• Login
  • Register
  • Search

Short-term Power Load Forecasting Based on LSTM Neural Network Model

Yanyan Wei

Abstract


This paper selects humidity, maximum temperature, minimum temperature, historical load data and holiday type as input variables, and constructs LSTM neural network model to predict the daily maximum load and daily minimum load of a place. The results show that LSTM is better than the minimum load in forecasting the daily maximum load.

Keywords


LSTM ; Power load forecasting; Meteorological factors

Full Text:

PDF

Included Database


References


[1] Peter G Z. Time series forecasting using a hybrid ARIMA and neuralnetwork model[J]. Neurocomputing, 2003(50): 159-175.

[2] Charlton N, Singleton C. Arefi ned parametric model for short term load forecasting. International Journal of Forecasting[J].2014(30): 364-368.

[3] Du Songhuai,Hou Zhijian,Jiang Chuanwen. Grey forecasting price, utation and its simulation[J]. Journal of Grey System,2003(1): 43-48.

[4] Granger C. Combining forecasts-twenty years later[J]. Journal of Forecasting, 1989, 8(3): 167-173.

[5] Ho Ku Long, Hsu Yuan Yih, Lee C E,etal. Short-term load forecasting of Tai-wan power system using aknowledge-based expert system[J]. IEEE PWRS, 1990, 5(4): 46-57.

[6] Hong C, Canizares C Z, Singh A. ANN-based short-term load forecasting in electricity markets[C]. 2001 IEEE Power Engineering Society Winter Meeting. Conference Proceedings, 2001(2): 411-415.

[7]RafiS.H,Nahid-Al-Masood,DeebaSR,etal.Ashort-termloadforecastingmethodusingintegratednnandlstmnetwork[J].IEEE Access,2021(9):32436-32448.

[8] Goswami K, Ganguly A, et al. Day ahead forecasting and peak load management using multivariate auto regression technique[C].2018 IEEE Applied Signal Processing Conference (ASPCON), 2018.




DOI: http://dx.doi.org/10.18686/ahe.v7i7.7607

Refbacks