• International Journal of Computer Science & Network Security
• /
• v.21 no.7
• /
• pp.324-330
• /
• 2021

Agriculture is one of the essential needs of human life on planet Earth. It is the source of food and earnings for many individuals around the world. The economy of many countries is associated with the agriculture sector. Lots of diseases exist that attack various fruits and crops. Apple Tree Leaves also suffer different types of pathological conditions that affect their production. These pathological conditions include apple scab, cedar apple rust, or multiple diseases, etc. In this paper, an automatic detection framework based on deep learning is investigated for apple leaves disease classification. Different pre-trained models, VGG16, ResNetV2, InceptionV3, and MobileNetV2, are considered for transfer learning. A combination of parameters like learning rate, batch size, and optimizer is analyzed, and the best combination of ResNetV2 with Adam optimizer provided the best classification accuracy of 94%.

• International Journal of Computer Science & Network Security
• /
• v.21 no.6
• /
• pp.77-88
• /
• 2021

Heart Failure represents a critical pathological case that is challenging to predict and discover at an early age, with a notable increase in morbidity and mortality. Machine Learning and Neural Network techniques play a crucial role in predicting heart attacks, diseases and more. These techniques give valuable perspectives for clinicians who may then adjust their diagnosis for each individual patient. This paper evaluated neural network models for heart attacks predictions. Several online learning methods were investigated to automatically and accurately predict heart attacks. The UCI dataset was used in this work to train and evaluate First Order and Second Order Online Learning methods; namely Backpropagation, Delta bar Delta, Levenberg Marquardt and QuickProp learning methods. An optimizer technique was also used to minimize the random noise in the database. A regularization concept was employed to further improve the generalization of the model. Results show that a three layers' NN model with a Backpropagation algorithm and Nadam optimizer achieved a promising accuracy for the heart attach prediction tasks.

• Earthquakes and Structures
• /
• v.22 no.4
• /
• pp.401-419
• /
• 2022

A novel hybrid extreme machine learning-multiverse optimizer (ELM-MVO) framework is proposed to predict the liquefaction phenomenon in seismically excited tunnel lining inside the sand lens. The MVO is applied to optimize the input weights and biases of the ELM algorithm to improve its efficiency. The tunnel located inside the liquefied sand lens is also evaluated under various near- and far-field earthquakes. The results demonstrate the superiority of the proposed method to predict the liquefaction event against the conventional extreme machine learning (ELM) and artificial neural network (ANN) algorithms. The outcomes also indicate that the possibility of liquefaction in sand lenses under far-field seismic excitations is much less than the near-field excitations, even with a small magnitude. Hence, tunnels designed in geographical areas where seismic excitations are more likely to be generated in the near area should be specially prepared. The sand lens around the tunnel also has larger settlements due to liquefaction.

• International journal of advanced smart convergence
• /
• v.13 no.2
• /
• pp.80-87
• /
• 2024

This paper focuses on improving accuracy in constrained computing settings by employing the ReLU (Rectified Linear Unit) activation function. The research conducted involves modifying parameters of the ReLU function and comparing performance in terms of accuracy and computational time. This paper specifically focuses on optimizing ReLU in the context of a Multilayer Perceptron (MLP) by determining the ideal values for features such as the dimensions of the linear layers and the learning rate (Ir). In order to optimize performance, the paper experiments with adjusting parameters like the size dimensions of linear layers and Ir values to induce the best performance outcomes. The experimental results show that using ReLU alone yielded the highest accuracy of 96.7% when the dimension sizes were 30 - 10 and the Ir value was 1. When combining ReLU with the Adam optimizer, the optimal model configuration had dimension sizes of 60 - 40 - 10, and an Ir value of 0.001, which resulted in the highest accuracy of 97.07%.

