• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.7
• /
• pp.943-948
• /
• 2018

In this paper, we propose a hybrid approach of combining $A^*$ and Genetic algorithm in the path search problem. In $A^*$, the cost from a start node to the intermediate node is optimized in principle but the path from that intermediate node to the goal node is generated and tested based on the cumulated cost and the next node in a priority queue is chosen to be tested. In that process, we adopt the genetic algorithm principle in that the group of nodes to generate the next node from an intermediate node is tested by its fitness function. Top two nodes are selected to use crossover or mutation operation to generate the next generation. If generated nodes are qualified, those nodes are inserted to the priority queue. The proposed method is compared with the original sequential selection and the random selection of the next searching path in $A^*$ algorithm and the result verifies the superiority of the proposed method.

• Journal of Korean Society of Steel Construction
• /
• v.10 no.1 s.34
• /
• pp.125-135
• /
• 1998

This study presents a design method for three-span continuous preflex composite girder bridges (3S-PCB) which imposes prestresses in the negative moment region by lifting or lowering interior supports and the design method is automated by a computer program which incorporates optimal design procedure. The objective function for the design of 3S-PCB minimizes the cost of construction materials and the constraint functions represent the limited dimensions of the design section and the allowable stress for each structural member as given in the specifications. Optimal design procedure used in this study is a modification of existing sequential unconstrained minimization technique (SUMT), a numerical analyses procedure for two-span continuous preflex composite bridges. The optimized design sections determined for each span length are compared with those of simple preflex composite beams (SPCB) and the optimal girder depth is determined by defining the relationship between girder depth and construction material costs.

• Journal of the Korean Society for Precision Engineering
• /
• v.28 no.12
• /
• pp.1388-1396
• /
• 2011

This paper is focused on the procedure of the development of a micro air vehicle which has vertical take-off and landing capability for indoor reconnaissance mission. Trade studies on mission feasibility led to the proposal of a coaxial rotorcraft configuration as the platform. The survey to provide a guide for preliminary design were conducted based on commercial off-the-shelf platform, and the rotor performance was estimated by the simple momentum theory. To determine the initial size of the micro air vehicle, the modified conventional fuel balance method was applied to adopt for electric powered vehicle, and the sizing problem was optimized with the sequential quadratic programming method using MATLAB. The designed rotor blades were fabricated with high strength carbon composite material and integrated with the platform. The developed coaxial rotorcraft micro air vehicle shows stable handling quality with manual flight test in indoor situation.

• Geomechanics and Engineering
• /
• v.35 no.2
• /
• pp.121-133
• /
• 2023

The demand for cement and limestone crushed materials has increased many folds due to the tremendous increase in construction activities in Pakistan during the past few decades. The number of cement production industries has increased correspondingly, and so the rock-blasting operations at the limestone quarry sites. However, the safety procedures warranted at these sites for the blast-induced ground vibrations (BIGV) have not been adequately developed and/or implemented. Proper prediction and monitoring of BIGV are necessary to ensure the safety of structures in the vicinity of these quarry sites. In this paper, an attempt has been made to predict BIGV using artificial neural network (ANN) at three selected limestone quarries of Pakistan. The ANN has been developed in Python using Keras with sequential model and dense layers. The hyper parameters and neurons in each of the activation layers has been optimized using randomized and grid search method. The input parameters for the model include distance, a maximum charge per delay (MCPD), depth of hole, burden, spacing, and number of blast holes, whereas, peak particle velocity (PPV) is taken as the only output parameter. A total of 110 blast vibrations datasets were recorded from three different limestone quarries. The dataset has been divided into 85% for neural network training, and 15% for testing of the network. A five-layer ANN is trained with Rectified Linear Unit (ReLU) activation function, Adam optimization algorithm with a learning rate of 0.001, and batch size of 32 with the topology of 6-32-32-256-1. The blast datasets were utilized to compare the performance of ANN, multivariate regression analysis (MVRA), and empirical predictors. The performance was evaluated using the coefficient of determination (R²), mean absolute error (MAE), mean squared error (MSE), mean absolute percentage error (MAPE), and root mean squared error (RMSE)for predicted and measured PPV. To determine the relative influence of each parameter on the PPV, sensitivity analyses were performed for all input parameters. The analyses reveal that ANN performs superior than MVRA and other empirical predictors, andthat83% PPV is affected by distance and MCPD while hole depth, number of blast holes, burden and spacing contribute for the remaining 17%. This research provides valuable insights into improving safety measures and ensuring the structural integrity of buildings near limestone quarry sites.

• Journal of Biosystems Engineering
• /
• v.35 no.4
• /
• pp.224-230
• /
• 2010

The purpose of this study was to develop position control system of an electric actuator for automatic shuttle shifting of a tractor. The electric actuator was installed at the link of the forward-reverse gearshift of the tractor transmission, and controlled in the ranges of forward, neutral, and reverse positions. The position control system of the electric actuator was developed based on PID (Proportional Integral Derivative) controller and transfer function of the electric actuator. The coefficients of the PID controller were determined by Ziegler-Nichols (Z-N) method and optimized using simulation program. The prototype AMT (Automated Manual Transmission) test unit of the tractor was installed and used to evaluate the performance of the position control. The evaluation system for the control performance consisted of forward-reverse actuator, motor driver, and controller. The tests were conducted as the controlled positions of the actuator were changed from neutral position to forward, neutral, and reverse positions in sequence. The sequential tests were repeated 20 times. The operations of changing the gearshift were considered as the step response of the control system. Maximum overshoot, settling time, and steady-state error were analyzed. The results showed that performance of the position control system was reasonable and qualified. The maximum overshoots, the steady-state errors, and the settling times of the position control system were 10~20%, 1~5%, and 0.92~1.49 sec, respectively. The modifications of the electric actuator will be required to enhance the performance of position control during field operation.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.14 no.1
• /
• pp.79-91
• /
• 2012

