• Title/Summary/Keyword: Slot Filling

Search Result 14, Processing Time 0.02 seconds

Analysis of Slot Leakage Reactance of Submersible Motor with Closed Slots during Starting Transient Operation

  • Bao, Xiaohua;Di, Chong;Fang, Yong
    • Journal of Electrical Engineering and Technology
    • /
    • v.11 no.1
    • /
    • pp.135-142
    • /
    • 2016
  • Generally, closed slots are adopted to reduce the water friction loss in both the stator and the rotor of water filling submersible motor due to the special environment of operation. One of the obvious differences between the traditional induction motors and water filling submersible motors is that the submersible motors only need relatively smaller starting torque. This paper aims to analyze the slot leakage reactance of water filling submersible motor during starting transient operation. An improved analytical method which considered the magnetic saturation of the slot bridge and the skin effect of rotor bars is proposed. The slot permeance factor which has a direct impact on the slot leakage reactance is calculated. Then finite element models with different stator slot types are constructed and search coils are introduced to measure the slot flux linkage. Moreover, the starting performances of the models with two typical stator slots are compared and the flux leakage characteristics are obtained. Finally, the results obtained by finite element method are very close to the results obtained by analytical method.

QUANTITATIVE ANALYSIS OF MARGINAL MICROLEAKAGE IN VARIOUS RETROGRADE FILLING MATERIALS AND PREPARATION TYPES (역행충전시 수복재와 와동 형태에 따른 변연누출의 정량적 분석)

  • Han, Chung-Kyeung;Yang, Hong-So
    • Restorative Dentistry and Endodontics
    • /
    • v.15 no.1
    • /
    • pp.97-105
    • /
    • 1990
  • When conventional root canal treatment is failed or contraindicated, retrograde root canal filling following apicoectomy is a valuable procedure, aimed at hermetically sealing the root canal against leakage of irritants from the canal into the periapical tissue. In this in vitro investigation, to analyze apical microleakage electrochemically in teeth with different retrograde filling materials and preparation types, single - rooted tooth was cut 2mm from the apex and each Class I and Slot preparation was prepared. Experimental groups : Group 1. Amalgam filling with cavity varnish in Class I preparation Group 2. Scotchbond 2+Silux filling in Class I preparation Group 3. Gutta percha filling with ZOE cement in Class I preparation Group 4. Amalgam filling with cavity varnish in Slot preparation Group 5. Scotchbond 2+Silux filling in Slot preparation Each specimens was immersed in 1% solution of KCl, and applied a potential of 9V external power supply. Measurements of the current flow were obtained at 1, 2, 3, 7, 9, 12, 14, 18, 21, 25 and 28 day after immerson. Marginal microleakage were compared and evaluated. The results were as follows ; 1. The group filled with composite resin with dentin bonding agent shows lower apical microleakage value than the group filled with amalgam following varnish application (P<0.01). 2. In the group filled with gutta percha, apical microleakage value was the hightest 3. There was no significant difference between Class I cavity and Slot type cavity regardless of the used materials.

  • PDF

Development of Korean dataset for joint intent classification and slot filling (발화 의도 예측 및 슬롯 채우기 복합 처리를 위한 한국어 데이터셋 개발)

  • Han, Seunggyu;Lim, Heuiseok
    • Journal of the Korea Convergence Society
    • /
    • v.12 no.1
    • /
    • pp.57-63
    • /
    • 2021
  • Spoken language understanding, which aims to understand utterance as naturally as human would, are mostly focused on English language. In this paper, we construct a Korean language dataset for spoken language understanding, which is based on a conversational corpus between reservation system and its user. The domain of conversation is limited to restaurant reservation. There are 7 types of slot tags and 5 types of intent tags in 6857 sentences. When a model proposed in English-based research is trained with our dataset, intent classification accuracy decreased a little, while slot filling F1 score decreased significantly.

