• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 하계학술대회 논문집 G
• /
• pp.2929-2931
• /
• 1999

In this paper, an adaptive fuzzy inference and HCM(Hard C-Means) clustering method are used for on-line fuzzy modeling of nonlinear and complex system. Here HCM clustering method is utilized for determining the initial parameter of membership function of fuzzy premise rules and also avoiding overflow phenomenon during the identification of consequence parameters. To obtain the on-line model structure of fuzzy systems. we use the recursive least square method for the consequent parameter identification. And the proposed on-line identification algorithm is carried out and is evaluated for sewage treatment process system.

• 제어로봇시스템학회논문지
• /
• 제6권3호
• /
• pp.273-283
• /
• 2000

In this paper we suggest an optimal design method of Fuzzy-Neural Networks(FNN) model for complex and nonlinear systems. The FNNs use the simplified inference as fuzzy inference method and Error Back Propagation Algorithm as learning rule. And we use a HCM(Hard C-Means) Clustering Algorithm to find initial parameters of the membership function. The parameters such as parameters of membership functions learning rates and momentum weighted value is proposed to achieve a sound balance between approximation and generalization abilities of the model. According to selection and adjustment of a weighting factor of an aggregate objective function which depends on the number of data and a certain degree of nonlinearity (distribution of I/O data we show that it is available and effective to design and optimal FNN model structure with a mutual balance and dependency between approximation and generalization abilities. This methodology sheds light on the role and impact of different parameters of the model on its performance (especially the mapping and predicting capabilities of the rule based computing). To evaluate the performance of the proposed model we use the time series data for gas furnace the data of sewage treatment process and traffic route choice process.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• 제13권3호
• /
• pp.278-290
• /
• 1991

Prediction of the sop tissue changes following hard tissue movements is very important in the preoperative analysis of surgical changes of the patient who have orthognathic surgery. This study examined post operative changes of the mid-facial sop tissues in Koreans depending upon two major positional changes of the maxilla following Le Fort - I type osteotomy for orthognathic purposes. Sixty patients(41 males and 19 females) of Koreans were selected and divided into two groups according to directional change of movement of anterior bony structures of the Maxilla as follows : Group I (44 patients) was mooed anteriorly and/or inferiorly, and Group II (16 patients) was mooed posteriorly and/or superioly. Postoperative changes of the sop tissue measurements following hard tissue changes were examined on pre - and post - operative cephalometrics by means of computerized digitation methods and the ratios of changes were analysed. The results were obtained as follows : 1. In Group I, all of the sop tissue measures except the Pn was closely followed by the changes of the hard tissue measures in the horizontal plane, but the Sn and the Cm were only correlated to the vertical changes(p<0.001). In group II, all of the sop tissue measures excluding of the N' and the Pn were significantly correlated to the hard tissue measures in horizontal plane(p<0.001), but the Ls and Stm were only correlated relatively to the vertical changes of the ANS(p<0.01). 2. Predictable ratio of the Sn was 66% of the ANS or 56% of the A in the horizontal plane and 89% of the A in the vertical plane in Group I. In Group II, the Sn was predictable as 85% of the ANS or 70% of the A in the horizontal plane but was not predictable in vertical plane. 3. Predictable ratio of the Cm was 28% of the ANS or 50% of the A in the horizontal plane and 56% of the ANS or 36% of the A in the vertical plane in Group I. In Group II, the Cm was predictable horizontally as 74% of the A. Predictable ratio of the Pn was 30% of the ANS or 38% of the A in horizontal plane in Group I, but it was not predictable both horizontally and vertically in Group II. 4. Predictable ratio of the Ls was 52% of the Pr in Group I and 77% in Group II in the horizontal plane. The Stm was predictable as 34% of the pr or 22% of the I in the horizontal plane in Group I, and was also predictable as 55% of the pr or 68% of the I horizontally and 21% of the pr or 65% of the I vertically in Group II. 5. All ratios of change in the thickness. length and area of the upper lip following maxillary movement were statiscally correlated, however, mangitudes of them were meaningful clinically.

