LSTM을 활용한 일반주택의 실내온도 예측 사례 연구
A Case Study on the Prediction of Indoor Temperature in General House Using LSTM
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
바닥 난방 시스템을 사용하는 일반주택에서, 급격한 외부 기상상태의 변화 및 주택과 보일러 특성에 따른 열 이동 등 다양한 변수를 고려하여 실내쾌적성을 유지함과 동시에 난방 에너지소비량을 최소화하는 것이 중요하다. 본 논문은 실제 사용자의 보일러 운영정보와 기상정보 및 LSTM 딥러닝 모델을 활용하여 일반주택의 실내온도 예측을 수행하였다. 시간에 따른 보일러 운영정보 및 기상정보 중 사용자의 실내온도와 상관관계가 있는 변수를 학습 인자로 선정하였다. 이후, LSTM 베이스라인을 구축하여 밀집층의 노드 개수, 드롭아웃 비율, L2 정규화 파라미터, 배치사이즈, 학습 시간 대역에 따른 성능 변화를 관찰하여 최적화를 수행하였다. 학습 파라미터들의 최적화를 수행한 결과, MAE는 0.3780, RMSE는 0.5306으로 직관적으로 평균 오차 0.38℃에 해당하는 우수한 성능을 확인하였다.
Abstract
In general houses using floor heating systems it is important to maintain indoor comfort and minimize heating energy consumption by considering various variables such as rapid changes in external weather conditions and heat movement according to housing and boiler characteristics. This paper performed indoor temperature prediction of general housing using actual user boiler operation information, weather information, and LSTM deep learning model. Among boiler operation information and weather information over time, a variable correlated with the user's indoor temperature was selected as a learning factor. After that, the LSTM baseline was established to perform optimization by observing performance changes according to the number of nodes in the dense layer, drop-out ratio, L2 regularization parameter, batch size, and learning time band. As a result of performing the optimization of the learning parameters, the MAE was 0.3780 and the RMSE was 0.5306, intuitively confirming excellent performance corresponding to the average error of 0.38°C.
참고문헌 (23)
- Chen, T., "Application of adaptive predictive control to a floor heating system with a large thermal lag", Energy and Buildings, 34(1), pp.45~51 (2002)
- Hong, G., "Development of a data-driven predictive model of supply air temperature in an air-handling unit for conserving energy", Energies, 11(2), pp.407~ (2018)
- Afroz, Z., "Real-time prediction model for indoor temperature in a commercial building", Applied energy, 231, pp.29~53 (2018)
- Paul, D., "IoT and machine learning based prediction of smart building indoor temperature", 4th Intl. Conf. on Computer and Information Sciences (ICCOINS) (2018)
- Mba, L., "Application of artificial neural network for predicting hourly indoor air temperature and relative humidity in modern building in humid region", Energy and Buildings, 121 (2016)
- Li, X., "Modeling for indoor temperature prediction based on time-delay and Elman neural network in air conditioning system", Building Engineering, 33, pp.101854~ (2021)
- Xu, C., "Improving prediction performance for indoor temperature in public buildings based on a novel deep learning method", Building and Environment, 148 (2019)
- Tsoumalis, G. I., "Minimization of natural gas consumption of domestic boilers with convolutional, long-short term memory neural networks and genetic algorithm", Applied energy, 299, pp.117256~ (2021)
- Lee H. W., "A Prediction of indoor temperature on a normal house using LSTM", Proc. of Korea Computer Congress, 520-522 (2022)
- Patro, S., "Normalization:A preprocessing stage" (2015)
- Hochreiter, S., "The vanishing gradient problem during learning recurrent neural nets and problem solutions", Fuzziness and Knowledge Based Systems, 6(02), pp.107~116 (1998)
- "LSTM Structure Diagram"[URL]
- Hochreiter, S., "Long short-term memory", Neural computation, 9(8), pp.1735~1780 (1997)
- Prechelt, L., "Neural Networks: Tricks of the trade", pp.55~69 (1998), Springer
- Willmott, C. J., "Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance", Climate research, 30(1), pp.79~82 (2005)
- Baldi, Pierre, "Understanding dropout", Advances in neural information processing systems, 26 (2013)
- "Visualization of Drop Out"[URL]
- Zaremba, Wojciech, "Recurrent neural network regularization" (2014)
- Cortes, Corinna, "L2 regularization for learning kernels" (2012)
- "Concept Diagram of Batch size"[URL]
- 이정근, "LSTM 오토인코더를 이용한 라디에이터 고장진단 사례연구", 한국차세대컴퓨팅학회 논문지, 16(6), pp.17~25 (2020), 한국차세대컴퓨팅학회[DOI]
- 배한준, "실제 이동에 따른 자기장 랜덤성을 반영한 LSTM 기반 실내 위치 인식 시스템", 한국차세대컴퓨팅학회 논문지, 17(3), pp.30~38 (2021), 한국차세대컴퓨팅학회[DOI]
- 이선미, "그린 데이터센터를 위한 스마트 에너지", 탄소 관리시스템의 개발, 2012, pp.8(2), 62~72[DOI]