• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 Proceedings of the Korea Automatic Control Conference, 11th (KACC); Pohang, Korea; 24-26 Oct. 1996
• /
• pp.295-298
• /
• 1996

In distillation column control, secondary measurements such as temperatures and flows are widely used in order to infer product composition. This paper addresses the design of static estimators using the secondary measurements for estimating the product compositions of the multicomponent distillation columns. Based on the unified framework for the estimator problems, the relationships among several typical static estimators are discussed including the effect of the measured inputs. Design guidelines for the composition estimator using PLS regression are also presented. The estimator based on the guidelines is robust to sensor noise and has a good predictive power.

• 한국생산제조학회지
• /
• 제8권5호
• /
• pp.70-75
• /
• 1999

Adaptive control in the robotic GMA(Gas Metal Arc) welding is employed to monitor the information about weld characteristics and process paramters as well as modification of those parameters to hold weld quality within the acceptable limits. Typical characteristics are the bead geometry composition micrrostructure appearance and process parameters which govern the quality of the final weld. The main objectives of this paper are to realize the mapping characteristicso f penetration through the learning. After learning the neural network can predict the pene-traition desired from the learning mapping characteristic. The design parameters of the neural network estimator(the number of hidden layers and the number of nodes in a layer) were chosen from an error analysis. partial-penetration single-pass bead-on-plate welds were fabricated in 12mm mild steel plates in order to verify the performance of the neural network estimator. The experimental results show that the proposed neural network estimator can predict the penetration with reasonable accuracy and gurarantee the uniform weld quality.

• 한국생산제조학회지
• /
• 제7권6호
• /
• pp.57-63
• /
• 1998

An adaptive control in the robotic GMA welding is employed to monitor information about weld characteristics and process parameters as well as to modify those parameters to hold weld quality within acceptable limits. Typical characteristics are the bead geometry, composition, microstructure, appearance, and process parameters which govern the quality of the final weld. The main objectives of this thesis are to realize the mapping characteristics of bead width through learning. After learning, the neural estimation can estimate the bead width desired form the learning mapping characteristic. The design parameters of the neural network estimator(the number of hidden layers and the number of nodes in a layer) are chosen from an estimation error analysis. A series of bead of bead-on-plate GMA welding experiments was carried out in order to verify the performance of the neural network estimator. The experimental results show that the proposed neural network estimator can predict the bead width with reasonable accuracy and guarantee the uniform weld quality.

• 한국정밀공학회지
• /
• 제12권4호
• /
• pp.28-38
• /
• 1995

The flank wear at the minor cutting edge significantly affects the geometric accuracy and surface roughness in finish machining. A fuzzy estimator based on a fuzzy inference algorithm with a max-min composition rule is introduced to evaluate the minor flank wear length. The features sensitive to minor flank wear are extracted from the dispersion analysis of a time series AR model of the feed directional acceleration signal. These features, dispersions, are used for constructing linguistic rules, and then the fuzzy inferences are carried out with test data sets collected under various cutting conditions. The proposed system turns out to be effective for estimating minor flank wear length.

• Animal Bioscience
• /
• 제34권4호
• /
• pp.558-566
• /
• 2021

Objective: Objectives were to estimate energy and protein requirements of dairy crossbred steers, as well as to evaluate equations previously described in the literature (HH46 and CS16) to predict the carcass and empty body chemical composition of crossbred dairy cattle. Methods: Thirty-three Holstein×Zebu steers, aged 19±1 months old, with an initial shrunk body weight (BW) of 324±7.7 kg, were randomly divided into three groups: reference group (n = 5), maintenance level (1.17% BW; n = 4), and the remaining 24 steers were randomly allocated to 1 of 4 treatments. Treatments were: intake restricted to 85% of ad libitum feed intake for either 0, 28, 42, or 84 d of an 84-d finishing period. Results: The net energy and the metabolizable protein requirements for maintenance were 0.083 Mcal/EBW0.75/d and 4.40 g/EBW0.75, respectively. The net energy (NEG) and protein (NPG) requirements for growth can be estimated with the following equations: NEG (Mcal/kg EBG) = $0.2973_{({\pm}0.1212)}{\times}EBW^{0.4336_{({\pm}0.1002)}$ and NPG (g/d) = 183.6(±22.5333)×EBG-2.0693(±4.7254)×RE, where EBW, empty BW; EBG, empty body gain; and RE, retained energy. Crude protein (CP) and ether extract (EE) chemical contents in carcass, and all the chemical components in the empty body were precisely and accurately estimated by CS16 equations. However, water content in carcass was better predicted by HH46 equation. Conclusion: The equations proposed in this study can be used for estimating the energy and protein requirements of crossbred dairy steers. The CS16 equations were the best estimator for CP and EE chemical contents in carcass, and all chemical components in the empty body of crossbred dairy steers, whereas water in carcass was better estimated using the HH46 equations.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1997년도 한국자동제어학술회의논문집; 한국전력공사 서울연수원; 17-18 Oct. 1997
• /
• pp.1553-1556
• /
• 1997

