同济大学“人工智能+大气海洋”实验室 AI + Climate

研究方向

本研究团队由同济大学软件学院和国家气象中心全球气象室组成，是目前国内 AI+ 气象交叉领域领先研究团队之一。研究团队的研究方向包括采用 AI 技术改善数值模式的预报技巧并提高预报效率和人工智能气象应用的物理一致性和可解释性研究等，覆盖 AI+ 气象的各个方面。具体成果包括天行气象大模型、数值模式-AI 模型融合的 NAO 集合预报系统及基于 CNOP 的台风集合预报系统、台风监测与预报、天气-次季节-季节尺度的 NAO 智能预测、ENSO 智能集合预测系统、北极海冰智能预测、数值模式预报结果的偏差订正与降尺度深度学习模型 、挖掘因果关系的通用深度学习组件 CAU 等。

高温热浪
高温热浪

通过AI参数化方案和智能数值模式对高温热浪进行预报。

降水
降水

应用AI参数化方案和智能数值模式进行次季节到季节降水预报。

台风
台风

结合深度学习和智能数值模式进行台风监测与预报。

北大西洋涛动 (NAO)
北大西洋涛动 (NAO)

采用深度学习与CESM耦合模式，结合了集合预报和敏感区识别。

北极海冰
北极海冰

通过AI智能预报和偏差订正方法监测北极海冰变化。

厄尔尼诺-南方涛动 (ENSO)
厄尔尼诺-南方涛动 (ENSO)

应用深度学习和物理引导进行智能预测。

主要研究成果

Identifying Optimal Precursors of NAO Using CNOP Method in CESM
Identifying Optimal Precursors of NAO Using CNOP Method in CESM Exploring Initial Perturbations and Nonlinear Processes in NAO Prediction

Authors: Bin Mu, Jing Li, Shijin Yuan, Xiaodan Luo, Guokun Dai
Publisher: European Geosciences Union
Date: 2020-07-01
ID: doi:10.5194/npg-2020-27

ENSO CNOP
Deep Neural Network for ENSO Prediction and Sensitive Area Identification
Deep Neural Network for ENSO Prediction and Sensitive Area Identification Understanding and Forecasting EP-type and CP-type ENSO Events

Authors: Bin Mu, Yuehan Cui, Shijin Yuan, Bo Qin
Publisher: European Geosciences Union
Date: 2022-05-25
ID: doi:10.5194/gmd-15-4105-2022

ENSO Prediction

全部

2024

A deep learning-based global tropical cyclogenesis prediction model and its interpretability analysis
A deep learning-based global tropical cyclogenesis prediction model and its interpretability analysis
Bin Mu, Xin Wang, Shijin Yuan, Yuxuan Chen, Guansong Wang, Bo Qin, Guanbo Zhou
Springer  ·  02 Aug 2024  ·  doi:10.1007/s11430-023-1383-6
Deep-Learning Model-Interpretability Prediction
Multivariate Upstream Kuroshio Transport (UKT) Prediction and Targeted Observation Sensitive Area Identification of UKT Seasonal Reduction
Multivariate Upstream Kuroshio Transport (UKT) Prediction and Targeted Observation Sensitive Area Identification of UKT Seasonal Reduction
Bin Mu, Yifan Yang-Hu, Bo Qin, Shijin Yuan
Ocean Modelling  ·  01 Jun 2024  ·  doi:10.1016/j.ocemod.2024.102344
UKT Kuroshio-Transport
Developing intelligent Earth System Models: An AI framework for replacing sub-modules based on incremental learning and its application
Developing intelligent Earth System Models: An AI framework for replacing sub-modules based on incremental learning and its application
Bin Mu, Zi-Jun Zhao, Shi-Jin Yuan, Bo Qin, Guo-Kun Dai, Guan-Bo Zhou
Atmospheric Research  ·  01 Jun 2024  ·  doi:10.1016/j.atmosres.2024.107306
AI Incremental-Learning
A deep learning-based bias correction model for Arctic sea ice concentration towards MITgcm
A deep learning-based bias correction model for Arctic sea ice concentration towards MITgcm
Shijin Yuan, Shichen Zhu, Xiaodan Luo, Bin Mu
Ocean Modelling  ·  01 Apr 2024  ·  doi:10.1016/j.ocemod.2024.102326
Arctic-Sea-Ice Deep-Learning
A generative adversarial network–based unified model integrating bias correction and downscaling for global SST
A generative adversarial network–based unified model integrating bias correction and downscaling for global SST
Shijin Yuan, Xin Feng, Bin Mu, Bo Qin, Xin Wang, Yuxuan Chen
Atmospheric and Oceanic Science Letters  ·  31 Jan 2024  ·  doi:10.1016/j.aosl.2023.100407
SST GAN Bias-Correction Downscaling

