• 한국컴퓨터정보학회논문지
• /
• 제19권9호
• /
• pp.33-38
• /
• 2014

FCM 기반하이브리드 RBF 네트워크는 서로 다른 학습 구조가 결합된 혼합형 모델로서, 입력층과 중간층의 학습 구조는 FCM 알고리즘을 적용하고, 중간층과 출력층 사이의 학습 구조는 Max_Min 알고리즘을 적용한다. 입력층과 중간층의 학습시 입력 벡터와 중간층의 노드 중에서 중심과 입력 벡터간의 가장 가까운 노드를 승자 노드로 선택하여 출력층으로 전달한다. 그리고 중간층과 출력층 사이의 학습구조인 Max_Min 신경망은 중간층의 승자 뉴런이 입력벡터로 적용된다. 그러나 많은 패턴이 입력벡터로 제시될 경우에는 학습성능이 저하되는 단점이 있다. 따라서 본 논문에서는 중간층과 출력층의 학습 구조인 Max_Min 알고리즘의 학습 성능을 향상시키기 위해 퍼지 논리 시스템을 이용한 학습률 자동 조정 방법을 제안한다. 제안된 방법은 목표값과 출력값의 차이에 대한 절대값이 0.1보다 적거나 같으면 정확성으로 분류하고 크면 부정확성으로 분류한다. 정확성의 총 개수를 퍼지 제어 시스템에 적용하여 학습률을 동적으로 조정한다. 제안된 방법의 학습 및 인식 성능을 평가하기 위해 컨테이너에서 추출한 숫자, 영문 식별자를 인식 및 성능평가 실험에 적용한 결과, 제안된 방법이 문자 패턴 인식에 효과적임을 확인할 수 있었다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 하계학술대회 논문집 B
• /
• pp.507-509
• /
• 1999

In this paper, we propose a new fuzzy model identification method that can yield a successful fuzzy rule base for fundamental approximations. The method in this paper uses a set of input-output data and is based on a hybrid messy genetic algorithm (mGA) with a fine-tuning scheme. The mGA processes variable-length strings, while standard GAs work with a fixed-length coding scheme. For successfully identifying a complex nonlinear system, we first use the mGA, which coarsely optimizes the structure and the parameters of the fuzzy inference system, and then the gradient descent method which tine tunes the identified fuzzy model. In order to demonstrate the superiority and efficiency of the proposed scheme, we finally show its application to a nonlinear approximation.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제49권8호
• /
• pp.423-431
• /
• 2000

This paper presents a systematic approach to the input-output data-based fuzzy modeling for the complex and uncertain nonlinear systems, in which the conventional mathematical models may fail to give the satisfying results. To do this, we propose a new method that can yield a successful fuzzy model using a mGA hybrid schemes with a fine-tuning method. We also propose a new coding method fo chromosome for applying the mGA to the structure and parameter identifications of fuzzy model simultaneously. During mGA search, multi-purpose fitness function with a penalty process is proposed and adapted to guarantee the accurate and valid fuzzy modes. This coding scheme can effectively represent the zero-order Takagi-Sugeno fuzzy model. The proposed mGA hybrid schemes can coarsely optimize the structure and the parameters of the fuzzy inference system, and then fine tune the identified fuzzy model by using the gradient descent method. In order to demonstrate the superiority and efficiency of the proposed scheme, we finally show its applications to two nonlinear systems.

• 한국정밀공학회지
• /
• 제18권9호
• /
• pp.71-81
• /
• 2001

This paper analyzes the effectiveness of minimizing vibration and sound transmission on/through a thin rectangular plate by both feedback control and hybrid control which combines adaptive feedforward control with a feedback loop. An experimental system identification technique using the matrix-fractional curve-fitting of the frequency response data is introduced for complex shaped structures. This identification technique reduces the model order o the MIMO(Multi-Input Multi-Output) system which simplifies the practical implementation. The adaptive feedforward control uses a Multiple filtered-x LMS(Least Mean Square) algorithm and the feedback control uses a multivariable digital LQG(Linear Quadratic Gaussian) algorithm. Experimental results show that an effective reduction of sound transmission is achieved by the hybrid control scheme when both vibration and noise measurement signals are incorporated in the controller.

• 제어로봇시스템학회논문지
• /
• 제4권5호
• /
• pp.651-657
• /
• 1998

In this study, an intelligent scheduling with hybrid architecture, which integrates expert system and neural network, is proposed. Neural network is trained with the data acquired from simulation model of FMC to obtain the knowledge about the relationship between the state of the FMC and its best dispatching rule. Expert system controls the scheduling of FMC by integrating the output of neural network, the states of FMS, and user input. By applying the hybrid system to a scheduling problem, the human knowledge on scheduling and the generation of non-logical knowledge by machine teaming, can be processed in one scheduler. The computer simulation shows that comparing with MST(Minimum Slack Time), there is a little increment in tardness, 5% growth in flow time. And at breakdown, tardness is not increased by expert system comparing with EDD(Earliest Due Date).

