• 제목/요약/키워드: automobile networks

검색결과 44건 처리시간 0.023초

입자 군집 최적화를 이용한 FCM 기반 퍼지 모델의 동정 방법론 (Identification Methodology of FCM-based Fuzzy Model Using Particle Swarm Optimization)

  • 오성권;김욱동;박호성;손명희
    • 전기학회논문지
    • /
    • 제60권1호
    • /
    • pp.184-192
    • /
    • 2011
  • In this study, we introduce a identification methodology for FCM-based fuzzy model. The two underlying design mechanisms of such networks involve Fuzzy C-Means (FCM) clustering method and Particle Swarm Optimization(PSO). The proposed algorithm is based on FCM clustering method for efficient processing of data and the optimization of model was carried out using PSO. The premise part of fuzzy rules does not construct as any fixed membership functions such as triangular, gaussian, ellipsoidal because we build up the premise part of fuzzy rules using FCM. As a result, the proposed model can lead to the compact architecture of network. In this study, as the consequence part of fuzzy rules, we are able to use four types of polynomials such as simplified, linear, quadratic, modified quadratic. In addition, a Weighted Least Square Estimation to estimate the coefficients of polynomials, which are the consequent parts of fuzzy model, can decouple each fuzzy rule from the other fuzzy rules. Therefore, a local learning capability and an interpretability of the proposed fuzzy model are improved. Also, the parameters of the proposed fuzzy model such as a fuzzification coefficient of FCM clustering, the number of clusters of FCM clustering, and the polynomial type of the consequent part of fuzzy rules are adjusted using PSO. The proposed model is illustrated with the use of Automobile Miles per Gallon(MPG) and Boston housing called Machine Learning dataset. A comparative analysis reveals that the proposed FCM-based fuzzy model exhibits higher accuracy and superb predictive capability in comparison to some previous models available in the literature.

CMAC 신경회로망을 이용한 가솔린 분사 제어 시스템에 관한 연구 (The injection petrol control system about CMAC neural networks)

  • 한아군;탁한호
    • 한국정보통신학회논문지
    • /
    • 제21권2호
    • /
    • pp.395-400
    • /
    • 2017
  • 본 논문에서는 산소 센서를 이용하여 CMAC 신경회로망 학습제어에 의한 차량의 연료분사 제어방법에 대해 논한다. 기본 차량 내연기관과 연료 분사 제어시스템의 동역학적인 비선형성으로 인하여 불연속적인 연로를 분사한다. 정밀 연료 분사량 제어에 어려움을 발생시키기 때문에 엔진성능은 저하된다. 본 연구에서는 CMAC 신경회로망을 이용한 연료 분사시스템을 제안한다. CMAC 신경회로망은 매우 넓은 범위의 함수로부터 비선형 관계를 학습 할 수 있고, 학습이 빠르며, 수렴 특성을 가지고 있다. 그리고 산소 센서의 출력특성을 파악하여 연료분사 속도를 계산해서 설정된 공연비 값을 유지시켜준다. 게다가 기존 가솔린 엔진의 구조변경이 없이 어떤 상황에서도 공연비를 정밀하게 제어할 수 있으며, 배기가스 배출량을 절감시킬 수 있다. 시뮬레이션을 통해 일반적인 차량의 제어 방법과 비교 분석하였고, 제안된 방법이 차량의 연비 향상과 친환경 성능 등에 더 효과적임을 확인하였다.

탄소계 경질 박막의 연구 및 산업 적용 동향 (Trend in Research and Application of Hard Carbon-based Thin Films)

