RISS 학술연구정보서비스

검색
다국어 입력

http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.

변환된 중국어를 복사하여 사용하시면 됩니다.

예시)
  • 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
  • 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
닫기
    인기검색어 순위 펼치기

    RISS 인기검색어

      조선소 최대전력수요 예측을 위한 알고리즘 비교분석 및 최적기법 도출 = Comparative Analysis of Algorithms and Optimal Model Selection for Shipyard Peak Power Demand Forecasting

      한글로보기

      https://www.riss.kr/link?id=T17368039

      • 0

        상세조회
      • 0

        다운로드
      서지정보 열기
      • 내보내기
      • 내책장담기
      • 공유하기
      • 오류접수

      부가정보

      다국어 초록 (Multilingual Abstract) kakao i 다국어 번역

      With the rise of Industry 4.0 and ESG management, energy efficiency is vital for the shipbuilding industry. In Korea, industrial electricity costs heavily depend on "Peak Power" demand, where the highest daily peak recorded during specific seasons determines the base rate for the entire year. Consequently, accurate daily forecasting is crucial for cost reduction, yet existing empirical methods fail to capture the non-linear fluctuations of shipyard power consumption.
      This study identifies the optimal forecasting model and proposes an ensemble approach using daily shipyard operating data from 2016 to 2023. The dataset integrates power consumption, production processes, and meteorological factors. Twelve models across five categories—RNN, Transformer, MLP, Gradient Boosting, and Koopa—were evaluated using MAE, RMSE, and R2 scores.
      Experimental results revealed that the N-HiTS (Neural Hierarchical Interpolation for Time Series) model achieved superior performance among single models (R2 0.8775, MAE 1,173kW). N-HiTS successfully captured multi-scale seasonality through hierarchical interpolation, outperforming complex Transformer-based models like TFT (R2 0.765) and traditional RNNs.
      To enhance stability, ensemble models were constructed using N-HiTS as the base learner. Statistical verification (t-test) indicated that the LSTM Meta Learner provided the most statistically significant improvement (p-value < 0.05), achieving an MAE of 1,243kW and R2 of 0.8653. This research validates the N-HiTS architecture and the LSTM ensemble as robust solutions for precise electricity cost management in smart shipyards.
      번역하기

      With the rise of Industry 4.0 and ESG management, energy efficiency is vital for the shipbuilding industry. In Korea, industrial electricity costs heavily depend on "Peak Power" demand, where the highest daily peak recorded during specific seasons det...

      With the rise of Industry 4.0 and ESG management, energy efficiency is vital for the shipbuilding industry. In Korea, industrial electricity costs heavily depend on "Peak Power" demand, where the highest daily peak recorded during specific seasons determines the base rate for the entire year. Consequently, accurate daily forecasting is crucial for cost reduction, yet existing empirical methods fail to capture the non-linear fluctuations of shipyard power consumption.
      This study identifies the optimal forecasting model and proposes an ensemble approach using daily shipyard operating data from 2016 to 2023. The dataset integrates power consumption, production processes, and meteorological factors. Twelve models across five categories—RNN, Transformer, MLP, Gradient Boosting, and Koopa—were evaluated using MAE, RMSE, and R2 scores.
      Experimental results revealed that the N-HiTS (Neural Hierarchical Interpolation for Time Series) model achieved superior performance among single models (R2 0.8775, MAE 1,173kW). N-HiTS successfully captured multi-scale seasonality through hierarchical interpolation, outperforming complex Transformer-based models like TFT (R2 0.765) and traditional RNNs.
      To enhance stability, ensemble models were constructed using N-HiTS as the base learner. Statistical verification (t-test) indicated that the LSTM Meta Learner provided the most statistically significant improvement (p-value < 0.05), achieving an MAE of 1,243kW and R2 of 0.8653. This research validates the N-HiTS architecture and the LSTM ensemble as robust solutions for precise electricity cost management in smart shipyards.

      더보기

      목차 (Table of Contents)

      • I. 서 론 1
      • 1. 연구 배경 및 목적 1
      • II. 배경 이론 4
      • 1. 머신러닝/딥러닝 이론적 배경 4
      • I. 서 론 1
      • 1. 연구 배경 및 목적 1
      • II. 배경 이론 4
      • 1. 머신러닝/딥러닝 이론적 배경 4
      • 1) RNN(순환 신경망) 기반 모델 4
      • 2) Transformer 기반 모델 7
      • 3) 다중 퍼셉트론(MLP) 기반 모델 12
      • 4) Gradient Boosting 모델 15
      • 5) 기타 시계열 모델 17
      • 2. 기존 문헌 고찰 18
      • 1) 산업용 및 선박 전력 수요 예측 연구 18
      • 2) 딥러닝(Deep Learning) 기반 시계열 예측 모델 21
      • 3) Transformer 기반 시계열 예측 모델 24
      • 4) MLP(Multi-layer Perceptron) 기반 시계열 예측 모델 28
      • 5) Gradient Boosting 기반 모델 30
      • 6) 기타 시계열 모델 31
      • 7) 비교 연구 및 벤치마킹 33
      • III. 본 론 35
      • 1. 연구 방법 35
      • 1) 방법론 및 절차 35
      • 2) 조선소 데이터 개요 및 전처리 36
      • 3) 실험 설계 38
      • 2. 탐색적 데이터 분석(EDA) 40
      • 1) 데이터셋 개요 40
      • 2) 기술 통계 분석 41
      • 3) 계절적 변동성 분석 42
      • 4) 공휴일 효과 42
      • 5) 상관 변수 42
      • 3. 단일 모델 실험 결과 44
      • 1) 모델 분류 및 평가 개요 44
      • 2) 순환 신경망(RNN) 계열 46
      • 3) Transformer 기반 계열 47
      • 4) MLP 기반 계열 48
      • 5) Gradient Boosting 기반 계열 49
      • 6) 기타 시계열 기반 51
      • 7) 계열별 성능 비교분석 51
      • 8) 단일 우수 모델: N-HiTS 53
      • 9) 한계점 및 개선 방향 54
      • 4. Ensemble 모델 실험 결과 55
      • 1) Ensemble 기법 개요 55
      • 2) Ensemble 모델 구성 55
      • 3) 실험 결과 및 분석 61
      • IV. 결론 및 향후 연구방향 63
      • 참 고 문 헌 66
      • ABSTRACT 70
      • 감사의 글 72
      더보기

      분석정보

      View

      상세정보조회

      0

      Usage

      원문다운로드

      0

      대출신청

      0

      복사신청

      0

      EDDS신청

      0

      동일 주제 내 활용도 TOP

      더보기

      주제

      연도별 연구동향

      연도별 활용동향

      연관논문

      연구자 네트워크맵

      공동연구자 (7)

      유사연구자 (20) 활용도상위20명

      이 자료와 함께 이용한 RISS 자료

      나만을 위한 추천자료

      해외이동버튼