scholarly journals Evaluation of Neural Network Model for Estimating Pile Load Capacity

2021 ◽  
Vol 21 (5) ◽  
pp. 221-228
Author(s):  
Byungsik Lee
Keyword(s):  
Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Load Capacity ◽  
Absolute Error ◽  
Descent Method ◽  
Training Data ◽  
Gradient Descent Method ◽  
Neural Network Models ◽  
Data Set

Neural network models based on deep learning algorithms are increasingly used for estimating pile load capacities as supplements of bearing capacity equations and field load tests. A series of hyperparameter tuning is required to improve the performance and reliability of developing a neural network model. In this study, the number of hidden layers and neurons, the activation functions, the optimizing algorithms of the gradient descent method, and the learning rates were tuned. The grid search method was applied for the tuning, which is a hyperpameter optimizer supplied by the developing platform. The cross-validation method was applied to enhance reliability for model validation. An appropriate number of epochs was determined using the early stopping method to prevent the overfitting of the model to the training data. The performance of the tuned optimum model evaluated for the test data set revealed that the model could estimate pile load capacities approximately with an average absolute error of 3,000 kN and a coefficient of determinant of 0.5.

scholarly journals RISC-V Virtual Platform-Based Convolutional Neural Network Accelerator Implemented in SystemC

Electronics ◽  
2021 ◽  
Vol 10 (13) ◽  
pp. 1514
Author(s):  
Seung-Ho Lim ◽  
WoonSik William Suh ◽  
Jin-Young Kim ◽  
Sang-Young Cho
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Network Model ◽  
Neural Network Model ◽  
Deep Neural Network ◽  
System Level ◽  
Neural Network Models ◽  
Data Set ◽  
Embedded Device ◽  
Virtual Platform

The optimization for hardware processor and system for performing deep learning operations such as Convolutional Neural Networks (CNN) in resource limited embedded devices are recent active research area. In order to perform an optimized deep neural network model using the limited computational unit and memory of an embedded device, it is necessary to quickly apply various configurations of hardware modules to various deep neural network models and find the optimal combination. The Electronic System Level (ESL) Simulator based on SystemC is very useful for rapid hardware modeling and verification. In this paper, we designed and implemented a Deep Learning Accelerator (DLA) that performs Deep Neural Network (DNN) operation based on the RISC-V Virtual Platform implemented in SystemC in order to enable rapid and diverse analysis of deep learning operations in an embedded device based on the RISC-V processor, which is a recently emerging embedded processor. The developed RISC-V based DLA prototype can analyze the hardware requirements according to the CNN data set through the configuration of the CNN DLA architecture, and it is possible to run RISC-V compiled software on the platform, can perform a real neural network model like Darknet. We performed the Darknet CNN model on the developed DLA prototype, and confirmed that computational overhead and inference errors can be analyzed with the DLA prototype developed by analyzing the DLA architecture for various data sets.

An explicit methodology for manufacturing cost–tolerance modeling and optimization using the neural network integrated with the genetic algorithm

2020 ◽  
Vol 34 (3) ◽  
pp. 430-443
Author(s):  
A. Saravanan ◽  
J. Jerald ◽  
A. Delphin Carolina Rani
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Network Model ◽  
Neural Network Model ◽  
Network Models ◽  
Manufacturing Cost ◽  
Genetic Operators ◽  
Neural Network Models ◽  
Data Set ◽  
The Neural Network

AbstractThe objective of the paper is to develop a new method to model the manufacturing cost–tolerance and to optimize the tolerance values along with its manufacturing cost. A cost–tolerance relation has a complex nonlinear correlation among them. The property of a neural network makes it possible to model the complex correlation, and the genetic algorithm (GA) is integrated with the best neural network model to optimize the tolerance values. The proposed method used three types of neural network models (multilayer perceptron, backpropagation network, and radial basis function). These network models were developed separately for prismatic and rotational parts. For the construction of network models, part size and tolerance values were used as input neurons. The reference manufacturing cost was assigned as the output neuron. The qualitative production data set was gathered in a workshop and partitioned into three files for training, testing, and validation, respectively. The architecture of the network model was identified based on the best regression coefficient and the root-mean-square-error value. The best network model was integrated into the GA, and the role of genetic operators was also studied. Finally, two case studies from the literature were demonstrated in order to validate the proposed method. A new methodology based on the neural network model enables the design and process planning engineers to propose an intelligent decision irrespective of their experience.