In the paper, we porpose a new mehtod of extending PLS(Partial Least Squares) regressiion method to nonlinear framework and apply it to the estimation of product compositions in high-purity distillation column. There have veen similar efforets to overcome drawbacks of PLS by using nonlinear-mapping ability of meural networks, however, they failed to show great improvement over PLS since they focused only in capturing nonlinear functional relationship between input data, not on nonlinear correlation inthe data set. By incorporating the structure of Robust Auto Associative Networks(RAAN) into that of previous nonlinear PLS, we can handle nonlinear correlation as well as nonlinear functional relationship. The application result shows that the proposed method performs better than previous ones even for nonlinearities caused by changing operating conditions, limited observations, and existence of meas-unrement noises.

• 한국산림과학회지
• /
• 제97권2호
• /
• pp.171-180
• /
• 2008

The characterization of tree species and forest stand conditions is useful in the planning of activities aimed to conserve biodiversity. The main objective of this study was to describe tree species diversity, species composition and stand structure of tropical deciduous forests distributed in three regions in Myanmar. Forest inventory was conducted in the Oktwin teak bearing forest, the Letpanpin community forest and Alaungdaw Kathapa National Park. According to the Jackknife estimator of species richness, 85 species (${\pm}18.16$), 70 species (${\pm}5.88$) and 186 species (${\pm}17.10$) belonging to 31 families were found in the Oktwin teak bearing forest, 33 families in Letpanpin community forest and 53 families in Alaungdaw Kathapa national park, respectively. Shannon's diversity indices were significantly different among the forests (p<0.05). It ranged from 3.36 to 4.36. Mean tree density (n/ha) of the Oktwin teak bearing forest, Letpanpin community forest and Alaungdaw Kathapa National Park were 488 (${\pm}18.6$), 535 (${\pm}15.6$) and 412 (${\pm}14.1$), while basal areas per hectare were $46.96m^2({\pm}3.23),\;49.01m^2({\pm}5.08)\;and\;60.03m^2({\pm}3.88)$, respectively. At the family level, Verbenaceae, Myrtaceae and Combretaceae occupied the highest importance value index, while at the species level it was Tectona grandis, Lagerstoremia speciosa and Xylia xylocarpa.

• 지능정보연구
• /
• 제16권4호
• /
• pp.67-84
• /
• 2010

미래에 대한 정확한 예측은 경영자, 또는 기업이 수행하는 경영의사결정에 매우 중요한 역할을 한다. 예측만 정확하다면 경영의사결정의 질은 매우 높아질 수 있을 것이다. 하지만 점점 가속화되고 있는 경영 환경의 변화로 말미암아 미래 예측을 정확하게 하는 일은 점점 더 어려워지고 있다. 이에 기업에서는 정확한 예측을 위하여 전문가의 휴리스틱뿐만 아니라 과학적 예측모형을 함께 활용하여 예측의 성과를 높이는 노력을 해 오고 있다. 본 연구는 사례기반추론모형을 예측을 위한 기본 모형으로 설정하고, 데이터 간의 유사도 측정에 퍼지 관계의 개념을 적용함으로써 개선된 예측성과를 얻고자 하였다. 특히, 독립변수 중 기호 데이터 형식의 속성을 가지는 변수들간의 유사도를 측정하기 위해 이진논리의 개념(일치여부의 판단)과 퍼지 관계 및 합성의 개념을 이용하여 도출된 유사도 매트릭스를 사용하였다. 연구 결과, 기호 데이터 형식의 속성을 가지는 변수들 간의 유사도 측정에서 퍼지 관계 및 합성의 개념을 적용하는 방법이 이진논리의 개념을 적용하는 방법과 비교하여 더 우수한 예측정확성을 나타내었다. 그러나 유사도 측정을 위해 다양한 퍼지합성방법(Max-min 합성, Max-product 합성, Max-average 합성)을 적용하여 예측하는 경우에는 예측정확성 측면에서 퍼지 합성방법 간의 통계적인 차이는 유의하지 않았다. 본 연구는 사례기반추론 모형의 구축에서 가장 중요한 유사도 측정에 있어서 퍼지 관계 및 퍼지 합성의 개념을 적용함으로써 유사도 측정 및 적용 방법론을 제시하였다는데 의의가 있다.