可解释的人工智能研究设计变量对双复合层材的静态特征的贡献
Saeid Saberi1, Hamid Nasiri2, Omid Ghorbani3
1Department of Mechanical Engineering, Isfahan University of Technology, Isfahan 84156-83111, Iran.
Materials (Basel, Switzerland)
|August 12, 2023
概括
使用SHapley添加式扩展 (SHAP) 的可解释AI (XAI) 揭示了影响双可变复合层状材料形状的关键因素. 热膨胀和湿度变化显著影响曲率和突破力.
科学领域:
- 复合材料科学 复合材料科学
- 计算力学 计算力学 计算力学
- 人工智能的人工智能
背景情况:
- 双面复合板材表现出由材料特性,几何形状和环境影响的复杂行为.
- 了解特征对层形状和穿力贡献的理解对于设计和应用至关重要.
- 现有的方法可能无法完全阐明输入参数和层状材料反应之间的复杂关系.
研究的目的:
- 为了研究各种输入特征对稳定平衡形状和可两性复合板材的突破力的影响.
- 应用一种新的可解释的人工智能 (XAI) 方法,SHapley添加式解释 (SHAP),用于特征重要性分析.
- 量化材料特性,几何尺寸和环境条件对层状材料特性的贡献.
主要方法:
- 应用最小能量的原则和雷利-里茨法来生成层状材料响应数据.
- 利用夏普利添加式解释 (SHAP) 进行特征重要性和贡献的详细分析.
- 采用极端梯度提升 (XGBoost) 进行特征重要性识别和有限元 (FM) 方法进行验证.
主要成果:
- SHAP分析有效地识别和排名了影响层层曲率和穿力输入特征的重要性.
- 发现横向热膨胀系数和湿度变化对模型的输出产生了最显著的影响.
- 特性重要性分析证实了特定材料特性和环境因素的主导作用.
结论:
- 可解释的人工智能,特别是SHAP,为复式复合板材的复杂行为提供了宝贵的见解.
- 横向热膨胀和湿度变化是设计人员必须考虑的关键参数,以控制层状板的形状和快穿行为.
- 结合理论方法,XAI和数值模拟的综合方法为理解和预测层层性能提供了强大的框架.
相关概念视频
Prismatic Beams: Problem Solving
145
In the design of a supported timber beam subjected to a distributed load, both the beam's physical dimensions and the timber's characteristics, such as its grade and species, are critical. These factors determine the allowable stress values, which are crucial for calculating the necessary beam depth to ensure structural integrity and safety.
The design begins with analyzing the beam as a free body to identify moments and force balances, thereby determining support reactions. Next, the...
The design begins with analyzing the beam as a free body to identify moments and force balances, thereby determining support reactions. Next, the...
145
Bending of Members Made of Several Materials
223
In analyzing a structural member composed of two different materials with identical cross-sectional areas, it is crucial to understand how their distinct elastic properties affect the member's response under load. The analysis involves assessing stress and strain distributions using the transformed section concept, which accounts for variations in material properties.
Hooke's Law determines stress in each material, stating that stress is proportional to strain but varies due to each...
Hooke's Law determines stress in each material, stating that stress is proportional to strain but varies due to each...
223
Design of Prismatic Beams for Bending
275
The design of prismatic beams, structural elements with a uniform cross-section, focuses on ensuring safety and structural integrity under load. The design process begins by determining the allowable stress, either from material properties tables, or by dividing the material's ultimate strength by a safety factor. This safety factor is essential for accommodating uncertainties, and varies depending on the material—timber, steel, or concrete—with each having unique strength and...
275
Factorial Design
13.1K
Factorial Analysis is an experimental design that applies Analysis of Variance (ANOVA) statistical procedures to examine a change in a dependent variable due to more than one independent variable, also known as factors. Changes in worker productivity can be reasoned, for example, to be influenced by salary and other conditions, such as skill level. One way to test this hypothesis is by categorizing salary into three levels (low, moderate, and high) and skills sets into two levels (entry level...
13.1K
Wood Products
108
Wood products encompass a broad range of materials crafted from wood strands, veneers, lumber, and even waste wood-like shreds, designed for both structural and nonstructural purposes. Various specialized wood products have been developed to enhance strength, durability, and versatility in building applications.
Glue-laminated wood, often referred to as glulam, combines multiple smaller pieces of dimensional lumber using adhesives to form a single, larger piece. Cross-laminated timber consists...
Glue-laminated wood, often referred to as glulam, combines multiple smaller pieces of dimensional lumber using adhesives to form a single, larger piece. Cross-laminated timber consists...
108
Experimental Designs
11.6K
An experimental design is a systematic process that allows researchers to evaluate the relationship between dependent and independent variables. There are three widely used types of experimental design - pre-experimental design, true experimental design, and quasi-experimental design. In pre-experimental design, the researcher compares the data before and after some interventions or treatments. The true-experimental design has more than one purposefully created group, a commonly measured...
11.6K


