• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2003.10a
• /
• pp.163-166
• /
• 2003

Hull form designers have to design a ship with satisfying an economical, technical and environmental demand. When it is concerned by a technical and environmental demand, there will be a economical demand left to criticize optimization. In this case, there were used to be requirements which needs to meet only a best performance not concerning about input of Human resource. Life cycle's cost contains building cost and operation cost so that now we need to check Man Hour cost in building a ship. This research shows a correlation between hull form information, i.e. curvature, length, breadth and thickness of surface and Man Hour of the Shell plating manufacture with using Artificial Neural Network and Gentic Programming. This study will support to classify initial work, to have a high assumption possible through predicting a Man Hour and to provide a guide book to infer a building cost and a economical optimization hull form.

• 한국신재생에너지학회:학술대회논문집
• /
• 2005.06a
• /
• pp.349-352
• /
• 2005

The energy management of an HEV using optimal control and global optimization is thought to be closest to the best operation of the system. However, there are some controversies on the ways of defining the optimization problems and constituting the optimal control simulators. Here, we presented a simulator which adopts the concept of equivalent fuel economy and leads the vehicle to run in a more efficient way. In order to realize the optimal operation of the HEV and check the validity of the control logics, we also developed a forward-facing simulator. The simulator was developed with the Cruise and MATLAB co-simulation interface. Especially, neural network controller was used for the hybrid control module in the simulator. With the simulator, the optimal operation could be converted into hybrid control rules and the validity of the operation was verified.

• Journal of Korean Society of Water and Wastewater
• /
• v.11 no.1
• /
• pp.21-32
• /
• 1997

Optimum design of water distribution network(WDN) in many times means just reducing redundancy. Given only a few situations are taken into consideration for such design, WDN deprived of inherited redundancy may not work properly in some unconsidered cases. Quantifying redundancy and incorporating it into the optimal design process will be a way of overcoming just reduction of redundancy. Expected shortage is developed as a reliability surrogate in WDN. It is an indicator of the frequency, duration and severity of failure. Using this surrogate, Expected Shortage Optimization Model (ESOM) is developed. ESOM is tested with an example network and results are analyzed and compared with those from other reliability models. The analysis results indicate that expected shortage is a quantitative surrogate measure, especially, good in comparing different designs and obtaining tradeoff between cost and. reliability. In addition, compared other models, ESOM is also proved useful in optimizing WDN with reliability and powerful in controlling reliability directly in the optimization process, even if computational burden is high. Future studies are suggested which focus on how to increase applicability and flexibility of ESOM.

• Journal of Power Electronics
• /
• v.18 no.5
• /
• pp.1409-1423
• /
• 2018

Compared with classical linear controllers, a nonlinear controller can result in better control performance for the nonlinear uncertainties of continuously variable transmission (CVT) systems that are driven by a synchronous reluctance motor (SynRM). Improved control performance can be seen in the nonlinear uncertainties behavior of CVT systems by using the proposed mingled revised recurrent Hermite polynomial neural network (MRRHPNN) control with mend ant colony optimization (ACO). The MRRHPNN control with mend ACO can carry out the overlooker control system, reformed recurrent Hermite polynomial neural network (RRHPNN) control with an adaptive law, and reimbursed control with an appraised law. Additionally, in accordance with the Lyapunov stability theorem, the adaptive law in the RRHPNN and the appraised law of the reimbursed control are established. Furthermore, to help improve convergence and to obtain better learning performance, the mend ACO is utilized for adjusting the two varied learning rates of the two parameters in the RRHPNN. Finally, comparative examples are illustrated by experimental results to confirm that the proposed control system can achieve better control performance.

• Journal of Korea Multimedia Society
• /
• v.21 no.8
• /
• pp.969-981
• /
• 2018

Recent advances in machine learning approaches such as deep neural network and reinforcement learning offer significant performance improvements in generating detailed and varied motions in physically simulated virtual environments. The optimization methods are highly attractive because it allows for less understanding of underlying physics or mechanisms even for high-dimensional subtle control problems. In this paper, we propose an efficient learning method for stochastic policy represented as deep neural networks so that agent can generate various energetic motions adaptively to the changes of tasks and states without losing interactivity and robustness. This strategy could be realized by our novel trajectory search method motivated by the trust region policy optimization method. Our value-based trajectory smoothing technique finds stably learnable trajectories without consulting neural network responses directly. This policy is set as a trust region of the artificial neural network, so that it can learn the desired motion quickly.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.24 no.8
• /
• pp.627-635
• /
• 2012

This study aims at developing an artificial neural network(ANN)-based predictive and adaptive temperature control method to control the openings at internal and external skins, and heating systems used in a building with double skin envelope. Based on the predicted indoor temperature, the control logic determined opening conditions of air inlets and outlets, and the operation of the heating systems. The optimization process of the initial ANN model was conducted to determine the optimal structure and learning methods followed by the performance tests by the comparison with the actual data measured from the existing double skin envelope. The analysis proved the prediction accuracy and the adaptability of the ANN model in terms of Root Mean Square and Mean Square Errors. The analysis results implied that the proposed ANN-based temperature control logic had potentials to be applied for the temperature control in the double skin envelope buildings.

• Journal of the Society of Naval Architects of Korea
• /
• v.33 no.2
• /
• pp.44-55
• /
• 1996

Since the use of computer for the ship structural design around mid 1960``s, specially many researches on the midship section optimum design were carried out from 1980. For a rule-based optimum design case, there has been a problem of handling a discrete design variable such as plate thickness for a practical use. To deal with the discrete design variable problems and to develop an effective new method using artificial neural network for the ship structural design applications, Neuro-Optimizer combing Hopfield Neural Network and other Simulated Annealing is proposed as a new optimization method and then applied to the fundamental skeletal structures and Midship section of Tanker. From the numerical results, it is confirmed that Neuro-Optimizer could be used effectively as a new optimization method for the structural design.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.17 no.2
• /
• pp.267-276
• /
• 1993

This paper discusses the neural network application in the study on the obstacle avoidance of robot manipulator during the trajectory planning. The collision problem of two robot manipulators which are simultaneously moving in the same workspace is investigated. Instead of the traditional modeling method, this paper processing based on the calculation of joint angle in the cartesian coordinate with constrained condition shows the possibility of real time control. The problem of the falling into the local minima is cleared by the adaptive weight factor control using the temperature adding method. Computer simulations are shown for the verification.

• Journal of Korea Technical Association of The Pulp and Paper Industry
• /
• v.37 no.4 s.112
• /
• pp.44-51
• /
• 2005

Paper mills use and discharge lots of water. And so now the papermaking industry could be classified into major water consuming industry In order to analyze the process water network and to establish the mass, water balance of duplex board mill, computer aided simulation was made using water pinch method. Based on the pinch analysis results, reuse of process water, after regenerating by microfilter as much as $140\;m^3/hr$, could be suggested without significant accumulation of contaminants in process water. According to this suggestion about $3000\;m^3/day$ of recycled process water could be sub stituted by regenerated water and consequently $30\%$ of energy cost is expected to be reduced.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.5
• /
• pp.462-466
• /
• 2008

In this paper, we propose a Data Dependant Swarm Intelligence Routing Algorithm(DSRA) based on "ant colony optimization" to improve routing performance in Mobile Ad hoc Network(MANET). DSRA generates a different routing path depending on data's characteristics: Realtime and Non-Realtime. DSRA achieves a reduced delay for Realtime data and an enhanced network lifetime from a decentralized path selection for Non-Realtime data. We demonstrate these results by an experimental study comparing with AODV, DSR and AntHocNet.