2023

IceTFT v1.0.0: interpretable long-term prediction of Arctic sea ice extent with deep learning
IceTFT v1.0.0: interpretable long-term prediction of Arctic sea ice extent with deep learning
Bin Mu, Xiaodan Luo, Shijin Yuan, Xi Liang
European Geosciences Union  ·  18 Aug 2023  ·  doi:10.5194/gmd-16-4677-2023
Arctic-Sea-Ice Deep-Learning IceTFT
Dimension shifting based intelligent algorithm framework to solve conditional nonlinear optimal perturbation
Dimension shifting based intelligent algorithm framework to solve conditional nonlinear optimal perturbation
Shijin Yuan, Yaxuan Liu, Huazhen Zhang, Bin Mu
Elsevier  ·  01 Jul 2023  ·  doi:10.1016/j.cageo.2023.105375
Dimension-Shifting CNOP Intelligent-Algorithm
Error Evolutions and Analyses on Joint Effects of SST and SL via Intermediate Coupled Models and Conditional Nonlinear Optimal Perturbation Method
Error Evolutions and Analyses on Joint Effects of SST and SL via Intermediate Coupled Models and Conditional Nonlinear Optimal Perturbation Method
Bin Mu, Xiaoyun Qin, Shijin Yuan, Bo Qin
MDPI  ·  24 Apr 2023  ·  doi:10.3390/jmse11050910
SST CNOP
PIRT: A Physics-Informed Red Tide Deep Learning Forecast Model Considering Causal-Inferred Predictors Selection
PIRT: A Physics-Informed Red Tide Deep Learning Forecast Model Considering Causal-Inferred Predictors Selection
Bin Mu, Bo Qin, Shijin Yuan, Xin Wang, Yuxuan Chen
IEEE  ·  01 Mar 2023  ·  doi:10.1109/LGRS.2023.3250642
Red-Tide Deep-Learning Causal-Inference

2022

ENSO-GTC: ENSO Deep Learning Forecast Model With a Global Spatial-Temporal Teleconnection Coupler
ENSO-GTC: ENSO Deep Learning Forecast Model With a Global Spatial-Temporal Teleconnection Coupler
Bin Mu, Bo Qin, Shijin Yuan
American Geophysical Union  ·  15 Dec 2022  ·  doi:10.1029/2022MS003132
ENSO Deep-Learning Teleconnection
Ensemble Forecast for Tropical Cyclone Based on CNOP-P Method: A Case Study of WRF Model and Two Typhoons
Ensemble Forecast for Tropical Cyclone Based on CNOP-P Method: A Case Study of WRF Model and Two Typhoons
Yuan Shijin, Shi Bo, Zhao Zijun, Mu Bin, Zhou Feifan, Duan Wansuo
Journal of Tropical Meteorology  ·  10 Jun 2022  ·  doi:10.46267/j.1006-8775.2022.010
ENSO typhoon CNOP
Simulation, precursor analysis and targeted observation sensitive area identification for two types of ENSO using ENSO-MC v1.0
Simulation, precursor analysis and targeted observation sensitive area identification for two types of ENSO using ENSO-MC v1.0
Bin Mu, Yuehan Cui, Shijin Yuan, Bo Qin
European Geosciences Union  ·  25 May 2022  ·  doi:10.5194/gmd-15-4105-2022
ENSO Prediction
The NAO Variability Prediction and Forecasting with Multiple Time Scales Driven by ENSO Using Machine Learning Approaches
The NAO Variability Prediction and Forecasting with Multiple Time Scales Driven by ENSO Using Machine Learning Approaches
Bin Mu, Jing Li, Shijin Yuan, Xiaodan Luo
Wiley  ·  15 Apr 2022  ·  doi:10.1155/2022/6141966
NAO ENSO Machine-Learning
Feature extraction-based intelligent algorithm framework with neural network for solving conditional nonlinear optimal perturbation
Feature extraction-based intelligent algorithm framework with neural network for solving conditional nonlinear optimal perturbation
Shijin Yuan, Huazhen Zhang, Yaxuan Liu, Bin Mu
Springer  ·  29 Jan 2022  ·  doi:10.1007/s00500-021-06639-8
ENSO intelligent-algorithm

