物理模拟与数据驱动混合的Z型立柱片剪切刚度认知预测方法

doi:10.3901/JME.2024.02.050

摘要/Abstract

摘要： 高层工业货架的货载量通常远大于结构自重，立柱片剪切刚度对货架整体的侧向稳定性至关重要。由于结构尺寸及连接方式的多样性，目前建立较为通用的立柱片剪切刚度计算模型还比较困难。以常见的Z型立柱片为例，提出了物理模拟与数据驱动混合的剪切刚度认知预测方法，以期提高工业货架抗侧倾设计的科学决策能力。首先，建立了基于力学性能试验的立柱片剪切刚度有限元仿真模型，并获取了不同立柱片结构的剪切刚度分析数据集；其次，采用XGBOOST算法建立了立柱片剪切刚度智能化预测模型；相比于卷积神经网络和长短期记忆网络，其预测效率较高且更接近于有限元数值计算，便于立柱片剪切性能的快速评估。以此为基础，选取SHAP(Shapley additive explanation, SHAP)值可视化分析方法对立柱片预测模型计算结果进行理解与认知。重要性分数显示，改善立柱片剪切性能的关键在于斜横撑结构的设计；而立柱片跨距对剪切性能影响相对有限，可以根据需要灵活配置。另外SHAP摘要图和依赖图显示，分离型斜横撑配置模式下增大结构件规格尺寸可以明显提高立柱片的剪切刚度值。

关键词: 数据驱动, 剪切刚度, XGBOOST, 认知预测, 立柱片

Abstract: The living load of high-rise industrial rack is usually much greater than the dead weight of structure, so the shear stiffness of upright frame is crucial for the overall lateral stability of the rack. Due to the variety of the size and the connection of structural components, it is still difficult to build a general analytical model at present. Taking the upright frame in Z-pattern as an example, based on the physical simulation and data-driven method, a cognitive prediction model on shear stiffness of upright frame is proposed. Firstly, based on the mechanical test, the finite element simulations of shear stiffness of upright frame were made to obtain the shear stiffness data sets of different upright frame structures. Secondly, the XGBOOST algorithm was used to train the intelligent prediction model of shear stiffness. In contrast with the convolutional neural network and long short-term memory network, the prediction results of the XGBOOST algorithm are more efficient and closer to the finite element method. On this basis, the results of prediction model are interpreted and understood with the visual method of SHapley additive exPlanation (SHAP). By the importance score analysis, it is found that the shear performance of the upright frame is mainly determined on the design of the bracing configuration, and the distance between two uprights has relatively limited influence on the shear performance which could be flexibly configured according to the user’s requirements. What’s more, the SHAP summary plots and dependency plots showed that the shear stiffness of upright frame is significantly increased by enlarging the dimensions of structural members under the separation configuration of bracing members.

Key words: data-driven, shear stiffness, XGBOOST, cognitive prediction, upright frame

中图分类号:

TH122

吕志军, 褚铭, 周亚勤, 肖雷, 李宏亮. 物理模拟与数据驱动混合的Z型立柱片剪切刚度认知预测方法[J]. 机械工程学报, 2024, 60(2): 50-61.

Lü Zhijun, CHU Ming, ZHOU Yaqin, XIAO Lei, LI Hongliang. A Cognitive Prediction Method of Shear Stiffness of Upright Frames in Z-pattern Based on the Physical Simulation and Data-driven[J]. Journal of Mechanical Engineering, 2024, 60(2): 50-61.

