Hanqing Zhao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Hanqing Zhao has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Lianlei Lin
School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin, China
11 shared publications - Junkai Wang
School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin 150001, China
9 shared publications Sheng Gao
School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin, China
7 shared publications- Zongwei Zhang
School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin, China
7 shared publications Junkai Wang
School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin, China
4 shared publicationsHangyi Yu
School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin, China
3 shared publications
Latest Publications
- Open Access
- 28 July 2026
- Published by Springer Nature in Communications Earth & Environment
- Vol. 7 (1)
Physics-guided neural network integrating uncertainty evolution for spatiotemporal wind speed prediction
- 28 July 2026
- Published by Springer Nature in Communications Earth & Environment
- Vol. 7 (1)
Open AccessPhysics-guided neural network integrating uncertainty evolution for spatiotemporal wind speed prediction
- Open Access
Bridging Equation-Based and Data-Driven Dynamics for Reliable Wind Speed Prediction in Energy Systems
Wind speed prediction is an essential spatiotemporal forecasting task in wind energy systems, yet it remains challenging due to the nonlinear and dynamic characteristics of atmospheric processes. The evolution of wind is governed by physical laws, which can be effectively described using partial differential equations (PDEs). To improve forecasting reliability and accuracy, this paper proposes a novel network model, termed DynWindNet, which integrates equation-based dynamics with data-driven dynamics within a unified framework. Specifically, an interactive dual-branch architecture is designed, where a Physics–Data Coupling Module (PDCM) enables adaptive information exchange between the two dynamics via attention-based gating mechanisms. In addition, a frequency-aware enhancement module (FAEM) is introduced to refine the representations of the data-driven branch by selectively emphasizing informative frequency components. Experimental results on the ERA5 dataset demonstrate that DynWindNet consistently outperforms representative baseline methods across atmospheric pressure levels. Overall, the proposed framework provides an effective approach for integrating physics-guided evolution modeling with deep spatiotemporal representation learning in wind field forecasting.
Open AccessBridging Equation-Based and Data-Driven Dynamics for Reliable Wind Speed Prediction in Energy Systems
Wind speed prediction is an essential spatiotemporal forecasting task in wind energy systems, yet it remains challenging due to the nonlinear and dynamic characteristics of atmospheric processes. The evolution of wind is governed by physical laws, which can be effectively described using partial differential equations (PDEs). To improve forecasting reliability and accuracy, this paper proposes a novel network model, termed DynWindNet, which integrates equation-based dynamics with data-driven dynamics within a unified framework. Specifically, an interactive dual-branch architecture is designed, where a Physics–Data Coupling Module (PDCM) enables adaptive information exchange between the two dynamics via attention-based gating mechanisms. In addition, a frequency-aware enhancement module (FAEM) is introduced to refine the representations of the data-driven branch by selectively emphasizing informative frequency components. Experimental results on the ERA5 dataset demonstrate that DynWindNet consistently outperforms representative baseline methods across atmospheric pressure levels. Overall, the proposed framework provides an effective approach for integrating physics-guided evolution modeling with deep spatiotemporal representation learning in wind field forecasting.
