware Implementation of Multi-Layer Perceptrons on FPGAs: A Parallel Approach

2012 ◽  
Vol 433-440 ◽  
pp. 5647-5653 ◽  
Author(s):  
Xiao Jun Li ◽  
Lin Li
Keyword(s):  
Neural Network ◽  
Hardware Implementation ◽  
Research Work ◽  
Universal Function ◽  
Multi Layer Perceptron ◽  
The Mean ◽  
Ip Cores ◽  
Artificial Neural Network Ann ◽  
Function Approximator ◽  
Mean Time

There’re many models derived from the famous bio-inspired artificial neural network (ANN). Among them, multi-layer perceptron (MLP) is widely used as a universal function approximator. With the development of EDA and recent research work, we are able to use rapid and convenient method to generate hardware implementation of MLP on FPGAs through pre-designed IP cores. In the mean time, we focus on achieving the inherent parallelism of neural networks. In this paper, we firstly propose the hardware architecture of modular IP cores. Then, a parallel MLP is devised as an example. At last, some conclusions are made.

scholarly journals Artificial Neural Network Approach for Solving Fuzzy Fractional order initial value problems under H-differentiability

2020 ◽  
Author(s):  
Somayeh Ezadi ◽  
Tofigh Allahviranloo
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Initial Conditions ◽  
Error Function ◽  
Back Propagation ◽  
Back Propagation Algorithm ◽  
Neural Network Approach ◽  
Error Back Propagation ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

This paper aims to solve the celebrated Fuzzy Fractional Differential Equations (FFDE) using an Artificial Neural Network (ANN) technique. Compared to the integer order differential equation, the proposed FFDE can better describe several real application problems of various physical systems. To accomplish the aforementioned aim, the error back propagation algorithm and a multi-layer feed forward neural architecture are utilized using the unsupervised learning in order to minimize the error function as well as the modification of the parameters such as weights and biases. By combining the initial conditions with the ANN, output provides an appropriate approximate solution of the proposed FFDE. Then, two illustrative examples are solved to confirm the applicability of the concept as well as to demonstrate both the precision and effectiveness of the developed method. By comparing with some traditional methods, the obtained results reveals a close match that confirms both accuracy and correctness of the proposed method.

scholarly journals Prediction of mean wave overtopping at simple sloped breakwaters using kernel-based methods

2021 ◽  
Author(s):  
Shabnam Hosseinzadeh ◽  
Amir Etemad-Shahidi ◽  
Ali Koosheh
Keyword(s):  
Neural Network ◽  
Sensitivity Analysis ◽  
Gaussian Process Regression ◽  
Structural Features ◽  
Support Vector ◽  
Ann Model ◽  
Wave Overtopping ◽  
Taylor Diagram ◽  
The Mean ◽  
Artificial Neural Network Ann

Abstract The accurate prediction of the mean wave overtopping rate at breakwaters is vital to have a safe design. Hence, providing a robust tool as a preliminary estimator can be useful for practitioners. Recently, soft computing tools such as artificial neural network (ANN) have been developed as alternatives to traditional overtopping formulae. The goal of this paper is to assess the capabilities of two kernel-based methods namely Gaussian process regression (GPR) and support vector regression for the prediction of mean wave overtopping rate at sloped breakwaters. An extensive dataset taken from EurOtop (2018) database, including rubble mound structures with permeable core, straight slopes, without berm, and crown wall, was employed to develop the models. Different combinations of the important dimensionless parameters representing structural features and wave conditions were tested based on the sensitivity analysis for developing the models. The obtained results were compared with those of the ANN model and the existing empirical formulae. The modified Taylor diagram was used to compare the models graphically. The results showed the superiority of kernel-based models, especially the GPR model over the ANN model and empirical formulae. In addition, the optimal input combination was introduced based on accuracy and the number of input parameters criteria. Finally, the physical consistencies of developed models were investigated the results, of which demonstrated the reliability of kernel-based models in terms of delivering physics of overtopping phenomenon.

scholarly journals Prediction of Ship Heave Motion Using Regularized BP Neural Network with Cross Entropy Error Function