Development of ordering chatbot that can process multiple keywords based on recursive slot-filling method (빈칸 되묻기 방식 기반 다중 키워드 처리가 가능한 주문용 챗봇 개발)

  • Choi, Hyeon-Jun;Bae, Seung-Ju;Jeong, Gu-Min
    • The Journal of Korea Institute of Information, Electronics, and Communication Technology
    • /
    • v.12 no.4
    • /
    • pp.440-448
    • /
    • 2019
  • In this paper, we propose an ordering chatbot that can process multiple keywords based on recursive slot-filling method. In general, in case of an order service using chatbots, the whole order process is performed only according to the sequence defined by the developer. That is, among all the information needed for the whole order process, only one input can be processed at one time. In order to reduce processing step for the order, we propose a recursive slot-filling method which fills out multiple slots per one time by extracting multiple keywords. First, a keyword array for the order is created according to the order related information. Next, from the input sentence of a user, multiple keywords is extracted. Corresponding slots for a keyword array will be filled with the extracted keywords. Finally, recursive routine will be executed to fill out all the blank in the keyword array. The usability and validity of the proposed method will be shown from the implementation of a smartphone application.

Adversarial Learning for Natural Language Understanding (자연어 이해를 위한 적대 학습 방법)

  • Lee, Dong-Yub;Whang, Tae-Sun;Lee, Chan-Hee;Lim, Heui-Seok
    • Annual Conference on Human and Language Technology
    • /
    • 2018.10a
    • /
    • pp.155-159
    • /
    • 2018
  • 최근 화두가 되고있는 지능형 개인 비서 시스템에서 자연어 이해(NLU) 시스템은 중요한 구성요소이다. 자연어 이해 시스템은 사용자의 발화로부터 대화의 도메인(domain), 의도(intent), 의미적 슬롯(semantic slot)을 분류하는 역할을 한다. 하지만 자연어 이해 시스템을 학습하기 위해서는 많은 양의 라벨링 된 데이터를 필요로 하며 새로운 도메인으로 시스템을 확장할 때, 새롭게 데이터 라벨링을 진행해야 하는 한계점이 존재한다. 이를 해결하기 위해 본 연구는 적대 학습 방법을 이용하여 풍부한 양으로 구성된 기존(source) 도메인의 데이터부터 적은 양으로 라벨링 된 데이터로 구성된 대상(target) 도메인을 위한 슬롯 채우기(slot filling) 모델 학습 방법을 제안한다. 실험 결과 적대 학습을 적용할 경우, 적대 학습을 적용하지 않은 경우 보다 높은 f-1 score를 나타냄을 확인하였다.

  • PDF

Quality Evaluations of Induction Motor Rotors during Die Casting Process (유도전동기 회전자 금형주조 시 품질평가)

  • Park, Sang-Chul
    • Journal of the Korea Academia-Industrial cooperation Society
    • /
    • v.19 no.10
    • /
    • pp.115-120
    • /
    • 2018
  • This study examined the cast quality of small-sized induction motor rotors during the die casting process. Numerical analyses with 3-dimensional half models were performed to investigate the filling patterns of aluminum molten metals into a mold after high-speed injections. The following were obtained from numerical analyses and experimental results. First, molten metals started to fill the lower end ring, then moved horizontally to fill the core slot and upper end ring, and finally stopped to fill the rotor core slot. Second, circulation of molten metals occurred at the lower end ring, resulting considerable porosity at the section of lower end ring from the experimental results. Third, further work for obtaining sound quality of rotor core cast is required to develop a new shape of the rotor core cast or improve the die casting conditions.

Quality Evaluations of Induction Motor Rotors during Die Casting Process II (유도전동기 회전자 금형주조 시 품질평가 II)

  • Park, Sang-Chul
    • Journal of the Korea Academia-Industrial cooperation Society
    • /
    • v.20 no.7
    • /
    • pp.347-352
    • /
    • 2019
  • This study focused on improving the cast quality of small-sized induction motor rotors during the die casting process. A new rotor core cast model was proposed based on previous research results and parametric studies. Numerical analyses using 3-dimensional half models were performed to evaluate the filling patterns of aluminum molten metals into a mold and on-site experiment performed to verify the newly proposed cast model. The following were obtained from numerical filling analyses and experimental results. First, molten metals started to fill the lower end ring, then moved on to fill the core slot and upper end ring and finally stopped to fill at the rotor core slot. Second, significant circulation of molten metals was not observed on the lower end ring, resulting in fewer defects at the section of the lower end ring from the experimental results. Third, the new shape of a rotor core cast was effective in producing rotors with sound cast quality, and reducing the end ring cast defect area by approximately 70 %.

