• Structural Engineering and Mechanics
• /
• v.7 no.6
• /
• pp.533-550
• /
• 1999

Optimal design of raft-pile foundations is examined by combining finite element technique and the optimization approach. The piles and soil medium are modeled by three dimensional solid elements while the raft is modelled by shell elements. Drucker-Prager criterion is adopted for the soil medium while the raft and the piles are assumed to be linear elastic. For the optimization process, the approximate semi-analytical method is used for calculating constraint sensitivities and a constraint approximation method which is a combination of the extended Bi-point approximation and Lagrangian polynomial approximation is used for predicting the behaviour of the constraints. The objective function of the problem is the volume of materials of the foundation while the design variables are raft thickness, pile length and pile spacing. The generalized reduced gradient algorithm is chosen for solving the optimization process. It is demonstrated that the method proposed in this study is promising for obtaining optimal design of raft-pile foundations without carrying out a large number of analyses. The results are also compared with those obtained from the previous study in which linear analysis was carried out.

• Computers and Concrete
• /
• v.11 no.5
• /
• pp.441-462
• /
• 2013

When using the conventional finite element method, a great number of grid nodes are necessary to describe the large and uneven temperature gradients in the concrete around cooling pipes when calculating the temperature field of mass concrete with cooling pipes. In this paper, the temperature gradient properties of the concrete around a pipe were studied. A new calculation method was developed based on these properties and an explicit iterative algorithm. With a small number of grid nodes, both the temperature distribution along the cooling pipe and the temperature field of the concrete around the water pipe can be correctly calculated with this new method. In conventional computing models, the cooling pipes are regarded as the third boundary condition when solving a model of concrete with plastic pipes, which is an approximate way. At the same time, the corresponding parameters have to be got by expensive experiments and inversion. But in the proposed method, the boundary condition is described strictly, and thus is more reliable and economical. And numerical examples were used to illustrate that this method is accurate, efficient and applicable to the actual engineering.

• Nuclear Engineering and Technology
• /
• v.51 no.4
• /
• pp.968-976
• /
• 2019

This paper presents a number of verification case studies for a recently developed sensitivity/uncertainty code package. The code package, ROMUSE (Reduced Order Modeling based Uncertainty/Sensitivity Estimator) is an effort to provide an analysis tool to be used in conjunction with reactor core simulators, in particular the Virtual Environment for Reactor Applications (VERA) core simulator. ROMUSE has been written in C++ and is currently capable of performing various types of parameter perturbations and associated sensitivity analysis, uncertainty quantification, surrogate model construction and subspace analysis. The current version 2.0 has the capability to interface with the Design Analysis Kit for Optimization and Terascale Applications (DAKOTA) code, which gives ROMUSE access to the various algorithms implemented within DAKOTA, most importantly model calibration. The verification study is performed via two basic problems and two reactor physics models. The first problem is used to verify the ROMUSE single physics gradient-based range finding algorithm capability using an abstract quadratic model. The second problem is the Brusselator problem, which is a coupled problem representative of multi-physics problems. This problem is used to test the capability of constructing surrogates via ROMUSE-DAKOTA. Finally, light water reactor pin cell and sodium-cooled fast reactor fuel assembly problems are simulated via SCALE 6.1 to test ROMUSE capability for uncertainty quantification and sensitivity analysis purposes.

• Wind and Structures
• /
• v.29 no.1
• /
• pp.65-75
• /
• 2019

A heated square cylinder (with height $A^*$) is kept parallel to the cold wall at a fixed gap height $0.5A^*$ from the wall. Another adiabatic rectangular cylinder (of same height $A^*$ and width $0.5A^*$) is placed upstream in an inline tandem arrangement. The spacing between the two cylinders is fixed at $3.0A^*$. The inlet flow is taken as Couette-Poiseuille flow based non-linear velocity profile. The conventional fluid (also known as base fluid) is chosen as water (W) whereas the nanoparticle material is selected as $Al_2O_3$. Numerical simulations are performed by using SIMPLE algorithm based Finite Volume approach with staggered grid arrangement. The dependencies of hydrodynamic and heat transfer characteristics of the cylinder on non-dimensional parameters governing the nanofluids and the fluid flow are explored here. A critical discussion is made on the mechanism of improvement/reduction (due to the presence of the upstream cylinder) of heat transfer and drag coefficient, in comparison to those of an isolated cylinder. It is observed that the heat transfer increases with the increase in the non-linearity in the incident velocity profile at the inlet. For the present range studied, particle concentration has a negligible effect on heat transfer.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.4
• /
• pp.121-128
• /
• 2021

In order to design a real time big data collection and analysis system of manufacturing data in a smart factory, it is important to establish an appropriate wired/wireless communication system and protocol. This paper introduces the latest communication protocol, OPC-UA (Open Platform Communication Unified Architecture) based client/server function, applied user interface technology to configure a network for real-time data collection through IoT Integration. Then, Database is designed in MES (Manufacturing Execution System) based on the analysis table that reflects the user's requirements among the data extracted from the new cutting process automation process, bush inner diameter indentation measurement system and tool monitoring/inspection system. In summary, big data analysis system introduced in this paper performs SPC (statistical Process Control) analysis and visualization analysis with interface of OPC-UA-based wired/wireless communication. Through AI learning modeling with XGBoost (eXtream Gradient Boosting) and LR (Linear Regression) algorithm, quality and visualization analysis is carried out the storage and connection to the cloud.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.1
• /
• pp.147-165
• /
• 2021