参考文献

[1] 《物流技术与应用》编辑部.疫情冲击下的淬炼与成长-2020年中国物流装备市场回顾与2021年展望(上)[J].物流技术与应用, 2021, 26(3):54.
Editorial Department of Logistics & Material Handling. Review of China's logistics equipment market in 2020 and prospect in 2021[J]. Logistics & Material Handling, 2021, 26(3):54.
[2] BALDASSINO N, BERNUZZI C. Analysis and behaviour of steel storage pallet racks[J]. Thin-Walled Structures, 2000, 37(4):277-304.
[3] GILBERT B P, RASMUSSEN K J R, BALDASSINO N, et al. Determining the transverse shear stiffness of steel storage rack upright frames[J]. Journal of Constructional Steel Research, 2012, 78(10):107-116.
[4] TALEBIAN N, GILBERT B P, BALDASSINO N, et al. Factors contributing to the transverse shear stiffness of bolted cold-formed steel storage rack upright frames with channel bracing members[J]. Thin-Walled Structures, 2019, 136(3):50-63.
[5] SAJJA S R, BEALE R G, GODLEY M H R. Cross-aisle stiffness tests on rack upright frames[C]//University of Missouri-Rolla. Twentieth International Specialty Conference on Cold-Formed Steel Structures, November 3 & 4, 2010, University of Missouri-Rolla, St. Louis, Missouri. St. Louis:University of Missouri-Rolla, 2010:367-381.
[6] SAJJA S R, BEALE R G, GODLEY M H R. Shear stiffness of pallet rack upright frames[J]. Journal of Constructional Steel Research, 2008, 64(7):867-874.
[7] Rack Manufacturers Institute. ANSI-RMI MH16.1 specification for the design, testing and utilization of industrial steel storage racks[S]. Charlotte:Rack Manufacturers Institute, 2021.
[8] TALEBIAN N, GILBERT B P, BALDASSINO N, et al. Finite element modeling of bolted cold-formed steel storage rack upright frames[J]. Applied Mechanics and Materials, 2016, 4097(5):251-257.
[9] ROURE F, PEKÖZ T, SOMALO M R, et al. Cross-aisle stiffness analysis of industrial welded cold-formed steel rack upright frames[J]. Thin-Walled Structures, 2019, 141(8):332-344.
[10] 徐格宁, 杨瑞刚.约界参数CAP对大型钢结构系统可靠性分析的影响[J].机械工程学报, 2005, 53(12):130-134.
XU Gening, YANG Ruigang. Influence of bounded parameter CAP on reliability analysis of large steel structure system[J]. Journal of Mechanical Engineering, 2005(12):130-134.
[11] 蒋琛, 邱浩波, 高亮.随机不确定性下的可靠性设计优化研究进展[J].中国机械工程, 2020, 31(2):190-205.
JIANG Chen, QIU Haobo, GAO Liang. Research progresses in reliability-based design optimization under aleatory uncertainties[J]. China Mechanical Engineering, 2020, 31(2):190-205.
[12] 张朝龙, 赵筛筛, 何怡刚.基于信息熵与PSO-LSTM的锂电池组健康状态估计方法[J].机械工程学报, 2022, 58(10):180-190.
ZHANG Chaolong, ZHAO Shaishai, HE Yigang. State-of-health estimate for lithium-ion battery using information entropy and PSO-LSTM[J]. Journal of Mechanical Engineering, 2022, 58(10):180-190.
[13] QU D, CAI X, CHANG W. Evaluating the effects of steel fibers on mechanical properties of ultra-high performance concrete using artificial neural networks[J]. Applied Sciences, 2018, 8(7):1120-1140.
[14] 宋一铭, 吕志军, 陈东, 等.非连续截面薄壁钢构件稳定性智能预测方法[J].机械强度, 2018, 40(3):653-659.
SONG Yiming, LÜ Zhijun, CHEN Dong, et al. Intelligent prediction method on stability of thin-wall steel component with non continuous cross section[J]. Journal of Mechanical Strength, 2018, 40(3):653-659.
[15] SHAH S N R, SULONG N H R, EL-SHAFIE A. New approach for developing soft computational prediction models for moment and rotation of boltless steel connections[J]. Thin-Walled Structures, 2018, 135(13):206-215.