scholarly journals On the Determination of Groundwater Level using Temporal and Spatial Parameters: Advanced Machine Learning Methods

2021 ◽  
Author(s):  
Amirhossein Najafabadipour ◽  
Gholamreza Kamali ◽  
Hossein Nezamabadi-pour
Keyword(s):  
Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Groundwater Level ◽  
Network Models ◽  
Machine Intelligence ◽  
Neural Network Models ◽  
Validation Data ◽  
Data Set ◽  
Spatial Parameters

Abstract Prediction of groundwater level is a useful tool for managing groundwater resources in the mining area. Water resources management requires identifying potential periods for groundwater drainage to prevent groundwater from entering the mine pit and reduce high costs. For this purpose, four multilayer perceptron (MLP) neural network models and four cascade forward (CF) neural network models optimized with Bayesian Regularization (BR), Levenberg-Marquardt (LM), Resilient Backpropagation (RB), and Scaled Conjugate Gradient (SCG), as well as a radial basis function (RBF) neural network model and a generalized regression (GR) neural network model were developed to predict groundwater level using 1377 data point. This data set includes 12 spatial parameters divided into two categories of sediments and bedrock, and besides, 6 time series parameters have been used. Also, to determine the best models and combine them, 165 extra validation data points have been used. After identifying the best models from the three candidate models with lower average absolute relative error (AARE) value, the committee machine intelligence system (CMIS) model has been developed. The proposed CMIS model predicts groundwater level data with high accuracy with an AARE value of less than 0.11%. Also, the proposed model was compared with ten other models through graphical and statistical error analysis. The results show that the developed CMIS model performs better than other existing models in terms of precision and validity range. The relevancy factor indicates that the electrical resistivity of sediments had the highest effect on the groundwater level. Eventually, the quality of the data used was investigated both statistically and graphically, and the results show satisfactory reliability of the data used.

scholarly journals Initial Results of Modeling and Improvement of BDS-2/GPS Broadcast Ephemeris Satellite Orbit Based on BP and PSO-BP Neural Networks

2021 ◽  
Vol 13 (23) ◽  
pp. 4801
Author(s):  
Hanlin Chen ◽  
Fei Niu ◽  
Xing Su ◽  
Tao Geng ◽  
Zhimin Liu ◽  
...  
Keyword(s):  
Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Bp Neural Network ◽  
Network Models ◽  
Neural Network Models ◽  
Orbit Error ◽  
Broadcast Ephemeris ◽  
Bp Neural Network Model ◽  
Orbit Errors

With the rapid development and gradual perfection of GNSS in recent years, improving the real-time service performance of GNSS has become a research hotspot. In GNSS single-point positioning, broadcast ephemeris is used to provide a space–time reference. However, the orbit parameters of broadcast ephemeris have meter-level errors, and no mathematical model can simulate the variation of this, which restricts the real-time positioning accuracy of GNSS. Based on this research background, this paper uses a BP (Back Propagation) neural network and a PSO (Particle Swarm Optimization)–BP neural network to model the variation in the orbit error of GPS and BDS broadcast ephemeris to improve the accuracy of broadcast ephemeris. The experimental results showed that the two neural network models in GPS can model the broadcast ephemeris orbit errors, and the results of the two models were roughly the same. The one-day and three-day improvement rates of RMS(3D) were 30–50%, but the PSO–BP neural network model was better able to model the trend of errors and effectively improve the broadcast ephemeris orbit accuracy. In BDS, both of the neural network models were able to model the broadcast ephemeris orbit errors; however, the PSO–BP neural network model results were better than those of the BP neural network. In the GEO satellite outcome of the PSO–BP neural network, the STD and RMS of the orbit error in three directions were reduced by 20–70%, with a 20–30% improvement over the BP neural network results. The IGSO satellite results showed that the PSO–BP neural network model output accuracy of the along- and radial-track directions experienced a 70–80% improvement in one and three days. The one- and three-day RMS(3D) of the MEO satellites showed that the PSO–BP neural network has a greater ability to resist gross errors than that of the BP neural network for modeling the changing trend of the broadcast ephemeris orbit errors. These results demonstrate that using neural networks to model the orbit error of broadcast ephemeris is of great significance to improving the orbit accuracy of broadcast ephemeris.

