丁基防水嵌缝胶的性能特征分析

发布时间 : 2026-8-11 18:40:55
Views:

(Summary description)Customized SILICONE STRUCTURAL SEALANT from China manufacturer refers to the adhesive with high strength, can bear large load, aging resistance, fatigue resistance and corrosion resistance, and has stable performance within the expected life. It is suitable for bearing the bonding of strong structural parts. Non structural adhesive has low strength and poor durability. It can only be used for bonding, sealing and fixing of ordinary and temporary properties, and cannot be used for structural adhesive bonding. NEUTRAL SILICONE WETHERPROOF SEALANT from China manufacturer is a structural adhesive with waterproof function. Silicone glue refers to the adhesive based on silicone rubber. According to different functions, it can be divided into structural adhesive, potting adhesive, sealant, thermal conductive adhesive, flame retardant adhesive and so on.

丁基防水嵌缝胶的性能特征分析

(Summary description)Customized SILICONE STRUCTURAL SEALANT from China manufacturer refers to the adhesive with high strength, can bear large load, aging resistance, fatigue resistance and corrosion resistance, and has stable performance within the expected life. It is suitable for bearing the bonding of strong structural parts. Non structural adhesive has low strength and poor durability. It can only be used for bonding, sealing and fixing of ordinary and temporary properties, and cannot be used for structural adhesive bonding. NEUTRAL SILICONE WETHERPROOF SEALANT from China manufacturer is a structural adhesive with waterproof function. Silicone glue refers to the adhesive based on silicone rubber. According to different functions, it can be divided into structural adhesive, potting adhesive, sealant, thermal conductive adhesive, flame retardant adhesive and so on.

发布时间 : 2026-8-11 18:40:55
Views:

XGBoost模型是一种基于决策树的机器学习算法，其通过梯度下降策略对损失函数进行优化，并融合正则化项与二阶导数的信息以提升预测精度。丁基防水嵌缝胶常温模量的XGBoost预测值与实际测试值的对比结果如图所示，丁基防水嵌缝胶常温拉伸强度及最大强度伸长率的XGBoost预测值与实际测试值的对比结果如图所示。

由图可知:在该模型下，丁基防水嵌缝胶常温模量、丁基防水嵌缝胶的常温拉伸强度和最大强度伸长率预测值与实际值具有较高一致性，因此该模型适用于硅酮丁基防水嵌缝胶的性能预测。决定系数Rz越趋近于1，表征模型预测结果与实际值的吻合度越高，模型适用性越强。

各算法对丁基防水嵌缝胶常温模量预测的决定系数R和均方误差MSE如表1所示，对丁基防水嵌缝胶常温最大强度伸长率预测的决定系数R和均方误差MSE如表所示，对丁基防水嵌缝胶常温最大强度伸长率预测的决定系数R`和均方误差MSE如表所示。

SHAP特征分析作为一种模型可解释性方法，其核心作用在于量化各输人特征对模型预测结果的边际贡献，从而阐明模型的决策机制，有效缓解机器学习模型的“黑箱”问题基于XGBoost模型的预测结果，丁基防水嵌缝胶与丁基防水嵌缝胶的性能特征分析结果如图所示。

由图可知:针对丁基防水嵌缝胶常温模量，加速老化后的拉伸强度、弹性模量及断裂伸长率对模型预测结果的贡献度最为显著。具体而言，加速后拉伸强度与模型预测的常温模量呈负相关，即该指标越低，模型倾向于输出更低的模量预测值;加速后模量则与预测结果呈正相关，指标越高，模型越倾向于高模量预测，反之亦然。此外，加速后断裂伸长率与预测模量也呈负相关，指标越高，模型预测值。http://www.sdyuantai.com/

Scan the QR code to read on your phone

上一篇: 硅烷改性聚氨酯密封胶热加速养护试件的各项性能指标

下一篇: 低模量耐候MS密封胶模型预测的机理解释

上一篇: 硅烷改性聚氨酯密封胶热加速养护试件的各项性能指标

下一篇: 低模量耐候MS密封胶模型预测的机理解释