Syllable-based Korean Named Entity Recognition and Slot Filling with ELECTRA (ELECTRA 모델을 이용한 음절 기반 한국어 개체명 인식과 슬롯 필링)

  • Do, Soojong;Park, Cheoneum;Lee, Cheongjae;Han, Kyuyeol;Lee, Mirye
    • Annual Conference on Human and Language Technology
    • /
    • 2020.10a
    • /
    • pp.337-342
    • /
    • 2020
  • 음절 기반 모델은 음절 하나가 모델의 입력이 되며, 형태소 분석을 기반으로 하는 모델에서 발생하는 에러 전파(error propagation)와 미등록어 문제를 회피할 수 있다. 개체명 인식은 주어진 문장에서 고유한 의미를 갖는 단어를 찾아 개체 범주로 분류하는 자연어처리 태스크이며, 슬롯 필링(slot filling)은 문장 안에서 의미 정보를 추출하는 자연어이해 태스크이다. 본 논문에서는 자동차 도메인 슬롯 필링 데이터셋을 구축하며, 음절 단위로 한국어 개체명 인식과 슬롯 필링을 수행하고, 성능 향상을 위하여 한국어 대용량 코퍼스를 음절 단위로 사전학습한 ELECTRA 모델 기반 학습방법을 제안한다. 실험 결과, 국립국어원 문어체 개체명 데이터셋에서 F1 88.93%, ETRI 데이터셋에서는 F1 94.85%, 자동차 도메인 슬롯 필링에서는 F1 94.74%로 우수한 성능을 보였다. 이에 따라, 본 논문에서 제안한 방법이 의미있음을 알 수 있다.

  • PDF

Multitask Transformer Model-based Fintech Customer Service Chatbot NLU System with DECO-LGG SSP-based Data (DECO-LGG 반자동 증강 학습데이터 활용 멀티태스크 트랜스포머 모델 기반 핀테크 CS 챗봇 NLU 시스템)

  • Yoo, Gwang-Hoon;Hwang, Chang-Hoe;Yoon, Jeong-Woo;Nam, Jee-Sun
    • Annual Conference on Human and Language Technology
    • /
    • 2021.10a
    • /
    • pp.461-466
    • /
    • 2021
  • 본 연구에서는 DECO(Dictionnaire Electronique du COreen) 한국어 전자사전과 LGG(Local-Grammar Graph)에 기반한 반자동 언어데이터 증강(Semi-automatic Symbolic Propagation: SSP) 방식에 입각하여, 핀테크 분야의 CS(Customer Service) 챗봇 NLU(Natural Language Understanding)을 위한 주석 학습 데이터를 효과적으로 생성하고, 이를 기반으로 RASA 오픈 소스에서 제공하는 DIET(Dual Intent and Entity Transformer) 아키텍처를 활용하여 핀테크 CS 챗봇 NLU 시스템을 구현하였다. 실 데이터을 통해 확인된 핀테크 분야의 32가지의 토픽 유형 및 38가지의 핵심 이벤트와 10가지 담화소 구성에 따라, DECO-LGG 데이터 생성 모듈은 질의 및 불만 화행에 대한 양질의 주석 학습 데이터를 효과적으로 생성하며, 이를 의도 분류 및 Slot-filling을 위한 개체명 인식을 종합적으로 처리하는 End to End 방식의 멀티태스크 트랜스포머 모델 DIET로 학습함으로써 DIET-only F1-score 0.931(Intent)/0.865(Slot/Entity), DIET+KoBERT F1-score 0.951(Intent)/0.901(Slot/Entity)의 성능을 확인하였으며, DECO-LGG 기반의 SSP 생성 데이터의 학습 데이터로서의 효과성과 함께 KoBERT에 기반한 DIET 모델 성능의 우수성을 입증하였다.

  • PDF

Development of Artificial Intelligence-based Legal Counseling Chatbot System

  • Park, Koo-Rack
    • Journal of the Korea Society of Computer and Information
    • /
    • v.26 no.3
    • /
    • pp.29-34
    • /
    • 2021
  • With the advent of the 4th industrial revolution era, IT technology is creating new services that have not existed by converging with various existing industries and fields. In particular, in the field of artificial intelligence, chatbots and the latest technologies have developed dramatically with the development of natural language processing technology, and various business processes are processed through chatbots. This study is a study on a system that provides a close answer to the question the user wants to find by creating a structural form for legal inquiries through Slot Filling-based chatbot technology, and inputting a predetermined type of question. Using the proposal system, it is possible to construct question-and-answer data in a more structured form of legal information, which is unstructured data in text form. In addition, by managing the accumulated Q&A data through a big data storage system such as Apache Hive and recycling the data for learning, the reliability of the response can be expected to continuously improve.