• 한국농공학회지
• /
• 제29권4호
• /
• pp.73-92
• /
• 1987

Formulation of the geometric optimization for truss structures based on the elasticity theory turn out to be the nonlinear programming problem which has to deal with the cross-sectional area of the member and the coordinates of its nodes simultaneously. A few techniques have been proposed and adopted for the analysis of this nonlinear programming problem for the time being. These techniques, however, bear some limitations on truss shapes, loading conditions and design criteria for the practical application to real structures. A generalized algorithm for the geometric optimization of the truss structures, which can eliminate the above mentioned limitations, is developed in this study. The algorithm proposed utilizes the two-levels technique. In the first level which consists of two phases, the cross-sectional area of the truss member is optimized by transforming the nonlinear problem into SUMT, and solving SUMT utilizing the modified Newton Raphson method. In the second level, which also consists of two phases the geometric shape is optimized utillzing the unindirectional search technique of the Powell method which make it possible to minimize only the objective functlon. The algorithm proposed in this study is numerically tested for several truss structures with various shapes, loading conditions and design criteria, and compared with the results of the other algorithms to examine its applicability and stability. The numerical comparisons show that the two- levels algorithm proposed in this study is safely applicable to any design criteria, and the convergency rate is relatively fast and stable compared with other iteration methods for the geometric optimization of truss structures. It was found for the result of the shape optimization in this study to be decreased greatly in the weight of truss structures in comparison with the shape optimization of the truss utilizing the algorithm proposed with the other area optimum method.

• Smart Structures and Systems
• /
• 제24권3호
• /
• pp.303-317
• /
• 2019

Magneto-rheological fluid (MRF) rotatory dampers are normally used for controlling the constant rotation of machines and engines. In this research, such a device is proposed to act as variable stiffness device to alleviate the rotational oscillation existing in the many engineering applications, such as motor. Under such thought, the main purpose of this work is to characterize the nonlinear torque-angular displacement/angular velocity responses of an MRF based variable stiffness device in oscillatory motion. A rotational hysteresis model, consisting of a rotatory spring, a rotatory viscous damping element and an error function-based hysteresis element, is proposed, which is capable of describing the unique dynamical characteristics of this smart device. To estimate the optimal model parameters, a modified whale optimization algorithm (MWOA) is employed on the captured experimental data of torque, angular displacement and angular velocity under various excitation conditions. In MWOA, a nonlinear algorithm parameter updating mechanism is adopted to replace the traditional linear one, enhancing the global search ability initially and the local search ability at the later stage of the algorithm evolution. Additionally, the immune operation is introduced in the whale individual selection, improving the identification accuracy of solution. Finally, the dynamic testing results are used to validate the performance of the proposed model and the effectiveness of the proposed optimization algorithm.

• 제어로봇시스템학회논문지
• /
• 제10권12호
• /
• pp.1210-1216
• /
• 2004

This paper deals with a human-like gait generation of a biped robot with a balancing weight of an inverted pendulum type by using genetic algorithm. The ZMP (Zero Moment Point) is the most important index in a biped robot's dynamic walking stability. To perform a stable walking of a biped robot, a balancing motion is required according to legs' trajectories and a desired ZMP trajectory. A dynamic equation of the balancing motion is nonlinear due to an inverted pendulum type's balancing weight. To solve the nonlinear equation by the FDM (Finite Difference Method), a linearized model of equation is proposed. And GA (Genetic Algorithm) is applied to optimize a human-like balancing motion of a biped robot. By genetic algorithm, the index of the balancing motion is efficiently optimized, and a dynamic walking stability is verified by the ZMP verification equation. These balancing motion are simulated and experimented with a real biped robot IWR-IV. This human-like gait generation will be applied to a humanoid robot, at future work.