A COMPARISON OF NEURAL NETWORK MODEL SELECTION STRATEGIES FOR THE PRICING OF S&P 500 STOCK INDEX OPTIONS

2007 ◽  
Vol 16 (06) ◽  
pp. 1093-1113 ◽  
Author(s):  
N. S. THOMAIDIS ◽  
V. S. TZASTOUDIS ◽  
G. D. DOUNIAS
Keyword(s):  
Neural Network ◽  
Model Selection ◽  
Network Model ◽  
Neural Network Model ◽  
Information Criteria ◽  
Stock Index ◽  
Statistical Hypothesis ◽  
Neural Network Models ◽  
Index Options ◽  
Pruning Technique

This paper compares a number of neural network model selection approaches on the basis of pricing S&P 500 stock index options. For the choice of the optimal architecture of the neural network, we experiment with a “top-down” pruning technique as well as two “bottom-up” strategies that start with simple models and gradually complicate the architecture if data indicate so. We adopt methods that base model selection on statistical hypothesis testing and information criteria and we compare their performance to a simple heuristic pruning technique. In the first set of experiments, neural network models are employed to fit the entire options surface and in the second they are used as parts of a hybrid intelligence scheme that combines a neural network model with theoretical option-pricing hints.

scholarly journals Exploiting the Relationship between Pruning Ratio and Compression Effect for Neural Network Model Based on TensorFlow

2020 ◽  
Vol 2020 ◽  
pp. 1-8
Author(s):  
Bo Liu ◽  
Qilin Wu ◽  
Yiwen Zhang ◽  
Qian Cao
Keyword(s):  
Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Prediction Accuracy ◽  
Computing Time ◽  
Experimental Results ◽  
Data Set ◽  
Compression Effect ◽  
Model Based ◽  
The Relationship

Pruning is a method of compressing the size of a neural network model, which affects the accuracy and computing time when the model makes a prediction. In this paper, the hypothesis that the pruning proportion is positively correlated with the compression scale of the model but not with the prediction accuracy and calculation time is put forward. For testing the hypothesis, a group of experiments are designed, and MNIST is used as the data set to train a neural network model based on TensorFlow. Based on this model, pruning experiments are carried out to investigate the relationship between pruning proportion and compression effect. For comparison, six different pruning proportions are set, and the experimental results confirm the above hypothesis.

scholarly journals PV Power Prediction, Using CNN-LSTM Hybrid Neural Network Model. Case of Study: Temixco-Morelos, México

Energies ◽  
2020 ◽  
Vol 13 (24) ◽  
pp. 6512
Author(s):  
Mario Tovar ◽  
Miguel Robles ◽  
Felipe Rashid
Keyword(s):  
Neural Network ◽  
Hybrid Model ◽  
Network Model ◽  
Neural Network Model ◽  
Short Term Memory ◽  
Lasso Regression ◽  
Neural Network Models ◽  
Model Case ◽  
Hybrid Neural Network ◽  
Photovoltaic Power

