当前位置:
X-MOL 学术
›
Tunn. Undergr. Space Technol.
›
论文详情
Our official English website, www.x-mol.net, welcomes your
feedback! (Note: you will need to create a separate account there.)
Decoupling analysis of inertia effects in tunnel boring machine using a data-physics driven approach
Tunnelling and Underground Space Technology ( IF 6.7 ) Pub Date : 2025-01-10 , DOI: 10.1016/j.tust.2025.106367
Yongsheng Li, Limao Zhang
Tunnelling and Underground Space Technology ( IF 6.7 ) Pub Date : 2025-01-10 , DOI: 10.1016/j.tust.2025.106367
Yongsheng Li, Limao Zhang
To quantify the “inertia effects” in construction equipment and reduce the time delay and violent disturbance phenomenon, an analysis method based on Support Energy in Generalized Inertia (SEGI) is proposed. In this paper, the theoretical expression of generalized inertia in Electric, Hydraulic, and Mechanical (EHM) systems is analyzed, where an Encoder-Decoder Neural Network (EDNN) model integrating Squeeze and Excitation (SE) and Shared Weight (SW) strategies is used to measure the SEGI. The Optimal Foraging Algorithm (OFA) realizes the system decoupling of inertia relaxation factors of EHM systems. The effectiveness of the proposed approach is verified in a tunnel boring machine (TBM). The results indicate that: (1) The proposed data-driven inertia measurement model can accurately predict the SEGI, where its R2 is 0.987; (2) Through decoupling optimization algorithm, the generalized inertia relaxation factors of EHM systems are obtained, which are 1.297, 0.97, 1.289, respectively; (3) Compared with the theoretical value and real value of SEGI, 81.94% of the sample errors are less than 5%. An important contribution of this paper is to propose a generalized inertia decoupling approach for complex system, which provides theoretical support for the precise control of construction equipment.
中文翻译:
使用数据物理驱动的方法对隧道掘进机的惯性效应进行解耦分析
为了量化建筑设备中的“惯性效应”,减少时间延迟和剧烈扰动现象,该文提出一种基于广义惯性支撑能(SEGI)的分析方法。本文分析了电气、液压和机械 (EHM) 系统中广义惯性的理论表达式,其中使用集成挤压和激励 (SE) 和共享权重 (SW) 策略的编码器-解码器神经网络 (EDNN) 模型来测量 SEGI。最优觅食算法 (OFA) 实现了 EHM 系统惯性弛豫因子的系统解耦。所提方法的有效性在隧道掘进机 (TBM) 中得到了验证。结果表明:(1)所提出的数据驱动惯量测量模型能够准确预测SEGI,其R2为0.987;(2)通过解耦优化算法,得到EHM系统的广义惯量弛豫因子,分别为1.297、0.97、1.289;(3) 与 SEGI 的理论值和实际值相比,81.94% 的样本误差小于 5%。本文的重要贡献是提出了一种复杂系统的广义惯性解耦方法,为施工设备的精确控制提供了理论支持。
更新日期:2025-01-10
中文翻译:

使用数据物理驱动的方法对隧道掘进机的惯性效应进行解耦分析
为了量化建筑设备中的“惯性效应”,减少时间延迟和剧烈扰动现象,该文提出一种基于广义惯性支撑能(SEGI)的分析方法。本文分析了电气、液压和机械 (EHM) 系统中广义惯性的理论表达式,其中使用集成挤压和激励 (SE) 和共享权重 (SW) 策略的编码器-解码器神经网络 (EDNN) 模型来测量 SEGI。最优觅食算法 (OFA) 实现了 EHM 系统惯性弛豫因子的系统解耦。所提方法的有效性在隧道掘进机 (TBM) 中得到了验证。结果表明:(1)所提出的数据驱动惯量测量模型能够准确预测SEGI,其R2为0.987;(2)通过解耦优化算法,得到EHM系统的广义惯量弛豫因子,分别为1.297、0.97、1.289;(3) 与 SEGI 的理论值和实际值相比,81.94% 的样本误差小于 5%。本文的重要贡献是提出了一种复杂系统的广义惯性解耦方法,为施工设备的精确控制提供了理论支持。