1.天津职业技术师范大学 汽车模具智能制造技术国家地方联合工程实验室,天津 300222
2.天津职业技术师范大学 天津市高性能精准成形制造技术与装备重点实验室,天津 300222
姚磊,男,1998年生,山西大同人,在读硕士研究生;主要研究方向为金属材料服役性能评价;E-mail:861492538@qq.com。
收稿:2024-03-11,
修回:2024-05-20,
纸质出版:2026-01-15
移动端阅览
姚磊,黎振,武川,等. 基于互信息特征选择和递归特征消除的钢材疲劳强度预测[J]. 机械强度,2026,48(1):72-78.
YAO Lei,LI Zhen,WU Chuan,et al. Prediction of steel fatigue strength based on mutual information feature selection and recursive feature elimination[J]. Journal of Mechanical Strength,2026,48(1):72-78.
姚磊,黎振,武川,等. 基于互信息特征选择和递归特征消除的钢材疲劳强度预测[J]. 机械强度,2026,48(1):72-78. DOI: 10.16579/j.issn.1001.9669.2026.01.009.
YAO Lei,LI Zhen,WU Chuan,et al. Prediction of steel fatigue strength based on mutual information feature selection and recursive feature elimination[J]. Journal of Mechanical Strength,2026,48(1):72-78. DOI: 10.16579/j.issn.1001.9669.2026.01.009.
目的
2
为了解决疲劳强度预测中由于数据量不足和高维数据而引起的挑战,提出一种互信息特征选择和递归特征消除(Mutual Information Feature Selection and Recursive Feature Elimination
MIFS-RFE)相结合的方法。
方法
2
首先,采用MIFS识别对模型预测最关键的特征。其次,在RFE阶段重点处理剩余特征,通过系统迭代过程选择最具信息量的特征,确保最终特征子集能够准确地支持疲劳强度预测。最后,将经过特征选择和消除处理得到的最终特征子集输入到随机森林回归(Random Forest Regression
RFR)、K最近邻回归(K Nearest Neighbor Regression
KNN)、支持向量回归(Support Vector Regression
SVR)和多层感知机(Multi-Layer Perceptron
MLP)模型中,综合分析各模型的预测性能。
结果
2
经过优化后,特征数据从原始的25维降到了13维。在测试过程中,RFR、KNN、SVR和MLP的决定系数
R
2
分别为0.977 7、0.972 5、0.961 3和0.976 6。相较于全部特征的测试结果,所提方法使得模型的
R
2
最大提升了0.020 8。最后基于模型的Shapley加法解释(Shapley Additive Explanation
SHAP)值分析最优特征的影响性,验证了MIFS-RFE方法的有效性,说明所提方法在维持相对高性能的同时成功降低了特征维度,为疲劳强度预测提供了更有效的优化方案。
Objective
2
To address the challenges in fatigue strength prediction caused by limited data and high-dimensional feature space
a method combining mutual information feature selection and recursive feature elimination (MIFS-RFE) is proposed.
Methods
2
Firstly
MIFS was used to identify crucial features for prediction. Subsequently
the remaining features were processed in the RFE stage. Through an iterative process
the most informative features were selected to ensure accurate fatigue strength prediction. The finalized feature subset was input into random forest regression (RFR)
K-nearest neighbors regression (KNN)
support vector regression (SVR)
and multilayer perceptron (MLP) models for performance analysis.
Results
2
After optimization
the feature dimensionality was reduced from 25 to 13. During the test process
RFR
KNN
SVR
and MLP achieved
R
2
values of 0.977 7
0.972 5
0.961 3
and 0.976 6
respectively. Compared with the test results of all features
the proposed method increased the maximum
R
2
of the model by 0.020 8. Finally
based on the SHAP value
the influence of effective features was analyzed
and the effectiveness of the combination of MIFS and RFE was validated. The results indicate that the proposed method maintains high performance while reducing feature dimensionality
offering an optimized solution for fatigue strength prediction.
李道航 , 吴学雷 , 万芳 , 等 . 综合考虑疲劳强度影响因素的寿命预测方法 [J]. 装备环境工程 , 2020 , 17 ( 7 ): 109 - 115 .
LI Daohang , WU Xuelei , WAN Fang , et al . Life prediction method synthetically considering the influence factors of fatigue strength [J]. Equipment Environmental Engineering , 2020 , 17 ( 7 ): 109 - 115 . (In Chinese)
邓海龙 , 刘兵 , 郭扬 , 等 . 基于内部失效机理预测评估渗碳Cr-Ni齿轮钢的超高周疲劳强度 [J]. 材料研究学报 , 2023 , 37 ( 1 ): 55 - 64 .
