• Proceedings of the KIEE Conference
• /
• 2008.10b
• /
• pp.247-248
• /
• 2008

본 논문에서는 TYPE-2 퍼지 추론 기반의 RBF 뉴럴 네트워크(TYPE-2 Radial Basis Function Neural Network, T2RBFNN)를 설계하고 PSO(Particle Swarm Optimization) 알고리즘을 이용하여 모델의 파라미터를 동정한다. 제안된 모델의 은닉층은 TYPE-2 가우시안 활성 함수로 구성되며, 출력층은 Interval set 형태의 연결가중치를 갖는다. 여기에서 규칙 전반부 활성함수의 중심 선택은 C-means 클러스터링 알고리즘을 이용하고, 규칙 후반부 Interval set 형태의 연결가중치 결정에는 경사 하강법(Gradient descent method)을 이용한 오류 역전파 알고리즘을 사용하여 학습한다. 또한, 최적의 모델을 설계하기 위한 학습율 및 활성함수의 활성화 영역 결정에는 입자 군집 최적화(PSO; Particle Swarm Optimization) 알고리즘으로 동조한다. 마지막으로, 제안된 모델의 평가를 위하여 모의 데이터 집합(Synthetic dadaset)을 적용하고 근사화 및 일반화 능력에 대하여 토의한다.

• Journal of Advanced Marine Engineering and Technology
• /
• v.26 no.6
• /
• pp.705-711
• /
• 2002

디젤기관은 실린더 내경의 크기, 실린더 수 및 회전수에 따라 착화지연, 연소지연 및 디젤기관의 각종 정수가 달라지므로 비선형이 심한 시스템이다. 본 연구에서는 신경회로망을 이용하여 발전기를 구동하는 디젤 기관의 속도를 제어하는 디젤기관 신경회로망 디지털조속기를 제안한다. 이를 위하여 3상 50㎾ 발전기를 구동하는 4행정 4실린더, 1800 rpm ISUTSU 디젤기관의 실제 운전데이터로부터 뉴럴에뮬레이터를 구한다. 최적치 뉴럴에뮬레이터 구성을 위하여 다양한 역전파알고리즘으로 학습을 행하고 결과를 비교한다. 또한 디젤기관의 역으로부터 뉴럴 제어기를 구성하고 뉴럴에뮬레이터로 시뮬레이션을 행한다. 외란이 존재하는 경우에도 효과적인 뉴럴제어기를 구성하기 위하여 선택적 뉴럴제어 기의 사용을 제안한다. 또한 응답성을 향상하고 정확한 목표치추종을 위하여 PI제어기를 보조제어기로 사용하는 하이브리드제어기를 구성하여 시뮬레이션을 통하여 성능이 향상됨을 보인다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 1999.11a
• /
• pp.210-213
• /
• 1999

In this paper, we propose a new neural network for the control of a robot manipulator The proposed neural network structure is that all of network outputs feed bark into hidden units and output units from feedback units The feedback units are only to memorize the previous activations of the hidden units and output units and can be considered to function as one-step time delays. The proposed neural network works standard back-propagation Loaming algorithm. The simulation and experiment results showed the effectiveness of using the modified neural network structure in the control of the robot manipulator.

• The KIPS Transactions:PartB
• /
• v.14B no.6
• /
• pp.401-406
• /
• 2007

This study discusses the development of a u-Health care system that can detect quickly and cope actively with emergent situations of chronic disease patients who lead everyday life. If a patient's emergent situation is detected by personal health care host(PHCH), which is composed of acceleration and vibration sensors, GPS, and CDMA communication module, a text message on the patient's current location is transmitted to the hospital and the guardian's mobile terminal so that they can cope with the situation immediately. Especially, the system analyzes data from sensors by using neural network and determines emergent situations such as syncope and convulsion promptly. The exact location of patients can also be found in the electronic map by using GPS information. The experiments show that this system is very effective to find emergencies promptly for chronic disease patients who cannot take care of themselves and it is expected to save many lives.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.1
• /
• pp.24-29
• /
• 2005

In the fuzzy control for the multi-variable system, it is difficult to obtain the fuzzy rule. Therefore, the parallel structure of the independent single input-single output fuzzy controller using a pairing between the input and output variable is applied to the multi-variable system. However, among the input/output variables which arc not paired the interactive effects should be taken into account. these mutual coupling of variables affect the control performance. Therefore, for the control system with a strong coupling property, the control performance is sometimes lowered. In this paper, the effect of mutual coupling of variables is considered by the introduction of a neuro-fuzzy controller using relative gain matrix. This proposed neuro-fuzzy controller automatically adjusts the mutual coupling weight between variables using a neural network which is realized by back-propagation algorithm. The good performance of the proposed nero-fuzzy controller is verified through computer simulations on 200MW boiler systems.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.18 no.6
• /
• pp.464-476
• /
• 1998

Determination of location, shape and size of a flaw from its eddy current testing signal is one of the fundamental issues in eddy current nondestructive evaluation of steam generator tubes. Here, we propose an approach to this problem; an inversion of eddy current flaw signal using neural networks trained by finite element model-based synthetic signatures. Total 216 eddy current signals from four different types of axisymmetric flaws in tubes are generated by finite element models of which the accuracy is experimentally validated. From each simulated signature, total 24 eddy current features are extracted and among them 13 features are finally selected for flaw characterization. Based on these features, probabilistic neural networks discriminate flaws into four different types according to the location and the shape, and successively back propagation neural networks determine the size parameters of the discriminated flaw.