2021

ENSO-ASC 1.0.0: ENSO deep learning forecast model with a multivariate air–sea coupler
ENSO-ASC 1.0.0: ENSO deep learning forecast model with a multivariate air–sea coupler
Bin Mu, Bo Qin, Shijin Yuan
European Geosciences Union  ·  17 Nov 2021  ·  doi:10.5194/gmd-14-6977-2021
ENSO Deep-Learning Air-Sea-Coupler
Efficient Executions of Community Earth System Model onto Accelerators Using GPUs
Efficient Executions of Community Earth System Model onto Accelerators Using GPUs
Shijin Yuan, Cheng Wang, Bin Mu, Xiaodan Luo
ACM  ·  09 Jun 2021  ·  doi:10.1145/3449301.3449334
CESM GPU CUDA
Typhoon Intensity Forecasting Based on LSTM Using the Rolling Forecast Method
Typhoon Intensity Forecasting Based on LSTM Using the Rolling Forecast Method
Shijin Yuan, Cheng Wang, Bin Mu, Feifan Zhou, Wansuo Duan
MDPI  ·  04 Mar 2021  ·  doi:10.3390/a14030083
LSTM Typhoon-Forecasting Rolling-Forecast

2020

A Climate Downscaling Deep Learning Model considering the Multiscale Spatial Correlations and Chaos of Meteorological Events
A Climate Downscaling Deep Learning Model considering the Multiscale Spatial Correlations and Chaos of Meteorological Events
Bin Mu, Bo Qin, Shijin Yuan, Xiaoyun Qin
Wiley  ·  16 Nov 2020  ·  doi:10.1155/2020/7897824
Climate-Downscaling Deep-Learning Meteorological-Events
Prediction of North Atlantic Oscillation Index Associated with the Sea Level Pressure Using DWT-LSTM and DWT-ConvLSTM Networks
Prediction of North Atlantic Oscillation Index Associated with the Sea Level Pressure Using DWT-LSTM and DWT-ConvLSTM Networks
Bin Mu, Jing Li, Shijin Yuan, Xiaodan Luo
Wiley  ·  05 Oct 2020  ·  doi:10.1155/2020/2413041
DWT-LSTM ConvLSTM NAO
Applying Convolutional LSTM Network to Predict El Niño Events: Transfer Learning from The Data of Dynamical Model and Observation
Applying Convolutional LSTM Network to Predict El Niño Events: Transfer Learning from The Data of Dynamical Model and Observation
Bin Mu, Shaoyang Ma, Shijin Yuan, Hui Xu
IEEE  ·  17 Jul 2020  ·  doi:10.1109/ICEIEC49280.2020.9152317
ConvLSTM El-Niño Transfer-Learning
Optimal Precursors Identification for North Atlantic Oscillation using CESM and CNOP Method
Optimal Precursors Identification for North Atlantic Oscillation using CESM and CNOP Method
Bin Mu, Jing Li, Shijin Yuan, Xiaodan Luo, Guokun Dai
European Geosciences Union  ·  01 Jul 2020  ·  doi:10.5194/npg-2020-27
ENSO CNOP

