• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.421-426
• /
• 2005

Fuzzy logic, neural network, fuzzy-neural network play an important as the key technology of linguistic modeling for intelligent control and decision making in complex systems. The fuzzy-neural network (FNN) learning represents one of the most effective algorithms to build such linguistic models. This paper proposes particle swarm optimization algorithm based optimal learning fuzzy-neural network (PSOA-FNN). The proposed learning scheme is the fuzzy-neural network structure which can handle linguistic knowledge as tuning membership function of fuzzy logic by particle swarm optimization algorithm. The learning algorithm of the PSOA-FNN is composed of two phases. The first phase is to find the initial membership functions of the fuzzy neural network model. In the second phase, particle swarm optimization algorithm is used for tuning of membership functions of the proposed model.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.1716-1722
• /
• 2005

Fuzzy logic, neural network, fuzzy-neural network play an important as the key technology of linguistic modeling for intelligent control and decision in complex systems. The fuzzy-neural network (FNN) learning represents one of the most effective algorithms to build such linguistic models. This paper proposes bacteria foraging algorithm based optimal learning fuzzy-neural network (BA-FNN). The proposed learning scheme is the fuzzy-neural network structure which can handle linguistic knowledge as tuning membership function of fuzzy logic by bacteria foraging algorithm. The learning algorithm of the BA-FNN is composed of two phases. The first phase is to find the initial membership functions of the fuzzy neural network model. In the second phase, bacteria foraging algorithm is used for tuning of membership functions of the proposed model.

• 대한기계학회논문집A
• /
• 제29권4호
• /
• pp.591-599
• /
• 2005

This paper proposes a systematic design methodology for high-speed/high-precision servomechanisms by using a disturbance observer. A multiplicative uncertainty model and a two degree-of-freedom controller composed of a disturbance observer (DOB) and a PD controller are considered as subsystems. Analysis of the system performance, such as internal stability and bandwidth of a servomechanism according to subsystem parameters is conducted for better understanding of the dynamic behavior and interactions among the subsystem parameters. Then, an integrated design methodology, where the interactions are considered simultaneously, is applied to design processes of the servomechanism. The tradeoff relationship between disturbance suppression and measurement noise rejection of the DOB is considered through the design process. Numerical case studies show the improved possibility to evaluate and optimize the dynamic motion performance of the servomechanism. Moreover, the disturbance observer designed based on the proposed design methodology yields excellent disturbance suppression performance.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2006년도 한국컴퓨터종합학술대회 논문집 Vol.33 No.1 (A)
• /
• pp.82-84
• /
• 2006

최근의 생물학적인 연구에 DNA microarray가 널리 쓰이고 있기 때문에, 이러한 DNA microarray를 구성하는데 필요한 probe design 작업의 중요성이 점차 커져가고 있다. 이 논문에서는 probe design 문제를 thermodynamic fitness function이 2개인 multi-objective optimization 작업으로 변환한 뒤, ${\epsilon}$-multiobjective evolutionary algorithm을 이용하여 probe set을 찾는다. 또한, probe 탐색공간의 크기를 줄이기 위하여 각 DNA sequence의 primer 영역을 찾는 작업을 진행하며, 사용자가 직접 프로그램을 테스트할 수 있는 웹사이트를 제공한다. 실험 대상으로는 mycoides를 선택하였으며, 이 논문에서 제안된 방법을 사용하여 성공적으로 probe set을 발견할 수 있었다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2003년도 하계학술대회 논문집 B
• /
• pp.855-857
• /
• 2003

A slotless Permanent Magnet Linear Synchronous Motor (PMLSM) has good control ability but thrust density is low. So, this paper proposes inserted core type of slotless PMLSM to improve its low thrust density. Inserting the core between windings of each phase, detent force is generated by the difference of magnetic resistance in an air gap. To minimize detent force and maxize thrust, this paper applies the neural network to inserted core type of slotless PMLSM.