Due to the intermittent nature of solar energy, accurate photovoltaic power predictions are very important for energy integration into existing energy systems. The evolution of deep learning has also opened the possibility to apply neural network models to predict time series, achieving excellent results. In this paper, a five layer CNN-LSTM model is proposed for photovoltaic power predictions using real data from a location in Temixco, Morelos in Mexico. In the proposed hybrid model, the convolutional layer acts like a filter, extracting local features of the data; then the temporal features are extracted by the long short-term memory network. Finally, the performance of the hybrid model with five layers is compared with a single model (a single LSTM), a CNN-LSTM hybrid model with two layers and two well known popular benchmarks. The results also shows that the hybrid neural network model has better prediction effect than the two layer hybrid model, the single prediction model, the Lasso regression or the Ridge regression.

Prediction Model of Micum Strength for Iron Ore Sinter

2011 ◽  
Vol 402 ◽  
pp. 476-479
Author(s):  
Wei Wang ◽  
Zhi Hui Xu ◽  
Long Long Yang ◽  
Zheng Liang Xue ◽  
Dong Nan Zhao ◽  
...  
Keyword(s):  
Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Absolute Error ◽  
Maximum Absolute Error ◽  
Important Indicator ◽  
The Neural Network ◽  
Iron Ore Sinter ◽  
Real Value ◽  
The Law

Micum strength is an important indicator of quality of sinter; BP artificial neural network model is built to predict the strength of sinter drum. The neural network use the main factors that influence the sinter drum as input data, and output is Micum strength. Experiment results shows that the maximum absolute error between the Micum strength predicted by neural network and real value from the sinter plant is 0.3346, and the average absolute error is 0.1154. These prove that the prediction is accuracy. In addition, because of the "black box" characteristic of the neural network model, the neural network model can not give the law of how the various factors affect the micum strength of sinter ore, this paper also uses the model to analysis the law of how TFe, SiO2 content affect the micum strength. The results not only consist with the sintering theory, but also verify the validity of the model.

scholarly journals The Performance Study on the Long-Span Bridge Involving the Wireless Sensor Network Technology in a Big Data Environment

Complexity ◽  
2018 ◽  
Vol 2018 ◽  
pp. 1-13
Author(s):  
Liwen Zhang ◽  
Chao Zhang ◽  
Zhuo Sun ◽  
You Dong ◽  
Pu Wei
Keyword(s):  
Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Bp Neural Network ◽  
Network Models ◽  
Neural Network Models ◽  
Long Span ◽  
Long Span Bridge ◽  
Vibration Responses ◽  
Bp Neural Network Model

The random traffic flow model which considers parameters of all the vehicles passing through the bridge, including arrival time, vehicle speed, vehicle type, vehicle weight, and horizontal position as well as the bridge deck roughness, is input into the vehicle-bridge coupling vibration program. In this way, vehicle-bridge coupling vibration responses with considering the random traffic flow can be numerically simulated. Experimental test is used to validate the numerical simulation, and they had the consistent changing trends. This result proves the reliability of the vehicle-bridge coupling model in this paper. However, the computational process of this method is complicated and proposes high requirements for computer performance and resources. Therefore, this paper considers using a more advanced intelligent method to predict vibration responses of the long-span bridge. The PSO-BP (particle swarm optimization-back propagation) neural network model is proposed to predict vibration responses of the long-span bridge. Predicted values and real values at each point basically have the consistent changing trends, and the maximum error is less than 10%. Hence, it is feasible to predict vibration responses of the long-span bridge using the PSO-BP neural network model. In order to verify advantages of the predicting model, it is compared with the BP neural network model and GA-BP neural network model. The PSO-BP neural network model converges to the set critical error after it is iterated to the 226th generation, while the other two neural network models are not converged. In addition, the relative error of predicted values using PSO-BP neural network is only 2.71%, which is obviously less than the predicted results of other two neural network models. We can find that the PSO-BP neural network model proposed by the paper in predicting vibration responses is highly efficient and accurate.

Sign in / Sign up

Export Citation Format

Share Document