相关实验视频
Updated: Sep 11, 2025
09:17
Surrogate Model Development for Digital Experiments in Welding
Published on: March 28, 2025
1.2K
基于Elman和feedforward神经网络的模型用于预测流形成AA6082管的机械性质
Tarak Nath De1, Bikramjit Podder2, Nirmal Baran Hui3
1Advanced Systems Laboratory (ASL), Hyderabad, 500058, India.
Scientific reports
|August 12, 2025
概括
本研究引入了H30管的预测模型,通过不进行破坏性测试的流形成参数估计机械性能. 一个前神经网络 (FNN) 实现了最高的准确性,最大限度地减少了实验需求.
科学领域:
- 材料科学 材料科学 材料科学
- 机械工程 机械工程
- 制造过程 制造过程 制造过程
背景情况:
- 破坏性测试通常需要测量流式产品的机械性能.
- 这在物理试验不可行或不可取时带来了局限性.
- 精确预测机械性能对于设计和制造优化至关重要.
研究的目的:
- 开发一个参数预测模型,用于通过流成型制造的H30管.
- 为了能够在不进行破坏性测试的情况下估计机械性能 (收益强度,UTS,延长).
- 为了方便选择最佳的流量形成参数 (输入速度比,轴向分离,输入).
主要方法:
- 关键输入参数的系统变化:输入速度 (FS) 比率,轴向分级 (AS) 和输入量 (IF).
- 测量输出机械性能:收益强度,最终抗拉强度 (UTS) 和百分比延长.
- 开发和评估三个预测模型:多变量回归 (MR),前神经网络 (FNN) 和埃尔曼神经网络 (ENN).
主要成果:
- 推进神经网络 (FNN) 模型表现出最高的预测准确性.
- FNN实现了7.45%的最大平均预测误差.
- FNN的表现优于埃尔曼神经网络 (ENN) (7.64%的误差) 和多变量回归 (MR) (12.4%的误差).
结论:
- 拟议的参数预测模型,特别是FNN,有效地估计了流形H30管的机械性能.
- 这种方法减少了对广泛的物理试验和破坏性测试的必要性.
- 该模型有助于设计人员选择最佳的流形成参数,以实现所需的材料特性.
相关概念视频
01:30Typical Model Studies
441
Fluid mechanics model studies often utilize scaled-down systems to predict fluid behavior in full-scale environments, such as river flows, dam spillways, and structures interacting with open surfaces. Maintaining Froude number similarity in river models is crucial, as it replicates surface flow features like wave patterns and velocities.
441
01:15Thin-Walled Hollow Shafts
238
In analyzing a thin-walled hollow shaft subjected to torsional loading, a segment with width dx is isolated for examination. Despite its equilibrium state, this segment faces torsional shearing forces at its ends. These forces are quantitatively described by the product of the longitudinal shearing stress on the segment's minor surface and the area of this surface, leading to the concept of shear flow. This shear flow is consistent throughout the structure, indicating a uniform distribution...
238
01:23Stresses under Combined Loadings
227
When analyzing a bent tube with a circular cross-section subjected to multiple forces, it is crucial to determine the stress distribution in order to maintain structural integrity under varied load conditions.
The process begins by slicing the tube at critical points and analyzing the internal forces and stress components at these sections, focusing on the centroid. Normal stresses, generated by axial forces and bending moments, are either compressive or tensile and vary across the section from...
The process begins by slicing the tube at critical points and analyzing the internal forces and stress components at these sections, focusing on the centroid. Normal stresses, generated by axial forces and bending moments, are either compressive or tensile and vary across the section from...
227