• ETRI Journal
• /
• 제44권2호
• /
• pp.194-207
• /
• 2022

In the last decade, substantial progress has been achieved in intelligent traffic control technologies to overcome consistent difficulties of traffic congestion and its adverse effect on smart cities. Edge computing is one such advanced progress facilitating real-time data transmission among vehicles and roadside units to mitigate congestion. An edge computing-based deep reinforcement learning system is demonstrated in this study that appropriately designs a multiobjective reward function for optimizing different objectives. The system seeks to overcome the challenge of evaluating actions with a simple numerical reward. The selection of reward functions has a significant impact on agents' ability to acquire the ideal behavior for managing multiple traffic signals in a large-scale road network. To ascertain effective reward functions, the agent is trained withusing the proximal policy optimization method in several deep neural network models, including the state-of-the-art transformer network. The system is verified using both hypothetical scenarios and real-world traffic maps. The comprehensive simulation outcomes demonstrate the potency of the suggested reward functions.

• 전기학회논문지
• /
• 제62권5호
• /
• pp.591-595
• /
• 2013

This paper presents a method of generation rescheduling using Newton's Approach which searches the solution of the Lagrangian function. The generation fuel cost and $CO_2$ emission cost functions are used as objective function to reallocate power generation while satisfying several equality and inequality constraints. The Pareto optimum in the fuel cost and emission objectives has a number of non-dominated solutions. The economic effects are analyzed under several different conditions, and $CO_2$ emission reductions offered by the use of storage are considered. The proposed approach can explore more efficient and noninferior solutions of a Multiobjective optimization problem. The method proposed is applied to a 4-machine 6-buses system to demonstrate its effectiveness.

• 대한기계학회논문집A
• /
• 제21권11호
• /
• pp.1765-1772
• /
• 1997

The layered composites have a character to change of structure stiffness with respect to the layup angles. The deformations in the fan-blades to be initially designed by considering efficiency and noise, etc., which arise due to the pressure during the fan operation, can make the fan inefficient. Thus, so as to minimize the deformations of the blades, it is needed to increase the stiffness of the blades. An investigation has been performed to develop the three dimensional layered composite shell element with the drilling degree of freedom and the optimization module for finding optimal layup angles with sensitivity analysis. And then they have been verified. In this study, the analysis model is engine cooling fan of automobile. In order to analyzes the stiffness of the composite fan blades, finite element analysis is performed. In addition, it is linked with optimal design process, and then the optimal angles that can maximize the stiffness of the blades are found. In the optimal design process, the deformations of the blades are considered as multiobjective functions, and this results minimum bending and twisting simultaneously.

• 전산구조공학
• /
• 제7권1호
• /
• pp.117-124
• /
• 1994

본 논문에서는 다단계다목적함수 최적화(MLMO)를 통해 철근콘크리트 뼈대구조의 최적해를 일단계단일 목적함수 최적화(SLSO)에 의한 결과와 비교하였다. MLMO방법에 의해서 간단히 가중치(Weighting factor)를 도모함으로써 경비와 처짐의 두가지 목적함수를 만족시키는 것이 가능했으며, 단계별로 제약조건식의 수를 감소시키고, 문제형성의 비선형성을 감소시킴으로써 최적화의 과정을 효율적으로 수행할 수 있었다. 또한 각 부구조물간의 설계변수의 변화에 의한 부재력의 변화를 제약조건에 반영하기 위하여 부재력변화량 추정을 하였고 부구조물의 최적화시 부재감 결합(coupling)이 가능하도록 하였다. 부구조물의 최적화시 선형화된 구조시스템의 선형화된 목적함수와 제약조건식을 사용하여 재해석 과정을 효율적으로 감소시킬 수 있었다. 최적화 과정중 초기에는 설계변수에 대한 비교적 큰 이동한계의 사용이 가능하였으며 반복회수 4호 정도에 최적해로의 효율적인 수렴이 이루어졌다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1994년도 하계학술대회 논문집 A
• /
• pp.165-167
• /
• 1994

This paper presents a new method for multiobjective optimization using Genetic Algorithm-Sexual Reproduction Model(SR model). In SR model, each individual consists of chromosome pairs. Sex cells(gametes) are produced through artificial meiosis in which crossover and mutation occur, The proposed method has two selection operators, one, individual selection which selects the individual to fertilize, and the other, gamete selection which makes zygote for offspring production, The two selection schemes are repectively conducted according to different fitness(or objective) function and consequently give a solution which is unbiased to any objectives. We apply the proposed method to optimization of the design parameters of Linear Induction Motor(LIM) and show its effectiveness.