Liqun Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Liqun Liu has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Haochen Li
School of Electronic Information Engineering, Taiyuan University of Science and Technology 1 , Taiyuan 030024, China
2 shared publications Qiusheng He
2 shared publicationsHaochen Li
School of Electronic Information Engineering, Taiyuan University of Science and Technology, Taiyuan, China
1 shared publicationJianfeng Zhang
1 shared publication- Kai Yang
Department of Automation, Taiyuan Institute of Technology, Taiyuan, China
1 shared publication Qingfeng Wu
1 shared publication
Latest Publications
- 20 June 2026
- Published by Springer Nature in Machine Vision and Applications
- Vol. 37 (5) , 101
YOLO-GHS: high-precision detection of insulator defects on transmission lines under severe weather conditions
- 20 June 2026
- Published by Springer Nature in Machine Vision and Applications
- Vol. 37 (5) , 101
YOLO-GHS: high-precision detection of insulator defects on transmission lines under severe weather conditions
- Open Access
- 10 February 2025
- Published by Elsevier in International Journal of Electrical Power & Energy Systems
- Vol. 165
Graph attention convolution network for power flow calculation considering grid uncertainty
- 10 February 2025
- Published by Elsevier in International Journal of Electrical Power & Energy Systems
- Vol. 165
Open AccessGraph attention convolution network for power flow calculation considering grid uncertainty
- Open Access
- 1 January 2024
- Published by AIP Publishing in Journal of Renewable and Sustainable Energy
- Vol. 16 (1)
A joint missing power data recovery method based on the spatiotemporal correlation of multiple wind farms
In reality, wind power data are often accompanied by data losses, which can affect the accurate prediction of wind power and subsequently impact the real-time scheduling of the power system. Existing methods for recovering missing data primarily consider the environmental conditions of individual wind farms, thereby overlooking the spatiotemporal correlations between neighboring wind farms, which significantly compromise their recovery effectiveness. In this paper, a joint missing data recovery model based on power data from adjacent wind farms is proposed. At first, a spatial–temporal module (STM) is designed using a combination of graph convolution network and recurrent neural networks to learn spatiotemporal dependencies and similarities. Subsequently, to provide a solid computational foundation for the STM, a Euclidean-directed graph based on Granger causality is constructed to reflect the hidden spatiotemporal information in the data. Finally, comprehensive tests on data recovery for both missing completely at random and short-term continuous missing are conducted on a real-world dataset. The results demonstrate that the proposed model exhibits a significant advantage in missing data recovery compared to baseline models.
- 1 January 2024
- Published by AIP Publishing in Journal of Renewable and Sustainable Energy
- Vol. 16 (1)
Open AccessA joint missing power data recovery method based on the spatiotemporal correlation of multiple wind farms
In reality, wind power data are often accompanied by data losses, which can affect the accurate prediction of wind power and subsequently impact the real-time scheduling of the power system. Existing methods for recovering missing data primarily consider the environmental conditions of individual wind farms, thereby overlooking the spatiotemporal correlations between neighboring wind farms, which significantly compromise their recovery effectiveness. In this paper, a joint missing data recovery model based on power data from adjacent wind farms is proposed. At first, a spatial–temporal module (STM) is designed using a combination of graph convolution network and recurrent neural networks to learn spatiotemporal dependencies and similarities. Subsequently, to provide a solid computational foundation for the STM, a Euclidean-directed graph based on Granger causality is constructed to reflect the hidden spatiotemporal information in the data. Finally, comprehensive tests on data recovery for both missing completely at random and short-term continuous missing are conducted on a real-world dataset. The results demonstrate that the proposed model exhibits a significant advantage in missing data recovery compared to baseline models.
- Open Access
- 21 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 131418-131434
A Spatiotemporal Coupling Calculation-Based Short-Term Wind Farm Cluster Power Prediction Method
Accurate short-term wind power prediction is of great significance to the real-time dispatching of power systems and the development of wind power generation plans. However, existing methods for wind power prediction have the following problems: (1) some studies aim at predicting wind power for a single wind farm, ignoring the correlations of the adjacent wind farms; (2) many studies tend to convert wind speed forecast results to wind power, increasing the conversion error; (3) almost all studies place emphasis on the capture of spatiotemporal features, neglecting the influence of spatiotemporal coupling. Therefore, to solve the above questions, this work proposes an adaptive graph neural network based on spatiotemporal attention calculation for short-term wind farm cluster power prediction, using only wind power data. Firstly, a dynamic undirected graph is established to sufficiently learn prior knowledge of spatial relationships. Next, the spatiotemporal coupling relationship and global temporal correlation between data can be computed by performing spatiotemporal cross-attention and temporal self-attention, respectively. Finally, a novel hybrid loss function is proposed to optimize the prediction model accurately. In a case study, compared with other benchmark methods, the proposed method shows excellent overall performance in predicting wind power.
- 21 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 131418-131434
Open AccessA Spatiotemporal Coupling Calculation-Based Short-Term Wind Farm Cluster Power Prediction Method
Accurate short-term wind power prediction is of great significance to the real-time dispatching of power systems and the development of wind power generation plans. However, existing methods for wind power prediction have the following problems: (1) some studies aim at predicting wind power for a single wind farm, ignoring the correlations of the adjacent wind farms; (2) many studies tend to convert wind speed forecast results to wind power, increasing the conversion error; (3) almost all studies place emphasis on the capture of spatiotemporal features, neglecting the influence of spatiotemporal coupling. Therefore, to solve the above questions, this work proposes an adaptive graph neural network based on spatiotemporal attention calculation for short-term wind farm cluster power prediction, using only wind power data. Firstly, a dynamic undirected graph is established to sufficiently learn prior knowledge of spatial relationships. Next, the spatiotemporal coupling relationship and global temporal correlation between data can be computed by performing spatiotemporal cross-attention and temporal self-attention, respectively. Finally, a novel hybrid loss function is proposed to optimize the prediction model accurately. In a case study, compared with other benchmark methods, the proposed method shows excellent overall performance in predicting wind power.