• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2002.10a
• /
• pp.41-46
• /
• 2002

In the face of global competition, the requirements fer the continuously increasing productivity, flexibility and quality (dimensional accuracy, mechanical properties and surface properties) have imposed a major change on steel manufacturing industries. The automation of hot rolling process requires the developments of several mathematical models for simulation and quantitative description of the industrial operations involved. To achieve this objectives, a new loaming method with neural network to improve the accuracy of rolling force prediction in hot rolling mill is developed. Also, Genetic Algorithm(GA) is applied to select the optimal structure of the neural network and compared with that of engineers experience. It is shown from this research that both structure selection methods can lead to similar results.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.12a
• /
• pp.131-134
• /
• 2002

일반적인 산업현장에서 많이 사용되는 이중탱크 시스템은 동작점 근방에서 선형화하는 고전제어기법을 사용한 것으로서 큰 시간지연과 비선형성으로 인해 정확한 수학적 모델링이 어렵고 모델링을 하더라도 넓은 동작 영역에서 만족스로운 결과를 얻기 어렵다. 따라서, 비교적 모델링에 대한 의존도가 낮은 퍼지, 신경회로망, 유전알고리즘 등의 지능제어 기법들도 제안되고 있다. 그러나 이들 제어기 역시 외란이나 다양한 동작 모드들에 따른 제어기 변수들의 적응성 저하로 인해 안정화 가능 영역이 협소해 지는 것은 물론 시스템의 불안정 현상도 초래한다. 이에 반해, SMC(sliding mode controller)는 변수의 변동, 외란에 둔감한 강점을 갖고 있지만, 시스템의 상태에 따른 슬라이딩 평면 설정의 곤란성과 채터링(chattering)이 존재하는 문제점 이 있다. 따라서 본 논문에서는 이중 탱크 시스템의 정밀한 수위 제어를 위하여, GA과 FLC를 사용하여 최적 변수로 설정 할 수 있게 하고, 채터링 저감을 위해 시스템 동특성 변동과 외란 에 강인한 GA-FSMC(genetic algorithm fuzzy sliding mode controller)를 제안하였다. 시뮬레이션을 통해 종래의 제어기의 제어결과와 비교함으로써 제안하는 GA-FSMC의 우수성을 입증하고자 한다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.12
• /
• pp.2613-2625
• /
• 1997

Theoretically, a multi-layered feedforward network has been known to be able to approximate a continuous function to an arbitrary degree of accuracy. However, these networks fail to approximate discontinuous functions when they are trained by well-known training algorithms. This paper presents a training algorithm which doesn't work consists of one or more modules, which are trained in a sequential order within subspaces of the input space, and is trained very rapidely once all modules are trained and merged. The experimantal results of applying this method indicates the proposed training algorithm is superior to traditional ones such as baskpagation.

• Proceedings of the KIEE Conference
• /
• 1996.11a
• /
• pp.401-403
• /
• 1996

In this paper, a new control strategy is presented that combines sliding mode control theory with a neural network. Sliding mode control theory requires the complete knowledge of the dynamics of the controlled system. However, in practice, one often bas only a small number of state measurements. This could be a serious limitation on the practical usefulness of sliding mode control theory. A multilayer neural network is employed to solve this kind of problem. The neural network serves as a compensator without a prior knowledge about the system. The proposed control algorithm is applied to a class of uncertain nonlinear system. The robustness against parameter uncertainty, nonlinearity and external disturbances, and the effectiveness is verified by the simulation results.

• Proceedings of the KIEE Conference
• /
• 1994.11a
• /
• pp.408-410
• /
• 1994

This Paper Presents a learning strategy for neural networks based on genetic algorithms and evolution strategies. Genetic algorithms and evolution strategies are used to train weights of feedforward neural network to solve problems faster than neural network, especially backpropagation. Simulations are performed exclusive-OR problem, full-adder problem, sine function generator to demonstrate the effectiveness of neural-GA-ES.

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.1730-1731
• /
• 2007

In this paper, we present a direct adaptive control method using neural networks for the control of nonlinear systems. The weights of neural networks are trained by an adaptive backpropagation algorithm based on Lyapunov stability theory. We develop the parameter update-laws using the neural network input and the error between the desired output and the output of nonlinear plant to update the weights of a neural network in the sense that Lyapunove stability theory. Beside the output tracking error is asymptotically converged to zero.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.11a
• /
• pp.189-192
• /
• 2006

본 논문에서는 ELM(Extreme Learning Machine)을 이용하여 계산속도 뿐만 아니라 성능면에서도 우수한 입력 속성선택 기법을 제안한다. 일반적으로 입력 속성 선택문제는 다양한 속성들의 영향을 고려함으로써 모든 입력속성들을 평가하는데 많은 계산량이 요구되는 단점이 있다. 이러한 문제점을 개선하기 위하여 학습속도가 기존의 신경회로망에 비하여 월등히 우수한 ELM 알고리즘을 적용한다. 입력속성 선택은 ELM으로부터 산출된 출력값을 이용하여 출력 오차에 영향이 큰 속성들 순으로 순위를 결정한 후, 전방향 선택이나 후방향 선택기법을 이용하여 입력속성을 선택한다. 제안된 방법은 다양한 데이터에 적용하여 타당성을 검증한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.720-722
• /
• 2009

Optimal operation of electric power generating plants is very essential for any power utility organization to reduce input costs and possibly the prices of electricity in general. This paper developed models for load forecasting using neural networks approach. This model is tested using actual load data of the Busan and weather data to predict the load of the Busan for one month in advance. The test results showed that the neural network forecasting approach is more suitable and efficient for a forecasting application.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.10
• /
• pp.1394-1397
• /
• 2020

The urine test used as a basic test method of in vitro diagnosis for health care has been used for a long time to be simple and convenient. The urine test method is using a color that appears depending on the change in the ion concentration that reacts over time buried in the standard color test paper(Strips) with a urine sample applied to some reaction reagents. In this paper, it was proposed a neural network algorithm to obtain a suitable and reproducibility and accuracy classifier suitable for the urine analysis system. The experimental results were compared with the visual colorimetric analysis, and the neural network algorithm showed better results.

• Journal of the Korean Society for Precision Engineering
• /
• v.10 no.1
• /
• pp.77-88
• /
• 1993

In this paper, we propose a stereo vision system to solve correspondence problem with large disparity and sudden change in environment which result from small distance between camera and working objects. First of all, a specific feature is divided by predfined elementary feature. And then these are combined to obtain coded data for solving correspondence problem. We use Neural Network to extract elementary features from specific feature and to have adaptability to noise and some change of the shape. Fourier transformation and Log-polar mapping are used for obtaining appropriate Neural Network input data which has a shift, scale, and rotation invariability. Finally, we use associative memory to obtain coded data of the specific feature from the combination of elementary features. In spite of specific feature with some variation in shapes, we could obtain satisfactory 3-dimensional data from corresponded codes.