• Journal of Electrical Engineering and Technology
• /
• 제9권6호
• /
• pp.2142-2153
• /
• 2014

Linearization of transducer characteristic plays a vital role in electronic instrumentation because all transducers have outputs nonlinearly related to the physical variables they sense. If the transducer output is nonlinear, it will produce a whole assortment of problems. Transducers rarely possess a perfectly linear transfer characteristic, but always have some degree of non-linearity over their range of operation. Attempts have been made by many researchers to increase the range of linearity of transducers. This paper presents a method to compensate nonlinearity of Linear Variable Displacement Transducer (LVDT) based on Extreme Learning Machine (ELM) method, Differential Evolution (DE) algorithm and Artificial Neural Network (ANN) trained by Genetic Algorithm (GA). Because of the mechanism structure, LVDT often exhibit inherent nonlinear input-output characteristics. The best approximation capability of optimized ANN technique is beneficial to this. The use of this proposed method is demonstrated through computer simulation with the experimental data of two different LVDTs. The results reveal that the proposed method compensated the presence of nonlinearity in the displacement transducer with very low training time, lowest Mean Square Error (MSE) value and better linearity. This research work involves less computational complexity and it behaves a good performance for nonlinearity compensation for LVDT and has good application prospect.

• 대한전자공학회논문지TE
• /
• 제39권4호
• /
• pp.389-394
• /
• 2002

본 논문에서는 비선형 확장 칼만 필터를 이용한 양방향 스테레오 정합 알고리즘을 제안한다. 제안 알고리즘은 비선형 칼만 필터를 사용하여 변이(disparity)를 예측하고 예측된 변이는 좌영상에서 우영상으로의 스테레오 정합에 적용된다. 변이 예측은 몇 번의 반복으로 구해지며 비선형 칼만 필터의 초기 상태 예측치에 큰 오차를 나타내는 단점을 극복하기 위하여 양방향 스테레오 정합 알고리즘을 사용하였다. 이웃 화소의 영향을 고려하여 선형 내삽법(intepolation)을 좌·우 영상에 적용함으로써 스테레오 정합에 강인한 결과를 나타내었다. 제안 알고리즘의 성능 평가를 위해서 기존의 SSD방법과 비교 검토하였다. 비교 결과 제안 알고리즘이 매우 우수한 정합 성능을 가짐을 알 수 있었다.

• Structural Engineering and Mechanics
• /
• 제14권1호
• /
• pp.21-38
• /
• 2002

A control algorithm combining viscous and non-linear Reid damping mechanisms has been recently proposed by the authors to command active friction dampers. In this paper, friction dampers and the proposed algorithm are applied to control the seismic responses of a nonlinear 20-story building. Piezoelectric stack actuators are used to implement the control algorithm. The capacity of each damper is determined by the practical size of piezoelectric actuators and the availability of power supply. The saturation effect of the actuators on the building responses is investigated. To minimize the peak story drift ratio or floor acceleration of the building structure, a practical sequential procedure is developed to sub-optimally place the dampers on various floors. The effectiveness of active friction dampers and the efficiency of the proposed sequential procedure are verified by subjecting the building structure to four earthquakes of various intensities. The performance of 80 dampers and 137 dampers installed on the structure is evaluated according to 5 criteria. Numerical simulations indicated that the proposed control algorithm effectively reduces the seismic responses of the uncontrolled 20-story building, such as inelastic deformation. The sub-optimal placement of dampers based on peak acceleration outperforms that based on peak drift ratio for structures subjected to near-fault ground motions. Saturation of piezoelectric actuators has adverse effect on floor acceleration.

• 대한전기학회논문지
• /
• 제45권4호
• /
• pp.543-552
• /
• 1996

The goal of this paper is to developed a DSP based power analyzing and control system by 3-Dimensional (3-D) space current co-ordinates. A developed system is made up of 486-PC and DSP (Digital Signal Processor) board, Active Power Filter, Non-linear thyristor load, and Power analyzing and control program for Windows. Power is analyzed using signal processing techniques based on the correlation between voltage and current waveforms. Since power analysis algorithm is performed by DSP, power analysis is achieved in real-time even under highly dynamic nonlinear loading conditions. Combining control algorithm with power analysis algorithm is performed by DSP, power analysis is achieved in real-time even under highly dynamic nonlinear loading conditions. Combining control algorithm with power analysis algorithm, flexibility of the proposed system which has both power analysis mode and control mode, is greatly enhanced. Non-active power generated while speed of induction motor is controlled by modulating firing angle of thyristor converter, is compensated by Active Power Filter for verifying a developed system. Power analysis results, before/after compensation, are numerically obtained and evaluated. From these results, various graphic screens for time/frequency/3-D current co-ordinate system are displayed on PC. By real-time analysis of power using a developed system, power quality is evaluated, and compared with that of conventional current co-ordinate system. (author). refs., figs. tabs.