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 19, 25624-25645
Hydrology-Aware Contrastive Meta-Learning for Domain Generalization Hyperspectral Wetland Image Classification
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 19, 25624-25645
Open AccessHydrology-Aware Contrastive Meta-Learning for Domain Generalization Hyperspectral Wetland Image Classification
- 19 December 2025
- Published by Elsevier in Neurocomputing
- Vol. 669
Asymmetric disentanglement for domain generalization in hyperspectral image classification
- 19 December 2025
- Published by Elsevier in Neurocomputing
- Vol. 669
Asymmetric disentanglement for domain generalization in hyperspectral image classification
- 17 September 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 298
Unbiased representation learning via feature decoupling network for cross-scene hyperspectral image classification
- 17 September 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 298
Unbiased representation learning via feature decoupling network for cross-scene hyperspectral image classification
- Open Access
- 22 May 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (11) , 1816
Spatial-Spectral Linear Extrapolation for Cross-Scene Hyperspectral Image Classification
In realistic hyperspectral image (HSI) cross-scene classification tasks, it is ideal to obtain target domain samples during the training phase. Therefore, a model needs to be trained on one or more source domains (SD) and achieve robust domain generalization (DG) performance on an unknown target domain (TD). Popular DG strategies constrain the model’s predictive behavior in synthetic space through deep, nonlinear source expansion, and an HSI generation model is usually adopted to enrich the diversity of training samples. However, recent studies have shown that the activation functions of neurons in a network exhibit asymmetry for different categories, which results in the learning of task-irrelevant features while attempting to learn task-related features (called “feature contamination”). For example, even if some intrinsic features of HSIs (lighting conditions, atmospheric environment, etc.) are irrelevant to the label, the neural network still tends to learn them, resulting in features that make the classification related to these spurious components. To alleviate this problem, this study replaces the common nonlinear generative network with a specific linear projection transformation, to reduce the number of neurons activated nonlinearly during training and alleviate the learning of contaminated features. Specifically, this study proposes a dimensionally decoupled spatial spectral linear extrapolation (SSLE) strategy to achieve sample augmentation. Inspired by the weakening effect of water vapor absorption and Rayleigh scattering on band reflectivity, we simulate a common spectral drift based on Markov random fields to achieve linear spectral augmentation. Further considering the common co-occurrence phenomenon of patch images in space, we design spatial weights combined with label determinism of the center pixel to construct linear spatial enhancement. Finally, to ensure the cognitive unity of the high-level features of the discriminator in the sample space, we use inter-class contrastive learning to align the back-end feature representation. Extensive experiments were conducted on four datasets, an ablation study showed the effectiveness of the proposed modules, and a comparative analysis with advanced DG algorithms showed the superiority of our model in the face of various spectral and category shifts. In particular, on the Houston18/Shanghai datasets, its overall accuracy was 0.51%/0.83% higher than the best results of the other methods, and its Kappa coefficient was 0.78%/2.07% higher, respectively.
- 22 May 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (11) , 1816
Open AccessSpatial-Spectral Linear Extrapolation for Cross-Scene Hyperspectral Image Classification
In realistic hyperspectral image (HSI) cross-scene classification tasks, it is ideal to obtain target domain samples during the training phase. Therefore, a model needs to be trained on one or more source domains (SD) and achieve robust domain generalization (DG) performance on an unknown target domain (TD). Popular DG strategies constrain the model’s predictive behavior in synthetic space through deep, nonlinear source expansion, and an HSI generation model is usually adopted to enrich the diversity of training samples. However, recent studies have shown that the activation functions of neurons in a network exhibit asymmetry for different categories, which results in the learning of task-irrelevant features while attempting to learn task-related features (called “feature contamination”). For example, even if some intrinsic features of HSIs (lighting conditions, atmospheric environment, etc.) are irrelevant to the label, the neural network still tends to learn them, resulting in features that make the classification related to these spurious components. To alleviate this problem, this study replaces the common nonlinear generative network with a specific linear projection transformation, to reduce the number of neurons activated nonlinearly during training and alleviate the learning of contaminated features. Specifically, this study proposes a dimensionally decoupled spatial spectral linear extrapolation (SSLE) strategy to achieve sample augmentation. Inspired by the weakening effect of water vapor absorption and Rayleigh scattering on band reflectivity, we simulate a common spectral drift based on Markov random fields to achieve linear spectral augmentation. Further considering the common co-occurrence phenomenon of patch images in space, we design spatial weights combined with label determinism of the center pixel to construct linear spatial enhancement. Finally, to ensure the cognitive unity of the high-level features of the discriminator in the sample space, we use inter-class contrastive learning to align the back-end feature representation. Extensive experiments were conducted on four datasets, an ablation study showed the effectiveness of the proposed modules, and a comparative analysis with advanced DG algorithms showed the superiority of our model in the face of various spectral and category shifts. In particular, on the Houston18/Shanghai datasets, its overall accuracy was 0.51%/0.83% higher than the best results of the other methods, and its Kappa coefficient was 0.78%/2.07% higher, respectively.
