• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.11a
• /
• pp.336-339
• /
• 2006

모델 예측 제어 시스템은 이동 제어 구간에서 원하는 출력과 예측된 출력의 차이를 최소화하는 현재의 제어 입력을 적용하는 방식을 사용한다. 제약조건이 있는 경우이거나 비선형 시스템 문제의 경우는 주어진 함수를 최소화하는 최적화 문제를 풀기가 힘들다. 본 논문에서는 모델 예측 제어 시스템의 최적화 문제를 실수 코딩 유전 알고리즘을 이용하여 효율적으로 구할 수 있음을 보인다. 또한 실수코딩 유전알고리즘이 여러 가지 면에서 디지털코딩 유전알고리즘보다 더 자연스럽고 유리함을 모의실험을 통해 보인다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.12a
• /
• pp.195-198
• /
• 2002

고농도오존이 발생되는 원인과 환경적 요인의 상호관계를 모델링하기 위해 신경회로 망과 같은 지능제어 기법들이 많이 적용되어 왔다 분석과 모델링을 위해 유전자 알고리즘과 같은 최적화 방법을 적용하기도 하지만, 고농도 오존이 발생되는 메커니즘이 매우 복잡하고, 비선형적이며, 패턴파악이 어렵기 때문에 고농도 오존의 예측 모델링에는 여전히 문제점이 있다 따라서 본 논문에서는 신뢰수준과 신뢰구간을 이용하여 초농도 오존을 예측할 수 있는 모델링 방법을 서술하였다 예측값의 신뢰수준의 평가는 예측에 대한 실측값을 구하여 신뢰구간내의 데이터의 개수를 파악함으로써 신뢰성을 평가할 수 있다. 또한 이 테스트는 우리가 가지고 있지 않은 데이터에 대한 유효성을 평가하는데 적용될 수 있다 그리고 본 논문에서는 GMDH(Group Method of data handling)의 전형적인 알고리즘에 바탕을 두고 있는 DPNN(Dynamic Polynomial Neural Network)를 이용하여 예측 모델을 구성하였다. DPNN은 데이터 해석이 용이하고 비선형적인 동적 시스템 예측에 유용하게 적용될 수 있는 장점을 가지고 있다.

• Journal of Bio-Environment Control
• /
• v.27 no.4
• /
• pp.326-331
• /
• 2018

Crop growth models are useful tools for understanding and integrating knowledge about crop growth. Models for predicting plant height, net photosynthesis rate, and plant growth of quinoa (Chenopodium quinoa Willd.) as a leafy vegetable in a closed-type plant factory system were developed using empirical model equations such as linear, quadratic, non-rectangular hyperbola, and expolinear equations. Plant growth and yield were measured at 5-day intervals after transplanting. Photosynthesis and growth curve models were calculated. Linear and curve relationships were obtained between plant heights and days after transplanting (DAT), however, accuracy of the equation to estimate plant height was linear equation. A non-rectangular hyperbola model was chosen as the response function of net photosynthesis. The light compensation point, light saturation point, and respiration rate were 29, 813 and $3.4{\mu}mol{\cdot}m^{-2}{\cdot}s^{-1}$, respectively. The shoot fresh weight showed a linear relationship with the shoot dry weight. The regression coefficient of the shoot dry weight was 0.75 ($R^2=0.921^{***}$). A non-linear regression was carried out to describe the increase in shoot dry weight of quinoa as a function of time using an expolinear equation. The crop growth rate and relative growth rate were $22.9g{\cdot}m^{-2}{\cdot}d^{-1}$ and $0.28g{\cdot}g^{-1}{\cdot}d^{-1}$, respectively. These models can accurately estimate plant height, net photosynthesis rate, shoot fresh weight, and shoot dry weight of quinoa.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.9A
• /
• pp.1299-1305
• /
• 2000

In B-ISDN, to realize ATM, the optimum control method of multi-media traffic must be proposed. Because there is not the traffic model of multi-media to make clear, the realization of optimum ATM congestion control is very difficult. In this paper, the traffic model is assumed to be slowly time-variable non-linear function and for real-time prediction of it, new model which is composed with parallel triple neural networks is proposed. And the simulation to predict assumed ATM traffic is executed. From the result, it's capability is shown that the proposed neural network model can be used in ATM congestion control.

• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.2401-2403
• /
• 2000

시스템의 특성을 이해하고 신뢰성 있는 제어를 위해서는 시스템에 대한 정확한 모델을 필요로 한다. 이러한 목적을 위해서 많은 연구자들에 의한 다양한 방법의 모델링 방법이 계속되어 연구되어지고 있다. 현재 많이 사용하는 모델링 방법 중에는 통계적 기법을 이용하는 것, first principle 방법을 이용하는 것, 지능형 기법을 이용하는 방법 등이 있다. 본 연구에서는 통계적 방법인 fractional factorial 방법을 이용한 모델, Taguchi 방법을 이용한 모델, 그리고 지능형 방법인 신경회로망을 이용한 모델의 3가지 모델을 사용해서 각 모델의 학습오차와 예측오차 등의 특성을 비교하였다. 모델에 사용된 데이터는 비선형 시스템인 플라즈마 화학 증착 장비(Plasma-Enhnaced Chemical Vapor Deposition : PECVD)에 의해 증착된 산화막 실험 데이터이다. 각 모델에 대해서 PECVD 데이터를 사용하여 모델을 만들었을 때 각 모델의 학습오차와 학습오차 변위, 그리고 예측오차와 예측오차변위를 조사하였다. 세가지 모델 모두 학습오차가 예측오차보다 작았으며 변위 또한 학습오차변위가 예측오차변위보다 작았다. 본 연구 결과는 일반적으로 신경회로망에 의한 오차가 다른 통계적인 방법에 의한 오차보다 작음을 보여준다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.5
• /
• pp.191-197
• /
• 2009

In this paper, multi-variable fuzzy model for efficient combustion control of refuse incineration plant is obtained. First, to obtain model of incineration plant which is complex and nonlinear multi-variable fuzzy modeling is performed. Obtained multi-variable fuzzy model predicts outputs of incinerator almost exactly. Then using multi-variable fuzzy model we can build simulator which is used as operation simulator for building of control strategy and training of operator.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.490-495
• /
• 1992

A neural model predictive control strategy combining a neural network for plant identification and a nonlinear programming algorithm for solving nonlinear control problems is proposed. A constrained nonlinear optimization approach using successive quadratic programming cooperates with neural identification network is used to generate the optimum control law for the complicate continuous/batch chemical reactor systems that have inherent nonlinear dynamics. Based on our approach, we developed a neural model predictive controller(NMPC) which shows excellent performances on nonlinear, model-plant mismatch cases of chemical reactor systems.

• Journal of Advanced Marine Engineering and Technology
• /
• v.41 no.1
• /
• pp.105-110
• /
• 2017

This paper introduces a heave compensation system for offshore crane when it subjected to unexpected disturbances such as ocean waves, tidal currents or winds and their external force. The dynamic model consists of a crane which is considered to behave in the same manner as a rigid body, a hydraulic driven winch, an elastic rope and a payload. To keep the payload from moving upwards and downwards, PD(Proportional-Derivative) control was applied by using linearization. In order to achieve a better performance, the sliding mode control and the nonlinear generalized predictive control algorithm was applied according to the time-delay. As a result, the oscillating amplitude of the payload was reduced by the control algorithm. Considering the time-delay involved in the system to be one second, nonlinear generalized predictive controller with a robust controller was a suitable control algorithm for this heave compensation system because it made the position of te payload reach the desired position with the minimum error. This paper presented a control algorithm using the robust control and its simulation results.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.5
• /
• pp.562-567
• /
• 2006

Model Predictive Control(MPC) system uses the current input which minimizes the difference between the desired output and the estimated output in the receding horizon scheme. In many cases (for example, system with constraints or nonlinear system), however, it is not easy to find the optimal solution to the above problem. In this paper, we show that real number coding genetic algorithm can be used to solve the optimal problem for MPC effectively. Also, we show by simulation that the real coding algorithm is mote natural and advantageous than the digital coding one.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.39 no.10
• /
• pp.985-993
• /
• 2015

Free-piston Stirling engines (FPSEs) have attracted much attention in the renewable energy field as a key device in the conversion from thermal to mechanical energy, and in the recycling of waste energy. Traditional Stirling engines consist of two pistons that are connected by a mechanical link, while FPSEs are formed as a vibration system by connecting each piston to a spring without a physical link. To ensure the correct design and control of operations, this requires elaborate dynamic-performance predictions. In this paper, we present the performance-prediction methodology using a linear and nonlinear dynamic analytical model considering the external load of FPSEs. We perform linear analyses to predict the operating point of the engine using the root locus technique. Using nonlinear analysis, we also predict the amplitude of pistons by performing numerical integration considering both the linear and nonlinear damping terms of the external load. We utilize the predicted dynamic behavior to predict the engine performance. In addition, we compare the experiment results and existing model predictions for RE-1000 to verify the reliability of the analytical model.