• 대한소아치과학회지
• /
• 제27권4호
• /
• pp.558-563
• /
• 2000

1960년대 최초로 Ruby 레이저의 개발 이래로, 치과용 드릴인 고속용 핸드피스를 대신해 치아 경조직 삭제에 레이저의 적용이 시도되어져 오고 있는데, 최근 Er:YAG레이저의 개발로 경조직 삭제시 이들의 삭제 효능과 치질의 온도상승에 관한 연구에 관심이 증대되었다. 과거 여러 종류의 레이저에 비해 2.94um Er:YAG 레이저는 파장의 특성이 치질의 수분과 하이드록시아파타이트에 매우 흡수율이 높아 치질내 수분이 고열에 의해 증기화되어 폭발하므로써 치질이 제거되므로 crack이나 탄화된 부위가 없이 와동 변연부가 명확하고 깨끗하며 효과적인 치질의 삭제가 가능하다. 삭제시 치질내 온도 상승이 레이저의 경조직 적용시 고려되는 중요한 문제점으로 대두되나 이는 물분사와 적적한 레이저 parameter의 사용으로 열 발생을 최소화하면 치수 손상이 발생하지 않는다고 보고 된다. 본증례는 치수에 근접하지 않은 우식 병소의 처치를 주소로 내원한 소아 환자들에 있어서 Er:YAG 레이저 적용시 기존의 고속용 삭제 기구에 비해 소음과 진동이 적어 소아 환자들의 치과 치료에 대한 공포감이나 동통을 최소로 하며 적절한 레이저 parameter와 물분사의 사용으로 치수에 대한 열자극이나 치질의 변색과 같은 임상적인 문제를 초래하지 않고 효과적으로 우식 병소의 제거 및 와동 형성을 시행할 수 있었다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2001년도 합동 추계학술대회 논문집 정보 및 제어부문
• /
• pp.420-423
• /
• 2001

In this paper, the Multi-FNNs(Fuzzy-Neural Networks) architecture is identified and optimized using HCM(Hard C-Means) clustering method and genetic algorithms. The proposed Multi-FNNs architecture uses simplified inference and linear inference as fuzzy inference method and error back propagation algorithm as learning rules. Here, HCM clustering method, which is carried out for the process data preprocessing of system modeling, is utilized to determine the structure of Multi-FNNs according to the divisions of input-output space using I/O process data. Also, the parameters of Multi-FNNs model such as apexes of membership function, learning rates and momentum coefficients are adjusted using genetic algorithms. An aggregate performance index with a weighting factor is used to achieve a sound balance between approximation and generalization abilities of the model. To evaluate the performance of the proposed model we use the time series data for gas furnace and the NOx emission process data of gas turbine power plant.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2003년도 하계학술대회 논문집 A
• /
• pp.552-554
• /
• 2003

The second harmonic component is commonly used for blocking differential relay in power transformers. However, it is difficult to distinguish between inrush and internal winding fault with differential current protective relaying. This paper proposed a new method using Nuro-Fuzzy System based on HCM(Hard C-Means). The proposed system is more objective and systematic than existing model. The data used in input are 3-phase primary voltage and fundamental harmonic of differential current. Various states of transformer are simulated using BCTRAN and HYSDAT of EMTP. As a result of the application of algorithm in various cases, the exact discrimination between internal winding fault and inrush is performed.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2014년도 추계학술발표대회
• /
• pp.894-897
• /
• 2014

The speed of processes and the extremely large amount of data to be used in Intelligence Transportations System (ITS) cannot be handling by humans without considerable automation. However, it is difficult to develop software with conventional fixed algorithms (hard-wired logic on decision making level) for effectively manipulate dynamically evolving real time transportation environment. This situation can be resolved by applying methods of artificial intelligence and data mining that provide flexibility and learning capability. This paper presents a brief introduction of data mining and artificial intelligence (AI) applications in Intelligence Transportation System (ITS), analyzing the prospects of enhancing the capabilities by means of knowledge discovery and accumulating intelligence to support in decision making.