• 대한교통학회지
• /
• 제6권2호
• /
• pp.7-20
• /
• 1988

Recent developments in combining transportation planning models and input-output approaches, together with inclusion of intensity of land uses, have made it possible to construct realistic comprehensive urban and regional activity models. These modes form the basis for a rigorous approach to studying the interactions among urban activities. However, efficient computational solution methods for implementing such comprehensive models are still not available. In this paper an efficient solution method for the urban activity model is developed by combining Evans' partial linearization technique with Powell's hybrid method. The solution algorithm is applied to a small but realistic urban area with a detailed transportation network.

• KEPCO Journal on Electric Power and Energy
• /
• 제8권1호
• /
• pp.21-29
• /
• 2022

Power generated from renewable energy has continuously increased recently. As the distributed generation begins to interconnect in the distribution system, an accurate generation forecasting has become important in efficient distribution planning. This paper explained method and current state of distributed power generation forecasting models. This paper presented selecting input and output variables for the forecasting model. In addition, this paper analyzed input variables and forecasting models that can use as mid-to long-term distributed power generation forecasting.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 2006년도 International Symposium on GPS/GNSS Vol.1
• /
• pp.277-282
• /
• 2006

It is well known that Kalman filtering is an optimal real-time data fusion method for GPS/INS integration. However, it has some limitations in terms of stability, adaptability and observability. A Kalman filter can perform optimally only when its dynamic model is correctly defined and the noise statistics for the measurement and process are completely known. It is found that estimated Kalman filter states could be influenced by several factors, including vehicle dynamic variations, filter tuning results, and environment changes, etc., which are difficult to model. Neural networks can map input-output relationships without apriori knowledge about them; hence a proper designed neural network is capable of learning and extracting these complex relationships with enough training. This paper presents a GPS/INS integrated system that combines Kalman filtering and neural network algorithms to improve navigation solutions during GPS outages. An Extended Kalman filter estimates INS measurement errors, plus position, velocity and attitude errors etc. Kalman filter states, and gives precise navigation solutions while GPS signals are available. At the same time, a multi-layer neural network is trained to map the vehicle dynamics with corresponding Kalman filter states, at the same rate of measurement update. After the output of the neural network meets a similarity threshold, it can be used to correct INS measurements when no GPS measurements are available. Selecting suitable inputs and outputs of the neural network is critical for this hybrid method. Detailed analysis unveils that some Kalman filter states are highly correlated with vehicle dynamic variations. The filter states that heavily impact system navigation solutions are selected as the neural network outputs. The principle of this hybrid method and the neural network design are presented. Field test data are processed to evaluate the performance of the proposed method.

• 한국통신학회논문지
• /
• 제33권6B호
• /
• pp.456-468
• /
• 2008

본 논문에서는 혼합형 2단 AWG 기반의 WDM-PON을 하드웨어적으로 구현하기 위한 초기 단계로서 입출력 모듈을 설계하고 로직 시뮬레이션을 통해 동작을 검증한다. 혼합형 2단 AWG 기반의 WDM-PON은 32개의 파장을 통하여 128개의 ONU에게 서비스를 제공한다. 이때, 하향 전송에서 각 ONU는 각기 할당된 별도의 파장을 이용하는 반면 상향 전송의 경우 4개의 ONU가 단일의 파장을 공유하는 형태이다. 설계한 WDM-PON MAC 칩은 sub-MAC을 기반으로 하며, 각 sub-MAC마다 제어부, 수신부 그리고 네 개의 송신부로 구성된다. 따라서 본 논문에서는 sub-MAC을 구성하는 송 수신부의 기능, 사용되는 핀, 제어 신호 및 타이밍을 정의하고 이를 기반으로 각 기능 모듈을 설계한다. 설계한 WDM-PON MAC 칩은 각 입출력 모듈이 1Gbps의 송수신률을 가지는 것을 목표로 하였으며 이 동작을 위하여 125MHz 구동 클럭에 맞도록 설계된다. WDM-PON MAC 칩의 설계과정은 FSM(Finite State Machine)을 이용한 설계 흐름을 따랐으며 설계한 sub-MAC의 입출력 기능의 검증 및 성능 평가를 위하여 ModelSIM에서 각 기능별로 시나리오를 작성하고 이를 기반으로 로직 시뮬레이션을 수행한다.

• Geomechanics and Engineering
• /
• 제22권1호
• /
• pp.81-95
• /
• 2020

Measurement of the unconfined compressive strength (UCS) of the rock is critical to assess the quality of the rock mass ahead of a tunnel face. In this study, extensive field studies have been conducted along 3,885 m of the new Nagasaki tunnel in Japan. To predict UCS, a hybrid model of artificial neural network (ANN) based on genetic algorithm (GA) optimization was developed. A total of 1350 datasets, including six parameters of the Measurement-While- Drilling data and the UCS were considered as input and output parameters respectively. The multiple linear regression (MLR) and the ANN were employed to develop contrast models. The results reveal that the developed GA-ANN hybrid model can predict UCS with higher performance than the ANN and MLR models. This study is of great significance for accurately and effectively evaluating the quality of rock masses in tunnel engineering.