Wei Zhou

Wei Zhou

Auto-generated profile, based on publicly available information

Wei Zhou

School of Mathematics and Computer Science, Yunnan Minzu University, Kunming 650504, People’s Republic of China

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 August 2025 in The Astrophysical Journal Supplement Series
Verified
Journal Article
Open access
WWWenhua Wang
Wenhua Wang
LDLinhua Deng
Linhua Deng
ZJZuo Jiang
Zuo Jiang
JLJinlei Li
Jinlei Li
XZXinhua Zhao
Xinhua Zhao
WZWeihong Zhou
Weihong Zhou
The hemispheric asymmetry of solar activity provides important diagnostics of solar dynamo processes. In this study, we present a phase-resolved statistical analysis of hemispheric sunspot number asymmetry over solar cycles 17–25 (1939–2024), using monthly mean data from the National Astronomical Observatory of Japan. By combining normalized asymmetry indices, cumulative deviation tracking, and year-by-year significance testing, we identify four key results: (1) the northern hemisphere consistently dominates during the ascending phases of most cycles, suggesting a preferential emergence of magnetic flux at the northern hemisphere; (2) the hemispheric asymmetry significantly diminishes during polarity reversal periods, indicating a balancing effect of global magnetic reconfiguration; (3) solar cycles 23 and 25 exhibit remarkably similar asymmetry patterns, implying solar cycle 25 is likely to be similar to solar cycle 23; and (4) the strength of hemispheric dominance correlates with the overall solar activity level, with stronger asymmetries observed during high-activity phases and weaker signals near solar minima. These findings offer new observational constraints for modeling hemispheric solar dynamics and establish a reproducible, scalable framework for future investigations of long-term solar magnetic asymmetries.

19 November 2025 in The Astrophysical Journal
Verified
Journal Article
Open access
WZWei Zhou
Wei Zhou
WZWeihong Zhou
Weihong Zhou
WWWenhua Wang
Wenhua Wang
JCJie Cao
Jie Cao
TXTingting Xu
Tingting Xu
PEPeng E
Peng E
The angular width of coronal mass ejections (CMEs) reflects their eruption scale, dynamics, and potential geoeffectiveness. To deepen our understanding of the hemispheric distribution characteristics of CME activities, we systematically investigate the north–south asymmetry of CMEs with different angular widths. Based on the white-light CME catalog from the Coordinated Data Analysis Workshop during solar cycles 23–25, our main results are as follows: (1) the temporal evolution of CME activity across solar cycles is primarily governed by normal and partial-halo CMEs, while narrow CMEs contribute relatively little to the overall variation; (2) CME activity in cycles 23 and 25 exhibits similar hemispheric patterns, with both cycles showing southern hemisphere dominance, in contrast to the northern dominance observed in cycle 24 (this difference may be attributed to the weaker heliospheric pressure during cycle 24); (3) the phase lag between CME activities in the two hemispheres is generally within five months. These findings reveal the solar activity cycle evolution of CME angular width distributions and hemispheric asymmetries, providing new physical constraints for understanding interhemispheric coupling in solar dynamo processes and improving space weather forecasting models.

26 December 2025 in arXiv
Verified
Preprint
Open access
PDF file
ZGZongyuan Ge
Zongyuan Ge
CZChenwaner Zhang
Chenwaner Zhang
WZWei Zhou
Wei Zhou
HZHongyu Zeng
Hongyu Zeng
GZGuiping Zhou
Guiping Zhou
Geomagnetic storms (GSTs) driven by solar wind-magnetosphere coupling can severely disrupt technological systems, motivating the need for improved prediction accuracy and longer warning times. In this study, we develop a physics-informed Long Short-Term Memory (LSTM) model that incorporates cosmic-ray flux modulation as an additional precursor signal. As coronal mass ejection (CME)-driven disturbances propagate through the heliosphere, enhanced turbulence and magnetic-field compression reduce galactic cosmic-ray (GCR) flux measured by ground-based neutron monitors (Forbush decreases), providing early information that can precede near-Earth solar-wind signatures by 1--3 days. We integrate multi-source space-weather data, spanning 1995-2020, including cosmic-ray observations, solar wind plasma parameters, interplanetary magnetic-field data, and geomagnetic indices. Based on these data, we construct a 19-dimensional feature vector that includes flux background levels, decrease-related indicators, and inter-station correlation measures as model inputs. Employing a 50-unit LSTM architecture, the proposed model achieves root-mean-square errors (RMSE) of 5.106 nT, 8.315 nT, 10.854 nT, 12.883 nT, and 14.788 nT for 2-, 6-, 12-, 24-, and 48-hour predictions, respectively. Incorporating cosmic-ray information further improves 48-hour forecast skill by up to 25.84% (from 0.178 to 0.224). These results demonstrate the value of physics-informed deep learning and cosmic-ray precursors for advancing space-weather forecasting.