A curved fiber reinforced polymer (FRP) panel is produced with a certain width depending on allowances of manufacturing processes and facilities. An targeted arch-shaped structure could be built by sequential connection of series of the FRP panels. The connection manner between the FRP panels could be given by chemical treatment, mechanical treatment and hybrid method. Among those, the connection between the panels by chemical treatment is commonly adopted. Therefore, For an optimized design of the connected part between FRP pannels, a number of direct shear tests have been undertaken in terms of a number of parameters: surface treatment conditions, bonding materials, etc.. As results, surface grinding condition by sand paper or surface treatment by sand blasting appear properly acceptable methods, and epoxy and acryl resins are shown to be effective bonding materials for the purpose in this study.

• Proceedings of the Korean Radioactive Waste Society Conference
• /
• 2003.11a
• /
• pp.54-59
• /
• 2003

Several radionuclides are considered as an object of the assesment to develop a scaling factor and a periodical verification method which are needed for the evaluation of radionuclide inventory of various radioactive wastes from nuclear power plants in Korea. A selective separation of $^{55}Fe$, $^{90}FSr$ 및 $^{94}Nb$ which should be recovered individually for the radiochemical analysis was described in detail. Sorption and desorption behaviours of ion exchange and extraction chromatographic resins for Fe, Sr, Nb and co-existing metal ions were Investigated using an artificial waste solution simulated of chemical composition of real radioactive wastes. Separation conditions available for the sequential recovery of these metal ions from a single sample were optimized to minimize a discharge of radioactive wastes produced through the analytical process and a radiation exposure to analysts. Their recovery yields were measured with reliability.

• Journal of Convergence for Information Technology
• /
• v.11 no.7
• /
• pp.31-38
• /
• 2021

In this paper, in order to obtain the optimization of the RNN model used for sentiment analysis, the correlation of each model was studied by observing the trend of loss and accuracy according to hyperparameter tuning. As a research method, after configuring the hidden layer with LSTM and the embedding layer that are most optimized to process sequential data, the loss and accuracy of each model were measured by tuning the unit, batch-size, and embedding size of the LSTM. As a result of the measurement, the loss was 41.9% and the accuracy was 11.4%, and the trend of the optimization model showed a consistently stable graph, confirming that the tuning of the hyperparameter had a profound effect on the model. In addition, it was confirmed that the decision of the embedding size among the three hyperparameters had the greatest influence on the model. In the future, this research will be continued, and research on an algorithm that allows the model to directly find the optimal hyperparameter will continue.

• Nuclear Engineering and Technology
• /
• v.53 no.1
• /
• pp.148-163
• /
• 2021

Timely fault identification is important for safe and reliable operation of the electric valve system. Many research works have utilized different data-driven approach for fault diagnosis in complex systems. However, they do not consider specific characteristics of critical control components such as electric valves. This work presents an integrated shallow-deep fault diagnostic model, developed based on signals extracted from DN50 electric valve. First, the local optimal issue of particle swarm optimization algorithm is solved by optimizing the weight search capability, the particle speed, and position update strategy. Then, to develop a shallow diagnostic model, the modified particle swarm algorithm is combined with support vector machine to form a hybrid improved particle swarm-support vector machine (IPs-SVM). To decouple the influence of the background noise, the wavelet packet transform method is used to reconstruct the vibration signal. Thereafter, the IPs-SVM is used to classify phase imbalance and damaged valve faults, and the performance was evaluated against other models developed using the conventional SVM and particle swarm optimized SVM. Secondly, three different deep belief network (DBN) models are developed, using different acoustic signal structures: raw signal, wavelet transformed signal and time-series (sequential) signal. The models are developed to estimate internal leakage sizes in the electric valve. The predictive performance of the DBN and the evaluation results of the proposed IPs-SVM are also presented in this paper.

• The Journal of Korean Society for Radiation Therapy
• /
• v.26 no.2
• /
• pp.207-216
• /
• 2014

Purpose : We present a method to reduce this gap and complete the treatment plan, to be made by the re-optimization is performed in the same conditions as the initial treatment plan different from Monaco treatment planning system. Materials and Methods : The optimization is carried in two steps when performing the inverse calculation for volumetric modulated radiation therapy or intensity modulated radiation therapy in Monaco treatment planning system. This study was the first plan with a complete optimization in two steps by performing all of the treatment plan, without changing the optimized condition from Step 1 to Step 2, a typical sequential optimization performed. At this time, the experiment was carried out with a pencil beam and Monte Carlo algorithm is applied In step 2. We compared initial plan and re-optimized plan with the same optimized conditions. And then evaluated the planning dose by measurement. When performing a re-optimization for the initial treatment plan, the second plan applied the step optimization. Results : When the common optimization again carried out in the same conditions in the initial treatment plan was completed, the result is not the same. From a comparison of the treatment planning system, similar to the dose-volume the histogram showed a similar trend, but exhibit different values that do not satisfy the conditions best optimized dose, dose homogeneity and dose limits. Also showed more than 20% different in comparison dosimetry. If different dose algorithms, this measure is not the same out. Conclusion : The process of performing a number of trial and error, and you get to the ultimate goal of treatment planning optimization process. If carried out to optimize the completion of the initial trust only the treatment plan, we could be made of another treatment plan. The similar treatment plan could not satisfy to optimization results. When you perform re-optimization process, you will need to apply the step optimized conditions, making sure the dose distribution through the optimization process.