使用LSTM-PINN混合方法进行地震响应预测的性能改进
Seunggoo Kim1, Donwoo Lee1, Seungjae Lee1
1School of Industrial Design & Architectural Engineering, Korea University of Technology & Education, 1600 Chungjeol-ro, Byeongcheon-myeon, Cheonan 31253, Republic of Korea.
Biomimetics (Basel, Switzerland)
|August 27, 2025
概括
一个新的混合人工智能模型结合了长期短期记忆 (LSTM) 网络和物理信息神经网络 (PINNs) 来改善地震结构反应预测. 这种LSTM-PINN模型提供比传统PINN更稳定,更准确的结果,即使培训较少.
科学领域:
- 结构工程
- 人工智能
- 地震分析
背景情况:
- 准确预测结构对地震负荷的反应对于安全至关重要.
- 像物理信息神经网络 (PINNs) 和长期短期记忆 (LSTM) 网络这样的深度学习模型显示出有希望但有局限性.
- 在非线性系统中,PINN缺乏长期时间依赖性捕获,而LSTM则缺乏物理解释性.
研究的目的:
- 开发一种混合LSTM-PINN模型,将LSTM和PINN的优势整合在一起.
- 增强在地震负荷下的动态结构行为的预测.
- 在结构响应建模中提高物理一致性和时间依赖性.
主要方法:
- 开发了一种混合模型,将LSTM用于时间学习和PINN用于基于物理的约束.
- 该模型在受到El-Centro地面运动的单自由度 (SDOF) 和多自由度 (MDOF) 系统上进行了评估.
- 使用平均误差和平均平方误差 (MSE) 来评估性能,与仅使用PINN的模型相比.
主要成果:
- 混合LSTM-PINN模型在整个时间域实现了更稳定,更精确的预测.
- 与基线PINN相比,它表现出优异的表现,在更少的培训时代 (10,000对50,000) 实现了高达50%的MSE.
- 该模型通过时间序列学习提高了概括能力.
结论:
- 混合LSTM-PINN模型有效地结合了时间学习和物理一致性来预测地震结构反应.
- 这种以物理为指导的时间序列人工智能方法在准确性和效率方面具有显著优势.
- 这些发现支持实时响应估计,结构健康监测和地震性能评估的潜力.
相关概念视频
01:35Long-term Potentiation
55.8K
Long-term potentiation, or LTP, is one of the ways by which synaptic plasticity—changes in the strength of chemical synapses—can occur in the brain. LTP is the process of synaptic strengthening that occurs over time between pre- and postsynaptic neuronal connections. The synaptic strengthening of LTP works in opposition to the synaptic weakening of long-term depression (LTD) and together are the main mechanisms that underlie learning and memory.
55.8K
01:19Survival Tree
159
Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
Building a Survival Tree
Constructing a...
Building a Survival Tree
Constructing a...
159