[16] LYU Z, ZHAO P, LU Q, et al. Prediction of the bending strength of boltless steel connections in storage pallet racks:An integrated experimental-FEM-SVM methodology[J]. Advances in Civil Engineering, 2020, 1:1-17.
[17] 郑怀亮, 王日新, 杨远涛, 等.数据驱动故障诊断方法泛化性能的经验性分析[J].机械工程学报, 2020, 56(9):102-117.
ZHENG Huailiang, WANG Rixin, YANG Yuantao, et al. An empirical analysis about the generalization performance of data-driven fault diagnosis methods[J]. Journal of Mechanical Engineering, 2020, 56(9):102-117.
[18] SUN H, BURTON H V, HUANG H. Machine learning applications for building structural design and performance assessment:State-of-the-art review[J]. Journal of Building Engineering, 2021, 33:101816.
[19] 丁浩.基于XGBoost多模型融合强化技术的个人信用评估研究[D].南京:南京信息工程大学, 2019.
DING Hao. Research on person credit evaluation based on enhanced technology of XGBoost multi-model fusion[D]. Nanjing:Nanjing University of Information Science and Technology, 2019.
[20] 刘诚然.基于深度学习的远场语音识别技术研究[D].郑州:战略支援部队信息工程大学, 2019.
LIU Chengran. Research on deep learning based far-filed speech recognition[D]. Zhengzhou:Information Engineering University, 2019.
[21] European Committee for Standardization. EN15512 Steel static storage systems[S]. Brussels:European Committee for Standardization, 2009.
[22] Standards Association of Australia. AS/NZS1252 High-strength steel bolts with associated nuts and washers for structural engineering[S]. Sydney:Standards Association of Australia, 1996.
[23] BARTH K E, ORBISON J G, NUKALA R, et al. Behavior of steel tension members subjected to uniaxial loading[J]. Journal of Constructional Steel Research, 2002, 58(5):1103-1120.
[24] 王青松, 谢兴生, 佘颢.基于CNN-XGBoost混合模型的短时交通流预测[J].测控技术, 2019, 38(4):37-40, 67.
WANG Qingsong, XIE Xingsheng, SHE Hao. Short-term traffic flow prediction based CNN-XGBoost hybrid model[J]. Measurement & Control Technology, 2019, 38(4):37-40, 67.
[25] ZHOU S, LIU Z, WANG M, et al. Impacts of building configurations on urban stormwater management at a block scale using XGBoost[J]. Sustainable Cities and Society, 2022, 87:104235
[26] 刘彧祺, 张智斌, 陈昊昱, 等.基于XGBoost集成的可解释信用评分模型[J].数据通信, 2019, 40(3):27-32.
LIU Jianqi, ZHANG Zhibin, CHEN Haoyu, et al. Interpretable credit scoring model based on XGBoost[J]. Dechnology Discussion, 2019, 40(3):27-32.
[27] MÖLLER A, RUHLMANN-KLEIDER V, LELOUP C, et al. Photometric classification of type Ia supernovae in the SuperNova Legacy Survey with supervised learning[J]. Journal of Cosmology & Astroparticle Physics, 2016(12):008.
[28] TAMAYO D, SILBURT A, VALENCIA D, et al. A machine learns to predict the stability of tightly packed planetary systems[J]. Astrophysical Journal, 2016, 832(2):L22.
[29] PETER B, HOUGAARD J L, SMILGINS A. Applied cost allocation:The DEA-aumann-shapley approach[J]. European Journal of Operational Research, 2016, 254(2):667-678.
[30] 陈洞天, 单杰, 周文丹.基于Xgboost的心血管疾病预测模型和指标分析研究[J].现代医院, 2021, 21(6):958-961.
CHEN Dongtian, SHAN Jie, ZHOU Wendan. An XGBoost-based cardiovascular disease prediction model and index analysis research[J]. Modern Hospitals, 2021, 21(6):958-961
[31] 罗妍, 王枞, 叶文玲.基于XGBoost和SHAP的急性肾损伤可解释预测模型[J].电子与信息学报, 2022, 44(1):27-38.
LUO Yan, WANG Cong, YE Wenling. An interpretable prediction model for acute kidney injury based on XGBoost and SHAP[J]. Journal of Electronics & Information Technology, 2022, 44(1):27-38.