- Open Access
- 17 April 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (8) , 1434
Neural Network Based on Dynamic Collaboration of Flows for Temporal Downscaling
Time downscaling is one of the most challenging topics in remote sensing and meteorological data processing. Traditional methods often face the problems of high computing cost and poor generalization ability. The framework interpolation method based on deep learning provides a new idea for the time downscaling of meteorological data. A deep neural network for the time downscaling of multivariate meteorological data is designed in this paper. It estimates the kernel weight and offset vector of each target pixel independently among different meteorological variables and generates output frames guided by the feature space. Compared with other methods, this model can deal with a large range of complex meteorological movements. The 2 h interval downscaling experiments for 2 m temperature, surface pressure, and 1000 hPa specific humidity show that the MAE of the proposed method is reduced by about 14%, 25%, and 18%, respectively, compared with advanced methods such as AdaCof and Zooming Slow-Mo. Performance fluctuates very little over time, with an average performance fluctuation of only about 1% across all metrics. Even in the downscaling experiment with a 6 h interval, the proposed model still maintains a leading performance advantage, which indicates that the proposed model has not only good performance and robustness but also excellent scalability and transferability in the downscaling task in the multivariate meteorological field.
- 17 April 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (8) , 1434
Open AccessNeural Network Based on Dynamic Collaboration of Flows for Temporal Downscaling
Time downscaling is one of the most challenging topics in remote sensing and meteorological data processing. Traditional methods often face the problems of high computing cost and poor generalization ability. The framework interpolation method based on deep learning provides a new idea for the time downscaling of meteorological data. A deep neural network for the time downscaling of multivariate meteorological data is designed in this paper. It estimates the kernel weight and offset vector of each target pixel independently among different meteorological variables and generates output frames guided by the feature space. Compared with other methods, this model can deal with a large range of complex meteorological movements. The 2 h interval downscaling experiments for 2 m temperature, surface pressure, and 1000 hPa specific humidity show that the MAE of the proposed method is reduced by about 14%, 25%, and 18%, respectively, compared with advanced methods such as AdaCof and Zooming Slow-Mo. Performance fluctuates very little over time, with an average performance fluctuation of only about 1% across all metrics. Even in the downscaling experiment with a 6 h interval, the proposed model still maintains a leading performance advantage, which indicates that the proposed model has not only good performance and robustness but also excellent scalability and transferability in the downscaling task in the multivariate meteorological field.
- Top Cited PapersOpen Access
- 3 April 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-18
A machine learning model for hub-height short-term wind speed prediction
Accurate short-term wind speed prediction is crucial for maintaining the safe, stable, and efficient operation of wind power systems. We propose a multivariate meteorological data fusion wind prediction network (MFWPN) to study fine-grid vector wind speed prediction, taking Northeast China as an example. Results show that MFWPN outperforms the ECMWF-HRES model regarding vector wind speed prediction accuracy within the first 6 h. Transfer experiments demonstrate the good generalized performance of the MFWPN, which can be quickly applied to offsite prediction. Efficiency experiments show that the MFWPN takes only 18 ms to predict vector wind speeds on a 24-hour fine grid over the future northeastern region. With its demonstrated accuracy and efficiency, the MFWPN can be an effective tool for predicting vector wind speeds in large regional wind centers and can help in ultrashort- and short-term deployment planning for wind power.
