에너지 인터넷을 위한 GRU기반 전력사용량 예측
Prediction of Power Consumptions Based on Gated Recurrent Unit for Internet of Energy
0
조회
0
다운로드
4
인용
0
참고문헌
키워드
초록
최근 에너지 인터넷에서 지능형 원격검침 인프라를 이용하여 확보된 대량의 전력사용데이터를 기반으로 효과적인 전력수요 예측을 위해 다양한 기계학습기법에 관한 연구가 활발히 진행되고 있다. 본 연구에서는 전력량 데이터와 같은 시계열 데이터에 대해 효율적으로 패턴인식을 수행하는 인공지능 네트워크인 Gated Recurrent Unit(GRU)을 기반으로 딥 러닝 모델을제안하고, 실제 가정의 전력사용량 데이터를 토대로 예측 성능을 분석한다. 제안한 학습 모델의 예측 성능과 기존의 LongShort Term Memory (LSTM) 인공지능 네트워크 기반의 전력량 예측 성능을 비교하며, 성능평가 지표로써 Mean SquaredError (MSE), Mean Absolute Error (MAE), Forecast Skill Score, Normalized Root Mean Squared Error (RMSE),Normalized Mean Bias Error (NMBE)를 이용한다. 실험 결과에서 GRU기반의 제안한 시계열 데이터 예측 모델의 전력량 수요 예측 성능이 개선되는 것을 확인한다.
Abstract
Recently, accurate prediction of power consumption based on machine learning techniques in Internet of Energy (IoE)has been actively studied using the large amount of electricity data acquired from advanced metering infrastructure(AMI). In this paper, we propose a deep learning model based on Gated Recurrent Unit (GRU) as an artificialintelligence (AI) network that can effectively perform pattern recognition of time series data such as the powerconsumption, and analyze performance of the prediction based on real household power usage data. In the performanceanalysis, performance comparison between the proposed GRU-based learning model and the conventional learning modelof Long Short Term Memory (LSTM) is described. In the simulation results, mean squared error (MSE), mean absoluteerror (MAE), forecast skill score, normalized root mean square error (RMSE), and normalized mean bias error (NMBE)are used as performance evaluation indexes, and we confirm that the performance of the prediction of the proposedGRU-based learning model is greatly improved.
참고문헌 (14)
- Y. Cheng, "PowerLSTM : Power demand forecasting using long short-term memory neural network", Proc. of Advanced Data Mining and Applications: 13th International Conference, pp.727~740 (2017)
- J. Y. Choi, "Combining LSTM network ensemble via adaptive weighting for improved time series forecasting", Mathematical Problems in Engineering, 2018, pp.1~8 (2018)[DOI]
- W. He, "Load forecasting via deep neural networks", Proc. of 5th International Conference on Information Technology and Quantitative Management, ITQM 2017, 122, pp.308~314 (2017)
- S. P. Kim, "Deep Learning First Step" (2016), Hanbit media
- I. Goodfellow, "Deep Learning" (2016), MIT Press
- K. Cho, "Learning phrase representations using RNN encoder-decoder for statistical machine translation", Proc. of 2014 Conference on Empirical Methods in Natural Language Processing, pp.1724~1734 (2014)
- R. Jozefowicz, "An empirical exploration of recurrent network architectures", Proc. of ICML’15 32nd International Conference on Machine Learning, pp.2342~2350 (2015)
- J. Chung, "Empirical evaluation of gated recurrent neural networks on sequence modeling", Proc. of NIPS 2014 Workshop on Deep Learning, pp.1~9 (2014)
- "UMass Trace Respository"[URL]
- S. Hansun, "A new approach of moving average method in time series analysis", Proc. of 2013 Conference on New Media Studies (ConMedia), pp.1~4 (2013)
- Zhou Wang, "Mean squared error: Love it or leave it? A new look at Signal Fidelity Measures", IEEE Signal Processing Magazine, 26(1), pp.98~117 (2009), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- T. Chai, "Root mean square error (RMSE) or mean absolute error (MAE)? – Arguments against avoiding RMSE in the literature", Geoscientific Model Development, 7(3), pp.1247~1250 (2014), Copernicus GmbH[DOI]
- Shikhar Srivastava, "A comparative study of LSTM neural networks in forecasting day-ahead global horizontal irradiance with satellite data", Solar Energy, 162, pp.232~247 (2018), Elsevier BV[DOI]
- B. Yildiz, "Household electricity load forecasting using historical smart meter data with clustering and classification techniques", Proc. of 2018 IEEE Innovative Smart Grid Technologies - Asia(ISGT Asia), pp.873~879 (2018)