2021 ◽  
Vol 14 (1) ◽  
Author(s):  
Hailun Wang ◽  
Fei Wu ◽  
Dongge Lei
Keyword(s):  
Neural Network ◽  
Loss Function ◽  
Error Function ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Cross Entropy ◽  
Saturation Phenomenon ◽  
Entropy Error ◽  
The Mean ◽  
Heave Motion

AbstractAccurate prediction of ship’s heave motion can greatly enhance the safety of offshore operation. Due to its complexity and nonlinearity, however, ship’s heave motion prediction is a difficult task to be solved. In this paper, a new method for predicting ship’s heave motion is proposed based on an improved back propagation neural network (IBPNN). To overcome the gradient saturation phenomenon of traditional BPNN, the mean square error (MSE) loss function is replaced with a cross entropy (CE) loss function in IBPNN. Meanwhile, the weights of IBPNN is regularized by $$L_2$$ L 2 norm to enhance the generalization ability of traditional BPNN. Finally, conjugate gradient method is adopted to train IBPNN. The IBPNN is used to predict ship’s heave motion and the prediction results prove its effectiveness.

Forecasting Coconut Yield: A Comparative Study between the Use of Traditional Forecasting and Feed Forward Back Propagation Artificial Neural Network

10.17158/320 ◽  
2014 ◽  
Vol 18 (2) ◽  
Author(s):  
Eric John G. Emberda ◽  
Den Ryan L. Dumas ◽  
Timothy Pierce M. Rentillo
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Back Propagation ◽  
Ann Model ◽  
Feed Forward ◽  
Feed Forward Back Propagation ◽  
Forecast Data ◽  
The Mean ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

<p>This study compared the use of Linear Regression and Feed Forward Backpropagation Artificial Neural Network (ANN) in forecasting the coconut yield and copra yield of a selected area in Davao region. Raw data were gathered from the Philippine Coconut Authority, Davao Research Center. An ANN model was created and tested repeatedly to the best combination of nodes. Accuracy of the forecast between the two methods was compared by looking at the mean square error and the standard error for variable x and y. Results showed that the use of Feed Forward Back Propagation Artificial Neural Network gives better accuracy of the forecast data.</p>

scholarly journals IDENTIFICATION OF FLOODED AREAS DUE TO SEVERE STORM USING ENVISAT ASAR DATA AND NEURAL NETWORKS

2019 ◽  
Vol 10 (2) ◽  
pp. 113-120
Author(s):  
A. Abhyankar ◽  
A. Patwardhan ◽  
M. Paliwal ◽  
A. Inamdar
Keyword(s):  
Neural Network ◽  
Andhra Pradesh ◽  
Water Area ◽  
Ann Model ◽  
Cyclonic Storm ◽  
Envisat Asar ◽  
Specific Objective ◽  
Guntur District ◽  
Artificial Neural Network Ann ◽  
Flooded Areas

The specific objective of the present study is to identify flooded areas due to cyclonic storm using Envisat ASAR VV polarized data and Artificial Neural Network (ANN). On October 30, 2006, the Ogni storm crossed the Indian coast. It impacted three coastal districts in Andhra Pradesh, including Guntur, Prakasam, and Krishna. The present study considers only nine mandals of Guntur district of Andhra Pradesh for identification of flooded areas. For this purpose, pre and post event images of study area were procured of Envisat satellite (April 23, 2006 and November 4, 2006). Field visit to the affected district after the disaster was carried out to gather landcover information. In all, 564 pixels landcover information was collected during the visit (These were corresponding to pre event Envisat image of April 23, 2006). Out of the 564 pixels, randomly 406 pixels (91 were water and the remaining 315 were non-water pixels) were used for training the Neural Network and the remaining for testing. Using the trained ANN model, the total water area in the nine mandals of Guntur using Envisat ASAR satellite imagery of April 23, 2006 was found to be 2.344 thousand hectares. The trained model was applied to the post event Envisat ASAR image of November 4, 2006 to obtain completely submerged and partial/non submerged areas under water. The completely submerged landcover under water in nine mandals of Guntur district on November 4, 2006 was found to be 13.2705 thousand hectares. Results suggest a high accuracy of classification and indicate that this may be a rapid tool for damage estimation and post disaster relief and recovery efforts.

Sign in / Sign up

Export Citation Format

Share Document