20 April 2026 in Research in Astronomy and Astrophysics
Verified
Journal Article
Closed access
ZGZongyuan Ge
Zongyuan Ge
CZChenwaner Zhang
Chenwaner Zhang
WZWei Zhou
Wei Zhou
HZHongyu Zeng
Hongyu Zeng
GZGuiping Zhou
Guiping Zhou
Reliable forecasting of geomagnetic storms (GSTs) remains a major challenge in space weather, particularly at lead times beyond several hours. Most existing models rely on near-Earth solar wind data, which inherently limits the prediction lead time. We extend the forecast horizon by exploiting galactic cosmic ray (GCR) modulation as an upstream precursor. A physics-motivated Long Short-Term Memory (LSTM) model is developed that integrates cosmic-ray flux, solar wind, interplanetary magnetic field, and geomagnetic parameters from a 1995–2020 dataset at 1-hour cadence. The model achieves RMSE values of 5.106–14.788 nT for lead times of 2–48 h, with the inclusion of cosmic-ray inputs improving the 48-hour F1 score by 25.84%. For ICME-driven storms, we further introduce an event-based classification sub-model that predicts whether an approaching ICME will trigger a significant storm 24–48 h in advance; the corresponding F1 scores reach 0.55 (24 h) and 0.48 (48 h). Permutation importance analysis shows that cosmic-ray-related variables contribute substantially to predictive skill, while additional statistical analysis reveals that stronger GCR suppression correlates with more negative minimum Dst, supporting the physical interpretability of GCR modulation as a meaningful precursor.

11 November 2026 in Research in Astronomy and Astrophysics
Verified
Journal Article
Open access
PDF file
WZWei Zhou
Wei Zhou
WZWeihong Zhou
Weihong Zhou
YYYun Yang
Yun Yang
JCJie Cao
Jie Cao
ZGZongyuan Ge
Zongyuan Ge
Geomagnetic storms are intense space weather events triggered by solar activity and can significantly affect spacecraft operations and critical infrastructure security. The disturbance storm time (Dst) index is an important indicator of geomagnetic storm intensity and evolution, and its forecasting accuracy and timeliness directly affect geomagnetic storm warning effectiveness; however, existing Dst forecasting studies still exhibit evident forecast lag during geomagnetic storms. To address this issue, this study constructs an Informer-based framework for 1–6 hr-ahead Dst forecasting using hourly OMNI 2 data from 1995 to 2024 and nine solar wind parameters as inputs, and designs a lag-aware composite loss to balance numerical forecasting accuracy and temporal alignment performance. The results show that: (1) during the storm-time period and in a typical intense geomagnetic storm event, the lag-aware composite loss improves temporal alignment performance; (2) over both the full test period and the storm-time period, the model trained with this loss achieves lower root mean square error values at most forecasting horizons, indicating preserved numerical forecasting capability; and (3) in the directly comparable 4 hr-ahead forecasting task during the storm-time period, the Informer-based model achieves a higher D on value than the reference GRU-based model, suggesting better temporal alignment performance. These results indicate that loss-function design oriented toward the forecast-lag problem can not only improve the temporal alignment performance of Dst forecasting, but also enhance forecasting accuracy to some extent, thereby providing a feasible approach for improving Dst forecasting performance.

20 February 2026 in The Astrophysical Journal
Verified
Journal Article
Open access
WZWei Zhou
Wei Zhou
WZWeihong Zhou
Weihong Zhou
YYYun Yang
Yun Yang
JCJie Cao
Jie Cao
ZGZong-Yuan Ge
Zong-Yuan Ge
YDYe Dong
Ye Dong
Coronal mass ejections (CMEs) are intense solar eruptions that, once reaching Earth, pose catastrophic risks to the near-Earth space environment. Thus, improving CME transit time forecast accuracy to mitigate damage to ground-based facilities is critical for space weather forecasting. In this study, we constructed a 3D convolutional neural network (3D CNN) framework that learns from Solar and Heliospheric Observatory Large Angle and Spectrometric Coronagraph white-light image sequences to predict CME transit times. Our main contributions are threefold. (1) Unlike prior work, we first integrate the temporal dimension and use 3D CNN for joint spatiotemporal modeling of CME sequences to predict transit times. (2) To address the uneven sequence-length variations across events and to optimize data set usage, we propose a method that samples multiple fixed-length subsequences for each event and applies mean pooling to their features, which enhances model performance and generalization. (3) To enhance interpretability, we apply gradient-weighted class activation mapping across all convolutional layers to reveal each layer’s attention focus. The results show that shallow layers focus on the CME body and evolve over time, while deeper layers show stronger responses in the background and weaker responses to the CME body. On an independent test set, the model achieves a mean absolute error (MAE) of 11.04 hr; in the transit time range with the most samples, MAE is 6.37 hr, comparable to prior studies. These results confirm 3D CNN’s effectiveness in modeling CME spatiotemporal evolution and predicting transit times, offering a promising path for physically interpretable, data-driven space weather forecasting.