• Proceedings of the KIEE Conference
• /
• 2006.04a
• /
• pp.294-296
• /
• 2006

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.3
• /
• pp.539-546
• /
• 2018

Recently, as the solar power generation market is rapidly increasing, interest is focused on research for minimizing the output of the solar cell module. The role of the power optimizer is important when inconsistencies occur in photovoltaic power generation. In the conventional system, centralized inverter method and microinverter method are mainly used. In this paper, we analyze the problem of power generation efficiency loss due to the incompatibility of existing system configuration methods. We also proposed a module - type power compensation method that can improve the mismatch caused by shading. The proposed module - based power optimizer is implemented and compared with the existing operation method. From the simulation result, it was confirmed that the efficiency of the proposed operation method is improved compared to the existing method.

• Composites Research
• /
• v.30 no.1
• /
• pp.52-58
• /
• 2017

This paper presents the optimum design methodology for composite wing structure which automatically calculates the safety margin using optimization framework integrating failure modes. Particularly, its framework is possible to optimize sizing procedure to prevent failure mode which has the greatest effect on reducing the sizing time of composite structure. The main failure mode was set as the first ply failure, buckling failure mode, and bolted joint stress field, and the margin was calculated to minimize the weight. The design variable is a laminate sequence database and the responses are strain, buckling, bolted joint stress field. The objective function is the mass of the wing structure. The results of buckling analysis were compared using the finite element model to verify the robustness and reliability of Composite Optimizer.

• Journal of Korean Society of Steel Construction
• /
• v.10 no.3 s.36
• /
• pp.487-495
• /
• 1998

Recently, the space truss has been attracted by many designers because of their ability to support significant loads with a minimum material. And it is relatively flexible to design the configuration of structures. This paper presents a volume optimization for the space truss on the basis of result evaluated from nonlinear analysis. The optimization of the truss is done by nonlinear optimum GINO(General Interactive Nonlinear Optimizer) program. The objective function considered is the volume of the steel bars. The constraints for optimum design are the design limits, such as the axial force strength, maximum slenderness, minimum thickness, allowable deflection and ratio of the external diameter to thickness of the circular tube bars.

• Wind and Structures
• /
• v.18 no.2
• /
• pp.195-213
• /
• 2014

Recently, the horizontal axis rotor performance optimizer (HARP_Opt) tool was developed in the National Renewable Energy Laboratory, USA. This innovative tool is becoming more popular in the wind turbine industry and in the field of academic research. HARP_Optwas developed on the basis of two fundamental modules, namely, WT_Perf, a performance evaluator computer code using the blade element momentum theory; and a genetic algorithm module, which is used as an optimizer. A pattern search algorithm was more recently incorporated to enhance the optimization capability, especially the calculation time and consistency of the solutions. The blade optimization is an aspect that is highly dependent on experience and requires significant consideration on rotor control strategies, wind data, and generator type. In this study, the effects of rotor control strategies including fixed speed and fixed pitch, variable speed and fixed pitch, fixed speed and variable pitch, and variable speed and variable pitch algorithms on optimal blade shapes and rotor performance are investigated using optimized blade designs. The effects of environmental wind data and the objective functions used for optimization are also quantitatively evaluated using the HARP_Opt tool. Performance indices such as annual energy production, thrust, torque, and roof-flap moment forces are compared.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.24 no.12
• /
• pp.3040-3052
• /
• 2000

For effective construction of second-order response surface models, an efficient quad ratic approximation method is proposed in the context of trust region model management strategy. In the proposed method, although only the linear and quadratic terms are uniquely determined using 2n+1 design points, the two-factor interaction terms are mathematically updated by normalized quasi-Newton formula. In order to show the numerical performance of the proposed approximation method, a sequential approximate optimizer is developed and solves a typical unconstrained optimization problem having 2, 6, 10, 15, 30 and 50 design variables, a gear reducer system design problem and two dynamic response optimization problems with multiple objectives, five objectives for one and two objectives for the other. Finally, their optimization results are compared with those of the CCD or the 50% over-determined D-optimal design combined with the same trust region sequential approximate optimizer. These comparisons show that the proposed method gives more efficient than others.