In this work, we empirically evaluated the efficiency of the recent EfficientNetB0 model to identify and diagnose malaria parasite infections in blood smears. The dataset used was collected and classified by relevant experts from the Lister Hill National Centre for Biomedical Communications (LHNCBC). We prepared our samples with minimal image transformations as opposed to others, as we focused more on the feature extraction capability of the EfficientNetB0 baseline model. We applied transfer learning to increase the initial feature sets and reduced the training time to train our model. We then fine-tuned it to work with our proposed layers and re-trained the entire model to learn from our prepared dataset. The highest overall accuracy attained from our evaluated results was 94.70% from fifty epochs and followed by 94.68% within just ten. Additional visualization and analysis using the Gradient-weighted Class Activation Mapping (Grad-CAM) algorithm visualized how effectively our fine-tuned EfficientNetB0 detected infections better than other recent state-of-the-art DCNN models. This study, therefore, concludes that when fine-tuned, the recent EfficientNetB0 will generate highly accurate deep learning solutions for the identification of malaria parasites in blood smears without the need for stringent pre-processing, optimization, or data augmentation of images.

• Journal of the Korea Safety Management & Science
• /
• v.23 no.4
• /
• pp.93-104
• /
• 2021

In this paper, we propose a method for diagnosing overload and working load of collaborative robots through performance analysis of machine learning algorithms. To this end, an experiment was conducted to perform pick & place operation while changing the payload weight of a cooperative robot with a payload capacity of 10 kg. In this experiment, motor torque, position, and speed data generated from the robot controller were collected, and as a result of t-test and f-test, different characteristics were found for each weight based on a payload of 10 kg. In addition, to predict overload and working load from the collected data, machine learning algorithms such as Neural Network, Decision Tree, Random Forest, and Gradient Boosting models were used for experiments. As a result of the experiment, the neural network with more than 99.6% of explanatory power showed the best performance in prediction and classification. The practical contribution of the proposed study is that it suggests a method to collect data required for analysis from the robot without attaching additional sensors to the collaborative robot and the usefulness of a machine learning algorithm for diagnosing robot overload and working load.

• International Journal of Advanced Culture Technology
• /
• v.10 no.4
• /
• pp.499-510
• /
• 2022

This study examines the reporting factors of crime against business in Korea and proposes a corresponding predictive model using machine learning. While many previous studies focused on the individual factors of theft victims, there is a lack of evidence on the reporting factors of crime against a business that serves the public good as opposed to those that protect private property. Therefore, we proposed a crime prevention model for the willingness factor of theft reporting in businesses. This study used data collected through the 2015 Commercial Crime Damage Survey conducted by the Korea Institute for Criminal Policy. It analyzed data from 834 businesses that had experienced theft during a 2016 crime investigation. The data showed a problem with unbalanced classes. To solve this problem, we jointly applied the Synthetic Minority Over Sampling Technique and the Tomek link techniques to the training data. Two prediction models were implemented. One was a statistical model using logistic regression and elastic net. The other involved a support vector machine model, tree-based machine learning models (e.g., random forest, extreme gradient boosting), and a stacking model. As a result, the features of theft price, invasion, and remedy, which are known to have significant effects on reporting theft offences, can be predicted as determinants of such offences in companies. Finally, we verified and compared the proposed predictive models using several popular metrics. Based on our evaluation of the importance of the features used in each model, we suggest a more accurate criterion for predicting var.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.1
• /
• pp.136-141
• /
• 2022

A very complex task in deep learning such as image classification must be solved with the help of neural networks and activation functions. The backpropagation algorithm advances backward from the output layer towards the input layer, the gradients often get smaller and smaller and approach zero which eventually leaves the weights of the initial or lower layers nearly unchanged, as a result, the gradient descent never converges to the optimum. We propose a two-factor non-saturating activation functions known as Bea-Mish for machine learning applications in deep neural networks. Our method uses two factors, beta (𝛽) and alpha (𝛼), to normalize the area below the boundary in the Mish activation function and we regard these elements as Bea. Bea-Mish provide a clear understanding of the behaviors and conditions governing this regularization term can lead to a more principled approach for constructing better performing activation functions. We evaluate Bea-Mish results against Mish and Swish activation functions in various models and data sets. Empirical results show that our approach (Bea-Mish) outperforms native Mish using SqueezeNet backbone with an average precision (AP50val) of 2.51% in CIFAR-10 and top-1accuracy in ResNet-50 on ImageNet-1k. shows an improvement of 1.20%.

• Structural Engineering and Mechanics
• /
• v.85 no.5
• /
• pp.607-620
• /
• 2023

The classical optimal control (COC) method has been widely used for linear quadratic regulator (LQR) problems of structural control. However, the equation of motion of the structure is incorporated into the optimization model as the constraint condition for the LQR problem, which needs to be solved through the Riccati equation under certain assumptions. In this study, an explicit optimal control (EOC) method is proposed based on the explicit time-domain method (ETDM). By use of the explicit formulation of structural responses, the LQR problem with the constraint of equation of motion can be transformed into an unconstrained optimization problem, and therefore the control law can be derived directly without solving the Riccati equation. To further optimize the weighting parameters adopted in the control law using the gradient-based optimization algorithm, the sensitivities of structural responses and control forces with respect to the weighting parameters are derived analytically based on the explicit expressions of dynamic responses of the controlled structure. Two numerical examples are investigated to demonstrate the feasibility of the EOC method and the optimization scheme for weighting parameters involved in the control law.