Journal Article10.1016/j.energy.2023.128565
A spatio-temporal forecasting model using optimally weighted graph convolutional network and gated recurrent unit for wind speed of different sites distributed in an offshore wind farm
Xuefang Xu,Shiting Hu,Huaishuang Shao,Peiming Shi,Ruixiong Li,Deguang Li +5 more
50
TL;DR: A novel spatio-temporal forecasting model combines optimally weighted graph convolutional network and gated recurrent unit to predict wind speed at different offshore wind farm sites, outperforming existing models by utilizing both geographic and temporal features.
read more
Abstract: Accurate wind speed forecasting plays an essential role in scheduling wind power generation. Currently, most existing models predict wind speed just based on temporal features and geographical information of wind turbine sites is usually neglected or not taken full use of, leading to poor prediction performance. To solve this issue, a novel spatio-temporal prediction model based on optimally weighted graph convolutional network (GCN) and gated recurrent unit (GRU) is proposed. First, data preprocessing based on tensor decomposition is applied to recover missing speed values. Second, geographic and dynamic time warping (DTW) distance are both calculated to construct an optimally weighted graph for different turbine sites, which not only takes fully into account the geographic information but also the temporal similarity of speed series. Third, based on the weighted graph, GCN is used to extract spatial features from speed series adequately. Afterwards, by inputting spatial features of each site into GRU, respectively, the spatio-temporal features can be effectively extracted. Finally, model parameters are solved by iteration with a Huber loss. To demonstrate the proposed model, performance indexes including RMSE, MAE, MAPE, R2 and TIC are calculated. Results show that the proposed model has the smallest prediction error of all sites for multi-step short-term forecasting compared with prevalent models. Therefore, the proposed model effectively enhances prediction accuracy by making full use of spatio-temporal features.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Single and simultaneous fault diagnosis of gearbox via wavelet transform and improved deep residual network under imbalanced data
Suiyan Wang,Jiaye Tian,Pengfei Liang,Xuefang Xu,Zhuoze Yu,Siyuan Liu,Delong Zhang +6 more
TL;DR: A novel method integrates wavelet transform and an improved deep residual network (IDRN) to diagnose gearbox faults under imbalanced data, achieving higher accuracy and generalization ability compared to existing intelligent diagnostic models.
36
A wind speed forcasting model based on rime optimization based VMD and multi-headed self-attention-LSTM
TL;DR: A hybrid wind speed forecasting model combining Rime optimization, VMD, multi-headed self-attention, and LSTM outperforms 19 comparative models on four wind farms in China and two US datasets, achieving higher prediction accuracy and smaller errors.
33
A hybrid short-term wind power point-interval prediction model based on combination of improved preprocessing methods and entropy weighted GRU quantile regression network
TL;DR: A hybrid wind power prediction model, QR-EGRU, combines improved preprocessing methods and entropy weighted GRU quantile regression network to accurately predict point and interval wind power values, outperforming existing models in tracking abrupt changes and improving prediction accuracy.
21
A novel spatio-temporal wind speed forecasting method based on the microscale meteorological model and a hybrid deep learning model
Dongqin Zhang,Gang Hu,Jie Song,Huanxiang Gao,Hehe Ren,Wenli Chen +5 more
TL;DR: A novel spatio-temporal wind speed forecasting method combines a microscale meteorological model with a hybrid deep learning model, utilizing ICEEMDAN signal decomposition and wind speed ratios to predict multi-step ahead wind speeds with high accuracy in both temporal and spatial domains.
13
GGNet: A novel graph structure for power forecasting in renewable power plants considering temporal lead-lag correlations
Nanyang Zhu,Ying Wang,Kun Yuan,Jiahao Yan,Yaping Li,Kaifeng Zhang +5 more
TL;DR: This paper proposes GGNet, a novel graph neural network that captures temporal lead-lag correlations among renewable power plants using a dynamic graph structure with multiple temporal granularity groups, achieving improved power forecasting performance with reduced RMSEN and MAEN.
12
References
ARMA based approaches for forecasting the tuple of wind speed and direction
Ergin Erdem,Jing Shi +1 more
TL;DR: In this paper, four approaches based on autoregressive moving average (ARMA) method are employed for short-term forecasting of wind speed and direction are employed to forecast wind turbine operation and efficient energy harvesting.
874
Scalable tensor factorizations for incomplete data
TL;DR: An algorithm called CP-WOPT (CP Weighted OPTimization) that uses a first-order optimization approach to solve the weighted least squares problem and is shown to successfully factorize tensors with noise and up to 99% missing data.
728
A new hybrid model for wind speed forecasting combining long short-term memory neural network, decomposition methods and grey wolf optimizer
TL;DR: Experimental results indicate that the proposed combined model can capture non-linear characteristics of WSTS, achieving better forecasting performance than single forecasting models, in terms of accuracy.
519
Repeated wavelet transform based ARIMA model for very short-term wind speed forecasting
TL;DR: This paper proposes a new Repeated WT based ARIMA (RWT-ARIMA) model, which has improved accuracy for very short-term wind speed forecasting and a comparison proves the superiority of the proposed RWT-ARimA model over other models in very Short-term Wind speed forecasting.
500
A hybrid model for spatiotemporal forecasting of PM2.5 based on graph convolutional neural network and long short-term memory.
TL;DR: A hybrid model based on deep learning methods that integrates Graph Convolutional networks and Long Short-Term Memory networks to model and forecast the spatiotemporal variation of PM2.5 concentrations is proposed.
490