2019

Parallel PCA-Based Bacterial Foraging Optimization Algorithm for Identifying Optimal Precursors of North Atlantic Oscillation
Parallel PCA-Based Bacterial Foraging Optimization Algorithm for Identifying Optimal Precursors of North Atlantic Oscillation
Bin Mu, Jing Li, Shijin Yuan, Xiaodan Luo, Guokun Dai
IEEE  ·  10 Aug 2019  ·  doi:10.1109/HPCC/SmartCity/DSS.2019.00165
PCA BFOA NAO
CNOP-P-based parameter sensitivity for double-gyre variation in ROMS with simulated annealing algorithm
CNOP-P-based parameter sensitivity for double-gyre variation in ROMS with simulated annealing algorithm
Shijin Yuan, Huazhen Zhang, Mi Li, Bin Mu
Springer  ·  28 May 2019  ·  doi:10.1007/s00343-019-7266-2
CNOP-P Parameter-Sensitivity Simulated-Annealing
Optimal precursors of double-gyre regime transitions with an adjoint-free method
Optimal precursors of double-gyre regime transitions with an adjoint-free method
Shijin Yuan, Mi Li, Qiang Wang, Kun Zhang, Huazhen Zhang, Bin Mu
Springer  ·  25 Mar 2019  ·  doi:10.1007/s00343-019-7235-9
Double-Gyre CNOP Adjoint-Free-Method

2018

Identifying Typhoon Targeted Observations Sensitive Areas Using the Gradient Definition Based Method
Identifying Typhoon Targeted Observations Sensitive Areas Using the Gradient Definition Based Method
Bin Mu, Juhui Ren, Shijin Yuan, Feifan Zhou
Springer  ·  26 Oct 2018  ·  doi:10.1007/s13143-018-0068-1
CNOP Gradient-Method Typhoon-Observations
A novel approach for solving CNOPs and its application in identifying sensitive regions of tropical cyclone adaptive observations
A novel approach for solving CNOPs and its application in identifying sensitive regions of tropical cyclone adaptive observations
Linlin Zhang, Bin Mu, Shijin Yuan, Feifan Zhou
European Geosciences Union  ·  13 Sep 2018  ·  doi:10.5194/npg-25-693-2018
CNOP PSO Tropical-Cyclones

2017

An Efficient Approach Based on the Gradient Definition for Solving Conditional Nonlinear Optimal Perturbation
An Efficient Approach Based on the Gradient Definition for Solving Conditional Nonlinear Optimal Perturbation
Bin Mu, Juhui Ren, Shijin Yuan
Wiley  ·  14 Nov 2017  ·  doi:10.1155/2017/3208431
Gradient CNOP Optimization
An improved effective approach for urban air quality forecast
An improved effective approach for urban air quality forecast
Bin Mu, Site Li, Shijin Yuan
IEEE  ·  29 Jul 2017  ·  doi:10.1109/FSKD.2017.8393403
Air-Quality Forecasting PCA
Parallel dynamic search fireworks algorithm with linearly decreased dimension number strategy for solving conditional nonlinear optimal perturbation
Parallel dynamic search fireworks algorithm with linearly decreased dimension number strategy for solving conditional nonlinear optimal perturbation
Bin Mu, Junhui Zhao, Shijin Yuan, Jinghao Yan
IEEE  ·  14 May 2017  ·  doi:10.1109/IJCNN.2017.7966136
dynFWA CNOP Parallelization

2016

PCAFP for Solving CNOP in Double-Gyre Variation and Its Parallelization on Clusters
PCAFP for Solving CNOP in Double-Gyre Variation and Its Parallelization on Clusters
Shijin Yuan, Mi Li, Bin Mu, Jingpeng Wang
IEEE  ·  14 Dec 2016  ·  doi:10.1109/HPCC-SmartCity-DSS.2016.0049
PCAFP CNOP Parallelization
Parallel Modified Artificial Bee Colony Algorithm for Solving Conditional Nonlinear Optimal Perturbation
Parallel Modified Artificial Bee Colony Algorithm for Solving Conditional Nonlinear Optimal Perturbation
Juhui Ren, Shijin Yuan, Bin Mu
IEEE  ·  14 Dec 2016  ·  doi:10.1109/HPCC-SmartCity-DSS.2016.0055
MABC CNOP Parallel-Computing

