- Climate variability and models
- Meteorological Phenomena and Simulations
- Integrated Energy Systems Optimization
- Model Reduction and Neural Networks
- Oceanographic and Atmospheric Processes
- Marine and coastal ecosystems
- Tropical and Extratropical Cyclones Research
- Turbomachinery Performance and Optimization
- Heat Transfer and Optimization
- Complex Network Analysis Techniques
- Fluid Dynamics and Turbulent Flows
- Advanced Image Processing Techniques
- Power Systems and Renewable Energy
- Complex Systems and Time Series Analysis
- Refrigeration and Air Conditioning Technologies
- Flow Measurement and Analysis
- Aerodynamics and Fluid Dynamics Research
- Calibration and Measurement Techniques
- Radiative Heat Transfer Studies
- Advanced Neural Network Applications
- Nuclear reactor physics and engineering
- Nuclear Engineering Thermal-Hydraulics
- Arctic and Antarctic ice dynamics
- Power Systems and Technologies
- Adversarial Robustness in Machine Learning
National Center for Climate Change Strategy and International Cooperation
2010-2024
China Meteorological Administration
2012-2024
Changshu Institute of Technology
2021-2023
China University of Mining and Technology
2023
Chinese People's Liberation Army
2023
First Affiliated Hospital of Kunming Medical University
2022
Kunming Medical University
2022
Xiangtan University
2022
Beihang University
2004-2006
Institute of Information Engineering
2004
Abstract This study reveals that the relationship between autumn north tropical Atlantic (NTA) sea surface temperature (SST) anomalies and Northeast China’s winter snowfall (NECWS) undergoes a remarkable interdecadal enhancement after 2001. Previous research confirmed NTA SST anomaly atmospheric circulation experienced changes 2000s suggested various reasons for this phenomenon. During 1961–2000, has significantly positive correlation with other oceans, especially Indian Ocean (TIO), latter...
Based on reanalysis data of National Center Environment Prediction America, temperature correlation networks different time scales are constructed and their dynamical statistical properties also analyzed. Results show that 1 30d belong to small world semi-globally coupled network, respectively. The scale condition is 11—12d for shift from one type the other. For d scale, total number links at each geographic location semi-symmetrical between north south hemisphere, equatorial nodes have a...
A climate network of six indices the North Pacific air—sea system is constructed during period 1948–2009. In order to find out inherent relationship between intrinsic mechanism index and important shift, synchronization behaviour coupling these are investigated. Results indicate that happened around beginning 1960s, in middle 1970s at beginnings 1990s 2000s separately. These states were always followed by decrease coefficient. Each was well associated with abrupt phase or trend changes...
Abstract Based on the complex network method, this study reveals temporal and spatial characteristics of synchronization extreme rainfall events in southwest China (SWC) during main season (May–September). Our results show significant pattern SWC, which is closely associated with convergence warm‐wet air cold‐dry corresponding occurrence SWC. The divergence indicates a dominated direction net influence from Sichuan Basin to Yunnan–Guizhou Plateau, implying moving north south SWC more...
Abstract It is well known that El Niño events can induce worldwide impacts. However, the fact strong do not necessarily impacts raises a new research question: how to estimate of in advance? To address this question, we studied from perspective complex network. By comparing results five with distinct impacts, found phase transition surface air temperature network over tropical Pacific closely related This phenomenon was used explain less‐than‐expected 2015/2016 Niño, which suggested more...
A climate network of extreme rainfall over eastern Asia is constructed for the period 1971–2000, employing tools complex networks and a measure nonlinear correlation called event synchronization (ES). Using this network, we predict several cases without delay with n-day (1 ≤ n 10). The prediction accuracy can reach 58% delay, 21% 1-day 12% (2 results reveal that low in years weak east summer monsoon (EASM) or 1 year later high strong EASM later. Furthermore, higher due to many more links...
The spatial and temporal stability of temperature correlation network is analyzed based on extreme temperature, El Niňo/La Niňa abnormal year, key areas change high frequent year respectively. Results show that, the connectivity decreases by filtering out property at inflexion P=0.1 or 0.9, might be defined as boundary temperature. La possesses more significan links higher clustering coefficient than Niňo network, which indicates that former communicative stable latter. And these statistical...
Droughts and floods have frequently occurred in Southwest China (SWC) during the past several decades. Yet, understanding of mechanism precipitation SWC is still a challenge, since East Asian monsoon Indian potentially influence rainfall this region. Thus, prediction has become difficult critical topic climatology. We develop novel multi-variable network-based method to delineate relations between global sea surface temperature anomalies (SSTA) (PA) SWC. Our results show that out-degree...
Abstract In 2021/2022 winter, an intraseasonal surface air temperature (SAT) reversal occurred in China, shifting from regional warm to cold phase late January 2022. The variations of winter SAT China are tightly connected with 500 hPa potential height. particular, the Arctic Oscillation (AO) positive is diagnosed be contributing reversal. Under its influence, build‐up and collapse blockings over Urals Mountain led surges February Seasonal prediction systems (SPSs) such as NCEP_CFSv2,...
Full-state estimation with a limited number of sen-sors is valuable and challenging task in monitoring con-trolling complex physical systems. Supervised learning methods based on deep neural networks have shown excellent performance by the nonlinear mapping from sparse observations to global field. However, The network black box weak explanation for processes, reconstruction architecture optimization network. This paper aims leverage structure laws inherent data reconstruct field solving...
An objective identification technique is used to detect regional extreme low temperature events (RELTE) in China during 1960–2009. Their spatial-temporal characteristics are analyzed. The results indicate that the lowest temperatures of RELTE, together with frequency distribution geometric latitude center, exhibit a double-peak feature. RELTE frequently happen near area 30°N and 42°N before mid-1980s, but shifted afterwards 30°N. During 1960–2009, frequency, intensity, maximum impacted show...
As one of the most basic and important functions in intelligent transportation system, related research management control has attracted extensive attention researchers from all over world. Good traffic guidance are inseparable accurate real-time flow prediction, at same time, reliable prediction is also key to transition system "passive control" "active The extraction analysis features support complete data sets. This paper considers correlation between missing available spatiotemporal...
Abstract As subsystems of the Asian summer monsoon, precipitation variations in India and northern part eastern China (NEC) are physically connected. This study noted that connection has been significantly enhanced after 1999 compared to 1979–98, which is due strengthened water vapor transportation between two regions. That associated with interdecadal combined effects El Niño–Southern Oscillation (ENSO) sea surface temperature anomalies (SSTAs) over tropical Indian Ocean (TIO) on northwest...
Abstract Recovering the global accurate complex physics field from limited sensors is critical to measurement and control of engineering system. General reconstruction methods for recovering field, especially deep learning with more parameters better representational ability, usually require large amounts labeled data which unaffordable in practice. To solve problem, this paper proposes uncertainty guided ensemble self-training (UGE-ST), using plentiful unlabeled improve performance reduce...