• 한국산학기술학회논문지
• /
• 제14권1호
• /
• pp.378-384
• /
• 2013

본 논문에서는 FCM 클러스터링 알고리즘을 기반으로 하는 퍼지뉴럴네트워크를 제안한다. 일반적으로, 퍼지규칙을 생성할 때 차원이 증가하면 퍼지 규칙의 수가 기하급수적으로 증가하는 문제를 가지고 있다. 이를 해결하기 위해, 제안된 네트워크의 퍼지 규칙은 FCM 클러스터링 알고리즘을 이용하여 입력 공간을 분산 형태로 분할함으로써 생성한다. 퍼지 규칙의 전반부 파라미터는 FCM 클러스터링 알고리즘에 의한 소속행렬로 결정된다. 퍼지 규칙의 후반부는 다항식 함수의 형태로 표현되며, 퍼지뉴럴네트워크의 학습은 뉴런의 연결을 조절함으로써 실현되고, 오류 역전파 알고리즘에 의해 행해진다. 마지막으로, 제안된 네트워크는 비선형 공정으로의 적용을 통해 성능을 평가한다.

• 한국지능시스템학회논문지
• /
• 제27권1호
• /
• pp.50-58
• /
• 2017

비선형 시스템 제어에 TSK 퍼지 모델을 이용하는 연구가 많다. 그러나 비선형 시스템을 TSK 퍼지 모델로 표현하기 위해서는 대상 시스템의 비선형 동특성 방정식을 알고 있거나 시스템으로부터 실험으로 입출력 공간에 충분히 분포된 데이터를 얻을 수 있어야 한다. 본 논문에서는 비선형 시스템의 TSK 퍼지 모델링에 계단 입력 응답과 GA만을 이용하는 방법을 제안한다. 연구 대상인 시스템은 제어 입력에 비선형인 경우와 출력에 비선형인 두 경우로 하였다. 제어 입력에 비선형인 경우 여러 입력 값에 대한 계단 입력 응답 데이터들로 모델링 하며, 제어 입력에 비선형인 경우에는 계단 입력 응답 데이터와 제어 입력 값이 0인 경우 응답 데이터를 이용하여 모델링 한다. 또한 제안한 방법으로 구해진 TSK 퍼지 모델로부터 퍼지 제어기를 설계하는 방법도 제시한다. 제안한 방법들을 예제들에 적용하여 모의 실험한 결과 원하는 제어 결과를 얻을 수 있었다.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제50권7호
• /
• pp.339-349
• /
• 2001

This paper suggests an optimal identification method for complex and nonlinear system modeling that is based on Fuzzy-Neural Networks(FNN). The proposed Hybrid Identification Algorithm is based on Yamakawa's FNN and uses the simplified inference as fuzzy inference method and Error Back Propagation Algorithm as learning rule. In this paper, the FNN modeling implements parameter identification using HCM algorithm and hybrid structure combined with two types of optimization theories for nonlinear systems. We use a HCM(Hard C-Means) clustering algorithm to find initial apexes of membership function. The parameters such as apexes of membership functions, learning rates, and momentum coefficients are adjusted using hybrid algorithm. The proposed hybrid identification algorithm is carried out using both a genetic algorithm and the improved complex method. Also, an aggregated objective function(performance index) with weighting factor is introduced to achieve a sound balance between approximation and generalization abilities of the model. According to the selection and adjustment of a weighting factor of an aggregate objective function which depends on the number of data and a certain degree of nonlinearity(distribution of I/O data), we show that it is available and effective to design an optimal FNN model structure with mutual balance and dependency between approximation and generalization abilities. To evaluate the performance of the proposed model, we use the time series data for gas furnace, the data of sewage treatment process and traffic route choice process.