DENG Hailong , LIU Bing , GUO Yang , et al . Prediction and evaluation of very-high cycle fatigue strength of carburized Cr-Ni gear steel based on interior failure mechanism [J]. Chinese Journal of Materials Research , 2023 , 37 ( 1 ): 55 - 64 . (In Chinese)
李永平 , 康贺铭 , 李明凯 , 等 . 基于均方误差修正的渗碳12CrNi 3 钢疲劳强度预测模型 [J]. 材料科学与工艺 , 2023 , 31 ( 2 ): 60 - 68 .
LI Yongping , KANG Heming , LI Mingkai , et al . Prediction model for fatigue strength of carburized 12CrNi 3 steel based on mean square error correction [J]. Materials Science and Technology , 2023 , 31 ( 2 ): 60 - 68 . (In Chinese)
徐友坤 , 鲁连涛 , 曾东方 , 等 . 基于临界距离法的含表面缺陷车轴试样疲劳强度预测 [J] . 机械强度 , 2022 , 44 ( 4 ): 825 - 830 .
XU Youkun , LU Liantao , ZENG Dongfang , et al . Fatigue strength prediction of railway axle specimen with surface notch by TCD [J]. Journal of Mechanical Strength , 2022 , 44 ( 4 ): 825 - 830 . (In Chinese)
孙振铎 , 侯东勃 , 李志远 . 渗碳Cr-Mn钢的夹杂分布特性及疲劳强度预测 [J]. 兵器材料科学与工程 , 2021 , 44 ( 1 ): 98 - 103 .
SUN Zhenduo , HOU Dongbo , LI Zhiyuan . Prediction for fatigue strength and distribution features of inclusion of carburized Cr-Mn steel [J]. Ordnance Material Science and Engineering , 2021 , 44 ( 1 ): 98 - 103 . (In Chinese)
张乐 , 马奔奔 , 何启源 , 等 . 基于机器学习的高强度螺栓疲劳寿命预测 [J] . 西华大学学报(自然科学版) , 2022 , 41 ( 6 ): 68 - 74 .
ZHANG Le , MA Benben , HE Qiyuan , et al . Fatigue life prediction of high-strength bolts based on machine learning [J]. Journal of Xihua University(Natural Science Edition) , 2022 , 41 ( 6 ): 68 - 74 . (In Chinese)
杨梦迪 , 贺敬良 , 陈勇 , 等 . 基于灰色神经网络的金属疲劳寿命预测研究 [J]. 冶金与材料 , 2021 , 13 ( 5 ): 14 - 16 .
YANG Mengdi , HE Jingliang , CHEN Yong , et al . Research on metal fatigue life prediction based on grey neural network [J]. Metallurgy and Materials , 2021 , 13 ( 5 ): 14 - 16 . (In Chinese)
刘志状 , 吴昊 . 一种基于参数影响的数据驱动下的疲劳寿命预测方法 [J]. 机械工程学报 , 2023 , 59 ( 4 ): 71 - 79 .
LIU Zhizhuang , WU Hao . Data-driven fatigue life prediction method based on the influence of parameters [J]. Journal of Mechanical Engi-neering , 2023 , 59 ( 4 ): 71 - 79 . (In Chinese)
RUIZ E , FERREÑO D , CUARTAS M , et al . Application of machine learning algorithms for the optimization of the fabrication process of steel springs to improve their fatigue performance [J]. International Journal of Fatigue , 2022 , 159 : 106785 .
LIAN Z H , LI M J , LU W C . Fatigue life prediction of aluminum alloy via knowledge-based machine learning [J]. International Journal of Fatigue , 2022 , 157 : 106716 .
GAN L , WU H , ZHONG Z . Fatigue life prediction considering mean stress effect based on random forests and kernel extreme learning machine [J]. International Journal of Fatigue , 2022 , 158 : 106761 .
AGRAWAL A , DESHPANDE P D , CECEN A , et al . Exploration of data science techniques to predict fatigue strength of steel from composition and processing parameters [J]. Integrating Materials and Manufacturing Innovation , 2014 , 3 ( 1 ): 90 - 108 .
GUO J W , LAO Z P , HOU M , et al . Mechanical fault time series prediction by using EFMSAE-LSTM neural network [J]. Measurement , 2021 , 173 : 108566 .
RTAYLI N , ENNEYA N . Enhanced credit card fraud detection based on SVM-recursive feature elimination and hyper-parameters optimization [J]. Journal of Information Security and Applications , 2020 , 55 : 102596 .
ANANDAN B , MANIKANDAN M . Machine learning approach with various regression models for predicting the ultimate tensile strength of the friction stir welded AA 2050-T8 joints by the K-Fold cross-validation method [J]. Materials Today Communications , 2023 , 34 : 105286 .
YAN F , SONG K , LIU Y , et al . Predictions and mechanism analyses of the fatigue strength of steel based on machine learning [J]. Journal of Materials Science , 2020 , 55 ( 31 ): 15334 - 15349 .
AGRAWAL A , CHOUDHARY A . An online tool for predicting fatigue strength of steel alloys based on ensemble data mining [J]. International Journal of Fatigue , 2018 , 113 : 389 - 400 .
0
浏览量
1
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621