2015

Robust ensemble feature extraction for solving conditional nonlinear optimal perturbation
Robust ensemble feature extraction for solving conditional nonlinear optimal perturbation
Shicheng Wen, Shijin Yuan, Hongyu Li, Bin Mu
Inderscience  ·  01 Dec 2015  ·  doi:10.1504/IJCSE.2015.073494
Ensemble Feature-Extraction CNOP
Parallel Cooperative Co-evolution Based Particle Swarm Optimization Algorithm for Solving Conditional Nonlinear Optimal Perturbation
Parallel Cooperative Co-evolution Based Particle Swarm Optimization Algorithm for Solving Conditional Nonlinear Optimal Perturbation
Shijin Yuan, Li Zhao, Bin Mu
Springer  ·  09 Nov 2015  ·  doi:10.1007/978-3-319-26535-3_11
PSO Co-evolution CNOP
PCAGA: Principal component analysis based genetic algorithm for solving conditional nonlinear optimal perturbation
PCAGA: Principal component analysis based genetic algorithm for solving conditional nonlinear optimal perturbation
Bin Mu, Linlin Zhang, Shijin Yuan, Hongyu Li
IEEE  ·  01 Oct 2015  ·  doi:10.1109/IJCNN.2015.7280522
CNOP Genetic-Algorithm Principal-Component-Analysis
PCGD: Principal components-based great deluge method for solving CNOP
PCGD: Principal components-based great deluge method for solving CNOP
Shicheng Wen, Shijin Yuan, Bin Mu, Hongyu Li, Juhui Ren
IEEE  ·  14 Sep 2015  ·  doi:10.1109/CEC.2015.7257067
CNOP Principal-Components Great-Deluge-Method
User-QoS-Based Web Service Clustering for QoS Prediction
User-QoS-Based Web Service Clustering for QoS Prediction
Fuxin Chen, Shijin Yuan, Bin Mu
IEEE  ·  17 Aug 2015  ·  doi:10.1109/ICWS.2015.83
QoS Web-Service Clustering

2014

Code parallel refactoring of the Zebiak-Cane model based on JASMIN
Code parallel refactoring of the Zebiak-Cane model based on JASMIN
Hongyu Li, Jiachuan Qin, Shijin Yuan, Jie Bian
IEEE  ·  01 Dec 2014  ·  doi:10.1109/ICCSNT.2013.6967098
JASMIN Zebiak-Cane-model Parallel-Computing
QoS-Aware Cloud Service Selection Based on Uncertain User Preference
QoS-Aware Cloud Service Selection Based on Uncertain User Preference
Bin Mu, Su Li, Shijin Yuan
Springer  ·  01 Oct 2014  ·  doi:10.1007/978-3-319-11740-9_54
Cloud-Service QoS User-Preference
SAEP: Simulated Annealing Based Ensemble Projecting Method for Solving Conditional Nonlinear Optimal Perturbation
SAEP: Simulated Annealing Based Ensemble Projecting Method for Solving Conditional Nonlinear Optimal Perturbation
Shicheng Wen, Shijin Yuan, Bin Mu, Hongyu Li, Lei Chen
Springer  ·  02 Aug 2014  ·  doi:10.1007/978-3-319-11194-0_58
CNOP Simulated-Annealing Nonlinear-Perturbation
Orthogonal Neighborhood Preservation Projection Based Method for Solving CNOP
Orthogonal Neighborhood Preservation Projection Based Method for Solving CNOP
Bin Mu, Shicheng Wen, Shijin Yuan, Hongyu Li
Springer  ·  01 Aug 2014  ·  doi:10.1007/978-3-319-09339-0_12
CNOP Numerical-Models Adjoint-based-Method
Parallel Optimization of the MM5 Adjoint Model
Parallel Optimization of the MM5 Adjoint Model
Hongyu Li, Jie Bian, Shijin Yuan, Jiachuan Qin
IEEE  ·  12 Jun 2014  ·  doi:10.1109/HPCC.and.EUC.2013.134
MM5 Adjoint-Model Parallel-Computing
ACStack: Adaptive Composite Stack for Adjoint Models in Code Optimization
ACStack: Adaptive Composite Stack for Adjoint Models in Code Optimization
Hongyu Li, Weiyu Zhou, Shijin Yuan, Shicheng Wen
IEEE  ·  12 Jun 2014  ·  doi:10.1109/HPCC.and.EUC.2013.193
Adjoint-Model Code-Optimization Numerical-Weather-Forecasting

2013

An optimization framework for adjoint-based climate simulations: A case study of the Zebiak-Cane model
An optimization framework for adjoint-based climate simulations: A case study of the Zebiak-Cane model
Shijin Yuan, Shicheng Wen, Qin Liu
Sage Journals  ·  11 Jul 2013  ·  doi:10.1177/1094342013495096
Adjoint-based-Simulations Climate-Simulations Zebiak-Cane-model