A deep reinforcement learning approach for wind speed forecasting
  • Metaadatok
Tartalom: http://hdl.handle.net/20.500.12944/105262
Archívum: Közszolgálati Tudásportál
Gyűjtemény: Publikációk
Államtudományi és Nemzetközi Tanulmányok Kar
Cím:
A deep reinforcement learning approach for wind speed forecasting
Létrehozó:
Band, Shahab S.
Lin, Ting Jia
Qasem, Sultan Noman
Ameri, Rasoul
Shahmirzadi, Danyal
Aslam, Muhammad Shamrooz
Pai, Hao-Ting
Salwana, Ely
Mousavi, Amir
Téma:
Renewable energy
deepreinforcement learning
windspeed forecasting
empiricalwavelet transform
longshort-term memory
arti?cialintelligence
machinelearning
Természettudományok
Természettudományok/Matematika- és számítástudományok
Tartalmi leírás:
The conventional wind forecasting methods often struggle to handle the non-stationary and incon-sistent wind patterns. This paper presents a hybrid method of Empirical Wavelet Transform (EWT)and Deep Reinforcement Learning (DRL) for wind speed modeling to overcome the forecasting chal-lenges. The EWT method transforms the original wind speed series into several independent modesand a residual series. In addition, the DRL method is utilised to optimise the weights associated withthree distinct supervised deep learning models, i.e., Long Short-Term Memory (LSTM), ConvolutionalNeural Networks with LSTM (CNN-LSTM), and CNN with Gated Recurrent Units (CNN-GRU). The per-formance of the proposed EWT-DRL is evaluated against deep learning models, including LSTM,CNN-LSTM, CNN-GRU, and their coupling with EWT. The combination of EWT and the DRL (EWT-DRL)method achieves a Mean Absolute Error (MAE) of 0.151, a Mean Squared Error (MSE) of 0.060, a RootMean Squared Error (RMSE) of 0.192, and a correlation coefficient (R) of 0.9913. These results indicatethe effectiveness of EWT-DRL in improving accuracy for wind speed modeling.
Típus:
info:eu-repo/semantics/other
Formátum:
application/pdf
Azonosító:
nke:1997-003X
nke:1994-2060
nke:10.1080/19942060.2025.2498355
nke:36149466
nke:10058509
nke:Engineering Applications of Computational Fluid Mechanics
nke:19
nke:1
nke:001483507000001
nke:105004694196
nke:ENG APPL COMP FLUID
Létrehozó:
info:eu-repo/semantics/openAccess