• 한국산학기술학회논문지
• /
• 제13권11호
• /
• pp.5379-5388
• /
• 2012

비선형 공정에 대한 퍼지 모델링에서, 퍼지 규칙은 일반적으로 입력 변수 선택, 공간 분할 수 및 소속 함수에 의해 형성된다. 비선형 공정에 대한 퍼지 규칙의 생성은 차원이 증가할수록 규칙의 수가 지수적으로 증가하는 문제를 가지고 있다. 이를 해결하기 위해, 입력 공간의 퍼지 분할에 의한 퍼지 규칙을 생성함으로써 복잡한 비선형 공정을 모델링 할 수 있다. 따라서 본 논문에서는 HCM 클러스터링 알고리즘을 이용하여 입력 공간을 분산 형태로 분할함으로써 비퍼지 추론 시스템의 규칙을 생성한다. 규칙의 전반부 파라미터는 HCM 클러스터링 알고리즘에 의한 소속행렬로 결정된다. 규칙의 후반부는 다항식 함수의 형태로 표현되며, 각 규칙의 후반부 파라미터들은 표준 최소자승법에 의해 동정된다. 마지막으로, 비선형 공정으로는 널리 이용되는 데이터를 이용하여 비선형 특성 및 성능을 평가한다. 본 실험을 통해 고차원의 비선형 시스템은 매우 적은 수의 규칙을 가지고 모델링할 수 있었다.

• Composites Research
• /
• 제37권1호
• /
• pp.46-52
• /
• 2024

운송 수단의 발전은 인간의 교통 편의 증진과 더불어 이동이 불편한 장애인들의 이동 반경 확대를 가능하게 하였다. 그러나 휠체어 탑재 차량의 경우 차량 사고 시 발생할 수 있는 안전성은 일반 승객 좌석에 비해 여전히 낮다. 특히 무방비 상태에서 발생할 수 있는 후방 추돌 사고의 경우 장애인 탑승객의 목 부상에 치명적으로 작용할 수 있다. 따라서 휠체어 탑재 차량에 적용될 headrest에는 보다 세밀한 설계안이 반영되어야 한다. 본 연구에서는 휠체어 운송 차량의 저속 후방 추돌 시 headrest의 국부적 압축 특성 분포 구현을 위해 multi-cell headrest가 제안되었다. 이후 해석을 통한 데이터셋 구축과 k-means clustering을 적용한 군집화 결과를 이용해 탑승객의 목 상해지수와 충격 에너지 흡수량 간 상관관계 분석이 수행되었다. 군집화 결과 유사한 특성을 지닌 데이터 군집이 형성된 것을 확인하였으며, 각 군집의 특성을 통한 목 상해지수와 충격 에너지 흡수량 간의 상관관계 분석이 수행되었다. 분석 결과 Mid3와 Mid6에서의 cell 압축 특성이 soft할수록 충격 에너지 흡수량이 증가하는 것을 확인하였으며, Front2, Mid3, Mid6에서의 cell 압축 특성이 hard할수록 목 상해지수 감소에 효과적임을 확인하였다.

• 대한치과보철학회지
• /
• 제34권3호
• /
• pp.539-573
• /
• 1996

This investigation was designed to determine the effectiveness of the posterior occlusal schemes on masticatory activity during mastication in complete denture. Twelve edentulous subjects were selected for this study. All subjects had no past history and no functional abnormality on masticatory system and TMjoint. And, they had residual ridge of favorable morphology, firm mucosa and Class I skeletal jaw relationship, Twelve experimental denture with interchangeable occlusions(0-degree teeth, 30-degree teeth, Levin teeth and S-A teeth) were constructed for this study. The masticatory performance was analyzed by means of standard sieve(10, 16, 20, 30sieve), and the electrical activity from selected muscles(Temporalis and Masseter muscle) was recorded simultaneously with electromyography(Bio-Pak system) as the subject masticated test foods (rice, peanut and gum) with four different occlusal schemes. Mandibular movement was, also, measured with Sirognathography(Bio-Pak system). These recordings were performed in immediately, after 1 week and after 2 weeks of insertion of complete denture. The results were as follows; 1. The average masticatory performance of 0-degree artificial teeth was higher than any other artificial teeth. 2. Masticatory performance in denture wearer was affected preferentially by food and artificial occlusal schemes. 3. During chewing, there was a statistical difference of EMG activity between masseter and temporal muscle(p<0.01). Especially, EMG activity of working masticatory muscle was highly affected by food rather than by artificial occlusal schemes. 4. In denture wearer, the velocity of opening was not affected by food, whereas, the velocity of closing was faster in soft food chewing than in hard food chewing, and the amount of vertical displacement was grater in chewing of soft and large bolus than in chewing of hard and small bolus. However, the amount of lateral displacement showed conversely (p<0.05). 5. It was considered that masticatory performance in denture wearer is not affected by the condition of residual ridge, the history of denture wear, the preference, the adaptation to artificial teeth and the total mesiodistal length of artificial posterior teeth.