Page 173 - 《振动工程学报》2026年第5期

第 5 期 倪 菲，等：物理信息机器学习驱动的高速磁浮列车悬浮系统动力学响应预测 1377

in Applied Mechanics and Engineering， 2024， 420： metrics， and comparisons[J]. Journal of Computational
116721. Physics，2023，477：111902.
[25] YANG Y B，PERDIKARIS P. Adversarial uncertainty quan- [32] SUN H K，XU C Y，WU G X，et al. A two-stage physical-
tification in physics-informed neural networks[J]. Journal of informed neural network approach for high-speed railway
Computational Physics，2019，394：136-152. track geometry irregularity maintenance[J]. IEEE Transac-
[26] LIU X， YAO W， PENG W， et al. Bayesian physics- tions on Intelligent Transportation Systems，2025，26（8）：
informed extreme learning machine for forward and inverse 12080-12092.
PDE problems with noisy data[J]. Neurocomputing， 2023，
[33] LIU Y F，YANG L，ZHANG S F，et al. Parameter identifi-
549：126425.
cation and platoon control for virtually coupled train set with
[27] DABROWSKI J J，PAGENDAM D E，HILTON J，et al.
physics-informed neural network dynamics model[J]. Vehicle
Bayesian physics informed neural networks for data assimila-
System Dynamics，2025，63（1）：71-92.
tion and spatio-temporal modelling of wildfires[J]. Spatial
[34] SUN H K， GAO Y， XU C Y， et al. High-speed railway
Statistics，2023，55：100746.
track geometry robust maintenance with Bayesian deep learn-
[28] PEREZ S，MADDU S，SBALZARINI I F，et al. Adaptive
ing-enhanced physical-informed neural networks[J].
weighting of Bayesian physics informed neural networks for
Computer-Aided Civil and Infrastructure Engineering，
multitask and multiscale forward and inverse problems[J].
2025，40（24）：3934-3952.
Journal of Computational Physics，2023，491：112342.
[35] 靖永志. 高速磁浮列车悬浮间隙传感器及其建模方法研究
[29] ZOU Z R，MENG X H，KARNIADAKIS G E. Correcting
[D]. 成都：西南交通大学，2014.
model misspecification in physics-informed neural networks
（PINNs）[J]. Journal of Computational Physics，2024，505： JING Yongzhi. Study on the gap sensor of high-speed maglev
train and its modeling method[D]. Chengdu：Southwest Jiao-
112918.
[30] LIU H L，WANG Z，DENG R，et al. Flow reconstruction tong University，2014.
with uncertainty quantification from noisy measurements
based on Bayesian physics-informed neural networks[J]. 第一作者：倪 菲（1985—），女，博士，副教授。
Physics of Fluids，2024，36（11）：117104. E-mail：fei.ni@tongji.edu.cn
[31] PSAROS A F，MENG X H，ZOU Z R，et al. Uncertainty 通信作者：范 琳（1999—），女，博士研究生。
quantification in scientific machine learning： methods， E-mail：fanlin@tongji.edu.cn