중기 전력수요 예측을 위한 LSTM 최적 모델 선정 = Optimal Model Selection of Long Short-Term Memory for Mid-Term Electric Load Forecasting

다국어 초록 (Multilingual Abstract)

This paper proposes an optimal electric load forecasting technique by applying several activation functions and optimization methods in the internal structure of the Long Short-Term Memory (LSTM) algorithm to forecast mid-term electric load in Korea. As the activation function, the rectified linear unit (ReLU) and hyperbolic tangent (Tanh) are used. Optimization methods such as stochastic gradient descent (SGD) and adaptive moment optimization (Adam) are compared in the paper. In the input data, daily peak load over past seven days, maximum and minimum temperatures are used. In order to increase the efficiency of learning, special weeks and special day load data are replaced. The proposed method shows stable and great performance with small average errors, compared to the RANN based forecasting method which is a previous study. Especially, the estimated error is improved in the summer and winter where the temperature changes significantly.

참고문헌 (Reference)

1 박정도 ; 송경빈, "하계 특수경부하기간의 단기 전력수요예측" 대한전기학회 62 (62): 482-488, 2013

2 정승민, "전력수급계획을 위한 국내 태양광 일사량 데이터 분석" 국제차세대융합기술학회 3 (3): 125-131, 2019

3 공성일 ; 백영식 ; 송경빈 ; 박지호, "온도에 대한 민감도를 고려한 하절기 일 최대전력수요 예측" 대한전기학회 53 (53): 358-364, 2004

4 고상준 ; 윤호영 ; 신동명, "양방향 LSTM을 활용한 전력수요 데이터 예측 기법 연구" 한국소프트웨어감정평가학회 14 (14): 33-40, 2018

5 이기현 ; 곽경일 ; 채우리 ; 고진덕 ; 이주연, "부하예측 및 태양광 발전예측을 통한 ESS 운영방안(Guide-line) 연구" 한국디지털정책학회 18 (18): 267-278, 2020

6 조세원 ; 박래준 ; 송경빈, "기상에 대한 동네예보를 활용한 제주도의 특수일 전력수요예측 알고리즘" 한국조명.전기설비학회 33 (33): 1-8, 2019

7 김시연 ; 정현우 ; 박정도 ; 백승묵 ; 김우선 ; 전경희 ; 송경빈, "계절 ARIMA 모형을 이용한 104주 주간 최대 전력수요예측" 한국조명.전기설비학회 28 (28): 50-56, 2014

8 "The 9th basic Plan for Long-term Electricity Supply and Demand" Ministry of Trade, Industry and Engergy 2020

9 Lee, K.Y, "Short-term load forecasting using an artificial neural network" 7 (7): 124-132, 1992

10 Kim Dohyun, "Short-term load forecasting based on LSTM and CNN" Konkuk University 2019

11 W. Kong, "Short-Term Residential Load Forecasting Based on LSTM Recurrent Neural Network" 10 (10): 841-851, 2019

12 Amanpreet Kaur, "Net load forecasting for high renewable energy penetration grids" 114 : 1073-1084, 2016

13 Baek, S. M., "Mid-term load pattern forecasting with recurrent artificial neural network" 7 : 172830-172838, 2019

14 Bottou, L., "Large-Scale Machine Learning with Stochastic Gradient Descent" 177-186, 2010

15 El Desouky, AA, "Hybrid adaptive techniques for electric-load forecast using ANN and ARIMA" 147 (147): 213-217, 2000

16 Kingma, Diederik P., "Adam : A Method for Stochastic Optimization" 2015

17 Korea Power Exchange, "A Study on Midterm Load Forecasting Technique based on Expert System and its Application" Korea Power Exchange 2014