  • 이경황;박종원;양지훈;정재인
    • 한국표면공학회:학술대회논문집
    • /
    • 한국표면공학회 2009년도 춘계학술대회 논문집
    • /
    • pp.111-112
    • /
    • 2009
  • Diamond-like carbon (DLC) is a convenient term to indicate the compositions of the various forms of amorphous carbon (a-C), tetrahedral amorphous carbon (ta-C), hydrogenated amorphous carbon and tetrahedral amorphous carbon (a-C:H and ta-C:H). The a-C film with disordered graphitic ordering, such as soot, chars, glassy carbon, and evaporated a-C, is shown in the lower left hand corner. If the fraction of sp3 bonding reaches a high degree, such an a-C is denoted as tetrahedral amorphous carbon (ta-C), in order to distinguish it from sp2 a-C [2]. Two hydrocarbon polymers, that is, polyethylene (CH2)n and polyacetylene (CH)n, define the limits of the triangle in the right hand corner beyond which interconnecting C-C networks do not form, and only strait-chain molecules are formed. The DLC films, i.e. a-C, ta-C, a-C:H and ta-C:H, have some extreme properties similar to diamond, such as hardness, elastic modulus and chemical inertness. These films are great advantages for many applications. One of the most important applications of the carbon-based films is the coating for magnetic hard disk recording. The second successful application is wear protective and antireflective films for IR windows. The third application is wear protection of bearings and sliding friction parts. The fourth is precision gages for the automotive industry. Recently, exciting ongoing study [1] tries to deposit a carbon-based protective film on engine parts (e.g. engine cylinders and pistons) taking into account not only low friction and wear, but also self lubricating properties. Reduction of the oil consumption is expected. Currently, for an additional application field, the carbon-based films are extensively studied as excellent candidates for biocompatible films on biomedical implants. The carbon-based films consist of carbon, hydrogen and nitrogen, which are biologically harmless as well as the main elements of human body. Some in vitro and limited in vivo studies on the biological effects of carbon-based films have been studied [$2{\sim}5$].The carbon-based films have great potentials in many fields. However, a few technological issues for carbon-based film are still needed to be studied to improve the applicability. Aisenberg and Chabot [3] firstly prepared an amorphous carbon film on substrates remained at room temperature using a beam of carbon ions produced using argon plasma. Spencer et al. [4] had subsequently developed this field. Many deposition techniques for DLC films have been developed to increase the fraction of sp3 bonding in the films. The a-C films have been prepared by a variety of deposition methods such as ion plating, DC or RF sputtering, RF or DC plasma enhanced chemical vapor deposition (PECVD), electron cyclotron resonance chemical vapor deposition (ECR-CVD), ion implantation, ablation, pulsed laser deposition and cathodic arc deposition, from a variety of carbon target or gaseous sources materials [5]. Sputtering is the most common deposition method for a-C film. Deposited films by these plasma methods, such as plasma enhanced chemical vapor deposition (PECVD) [6], are ranged into the interior of the triangle. Application fields of DLC films investigated from papers. Many papers purposed to apply for tribology due to the carbon-based films of low friction and wear resistance. Figure 1 shows the percentage of DLC research interest for application field. The biggest portion is tribology field. It is occupied 57%. Second, biomedical field hold 14%. Nowadays, biomedical field is took notice in many countries and significantly increased the research papers. DLC films actually applied to many industries in 2005 as shown figure 2. The most applied fields are mold and machinery industries. It took over 50%. The automobile industry is more and more increase application parts. In the near future, automobile industry is expected a big market for DLC coating. Figure 1 Research interests of carbon-based filmsFigure 2 Demand ratio of DLC coating for industry in 2005. In this presentation, I will introduce a trend of carbon-based coating research and applications.

  • PDF

목표물의 고속 탐지 및 인식을 위한 효율적인 신경망 구조 (Effcient Neural Network Architecture for Fat Target Detection and Recognition)

  • 원용관;백용창;이정수
    • 한국정보처리학회논문지
    • /
    • 제4권10호
    • /
    • pp.2461-2469
    • /
    • 1997
  • 목표물 탐지 및 인식은 신경망의 적용이 활발한 하나의 분야로서, 일반적인 형태인식 문제들의 요구 사항에 추가적으로 translation invariance와 실시간 처리를 요구한다. 본 논문에서는 이러한 요구 사항을 만족하는 새로운 신경망의 구조를 소개하고, 이의 효과적인 학습 방법을 설명한다. 제안된 신경망은 특징 추출 단계와 형태 인식 단계가 연속(Cascade)된 가중치 공유 신경망(Shared-weight Neural Network)을 기본으로하여 이를 확장한 형태이다. 이 신경망의 특징 추출 단계는 입력에 가중치 창(weight kernel)으로 코릴레이션 형태의 연산을 수행하며, 신경망 전체를 하나의 2차원 비선형 코릴레이션 필터로 볼 수 있다. 따라서, 신경망의 최종 출력은 목표물 위치에 첨예(peak)값을 갖는 코릴레이션 평면이다. 이 신경망이 갖는 구조는 병렬 또는 분산 처리 컴퓨터로의 구현에 매우 적합하며, 이러한 사실은 실시간 처리가 중요한 요인이 되는 문제에 적용할 수 있음을 의미한다. 목표물과 비목표물간의 숫자상 불균형으로 인하여 초래되는 오경보(false alarm) 발생의 문제를 극복하기 위한 새로운 학습 방법도 소개한다. 성능 검증을 위하여 제안된 신경망을 주차장내에서 이동하는 특정 차량의 탐지 및 인식 문제에 적용하였다. 그 결과 오경보 발생이 없었으며, 중형급 컴퓨터를 이용하여 약 190Km로 이동하는 차량의 추적이 가능한 정도의 빠른 처리 결과를 보여 주었다.

  • PDF