- 3 April 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-18
Top Cited PapersOpen AccessA machine learning model for hub-height short-term wind speed prediction
Accurate short-term wind speed prediction is crucial for maintaining the safe, stable, and efficient operation of wind power systems. We propose a multivariate meteorological data fusion wind prediction network (MFWPN) to study fine-grid vector wind speed prediction, taking Northeast China as an example. Results show that MFWPN outperforms the ECMWF-HRES model regarding vector wind speed prediction accuracy within the first 6 h. Transfer experiments demonstrate the good generalized performance of the MFWPN, which can be quickly applied to offsite prediction. Efficiency experiments show that the MFWPN takes only 18 ms to predict vector wind speeds on a 24-hour fine grid over the future northeastern region. With its demonstrated accuracy and efficiency, the MFWPN can be an effective tool for predicting vector wind speeds in large regional wind centers and can help in ultrashort- and short-term deployment planning for wind power.
- 5 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-17
Spatiotemporal MultiWaveNet for Efficiently Generating Environmental Spatiotemporal Series
Real-time and accurate modeling of environmental variables in a specific region is of great significance to human production activities. In the era of intelligence, there is an increasing demand for accuracy and real-time performance of modeling environmental variables. However, due to the lack of a priori knowledge of spatiotemporal features and the high computational complexity, the existing deep learning methods have poor accuracy, and the real-time performance needs to be improved. To solve the above problems, this paper proposes a novel SpatioTemporal MultiWaveNet based on 3D convolution for end-to-end environmental spatiotemporal variables generation. The Spatiotemporal MultiWave Layer is designed based on the principle of wavelet transform to extract multi-frequency spatiotemporal features from the series. The MultiScale Embedding Module, Time InfoFusion Module, and Embedding Integration Module are proposed to further extract and integrate the spatiotemporal features to improve the accuracy of the model. Considering the short-term and long-term temporal characteristics of spatiotemporal series, the date encoding strategy and the cross-attention head are proposed to integrate relative and absolute date information. The improved loss function is proposed to optimize the model learning. Comparative experiments on three datasets with different physical fields, different spatial scales, and resolutions show that the proposed method can accurately generate the mesoscale environmental spatiotemporal variables in real time for different tasks with the generation error lower than other methods by 1.02∼83.52% on RMSE, and 1.67∼83.68% on MAE. Further experiments show that the proposed method can maintain its performance in practical applications.
- 5 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-17
Spatiotemporal MultiWaveNet for Efficiently Generating Environmental Spatiotemporal Series
Real-time and accurate modeling of environmental variables in a specific region is of great significance to human production activities. In the era of intelligence, there is an increasing demand for accuracy and real-time performance of modeling environmental variables. However, due to the lack of a priori knowledge of spatiotemporal features and the high computational complexity, the existing deep learning methods have poor accuracy, and the real-time performance needs to be improved. To solve the above problems, this paper proposes a novel SpatioTemporal MultiWaveNet based on 3D convolution for end-to-end environmental spatiotemporal variables generation. The Spatiotemporal MultiWave Layer is designed based on the principle of wavelet transform to extract multi-frequency spatiotemporal features from the series. The MultiScale Embedding Module, Time InfoFusion Module, and Embedding Integration Module are proposed to further extract and integrate the spatiotemporal features to improve the accuracy of the model. Considering the short-term and long-term temporal characteristics of spatiotemporal series, the date encoding strategy and the cross-attention head are proposed to integrate relative and absolute date information. The improved loss function is proposed to optimize the model learning. Comparative experiments on three datasets with different physical fields, different spatial scales, and resolutions show that the proposed method can accurately generate the mesoscale environmental spatiotemporal variables in real time for different tasks with the generation error lower than other methods by 1.02∼83.52% on RMSE, and 1.67∼83.68% on MAE. Further experiments show that the proposed method can maintain its performance in practical applications.
- 7 June 2024
- Published by Elsevier in Renewable and Sustainable Energy Reviews
- Vol. 201
Wind speed prediction in China with fully-convolutional deep neural network
- 7 June 2024
- Published by Elsevier in Renewable and Sustainable Energy Reviews
- Vol. 201
Wind speed prediction in China with fully-convolutional deep neural network