基于改进深度代理模型的挖掘机铲斗结构优化设计

doi:10.13229/j.cnki.jdxbgxb.20240322

摘要/Abstract

摘要：

为优化挖掘机铲斗结构设计并降低挖掘能耗，搭建了基于SOLIDWORKS、ADAMS、EDEM及MATLAB的联合仿真与优化平台，建立了以铲斗所受最大挖掘载荷、强度为约束条件，以铲斗质量、单位挖掘能耗为优化目标的某中型反铲液压挖掘机全局优化方法。通过多体动力学与颗粒动力学联合仿真，得到不同工况下挖掘机作业时的载荷特性；皮尔逊相关系数表明，试验与仿真得到的斗杆、动臂关键铰点载荷具有极强相关性，验证了以联合仿真代替真实挖掘作业的可行性。采用多层感知器（MLP）作为代理模型，并利用Adam、Hyperband算法优化MLP的参数和超参数。工况Ⅱ下，MLP对铲斗质量、最大挖掘载荷、挖掘质量及挖掘能耗的决定系数R²分别为0.999、0.943、0.933、0.984；工况Ⅳ下，MLP的R²分别为0.999、0.944、0.918、0.925。将优化后的MLP结合NSGA-Ⅱ算法对铲斗结构进行迭代优化，结果表明：优化后的铲斗在满足最大挖掘载荷及强度要求的前提下，质量减小7.06%，单位挖掘能耗减小6.47%。

关键词: 机械设计及理论, 挖掘机铲斗, 代理模型, 优化设计, 适应性矩估计算法

Abstract:

In order to optimize the design of excavator bucket structure and reduce the excavation energy consumption， this paper builds a joint simulation and optimization platform based on SOLIDWORKS， ADAMS， EDEM and MATLAB， and establishes a global optimization method for a medium-sized backhoe hydraulic excavator， which takes the maximal excavation load and strength of the bucket as the constraints， and the quality of the bucket and the unit excavation energy consumption as the optimization objectives. Through combined simulation of multibody dynamics and particle dynamics， load characteristics of the excavator during operation under different working conditions were obtained. Pearson correlation coefficients show a very strong correlation between experimental and simulated loads on the key hinge points of the bucket rod and boom， confirming the feasibility of using combined simulation to replace actual excavation operations. Multilayer perceptron （MLP） is used as the surrogate model and the parameters and hyper-parameters of MLP are optimized using Adam， Hyperband algorithm. The coefficients of determination R² of MLP for bucket mass， maximum excavation load， excavation mass and excavation energy consumption are 0.999， 0.943， 0.933， 0.984 for Case Ⅱ； the R² of MLP are 0.999， 0.944， 0.918， 0.925 for Case Ⅳ. The optimized MLP is combined with the NSGA-Ⅱ algorithm to iteratively optimize the bucket structure. The results show that the optimized bucket achieves a mass reduction of 7.06 % and a unit excavation energy consumption reduction of 6.47 % under the premise of meeting the requirements of the maximum excavation load and strength.

Key words: machine design and theory, excavator bucket, surrogate model, optimized design, Adam algorithm

中图分类号:

TU621

陈一馨,陈再续,刘永生,杨帅,郭浩杰,贾晋三. 基于改进深度代理模型的挖掘机铲斗结构优化设计[J]. 吉林大学学报(工学版), 2025, 55(11): 3485-3497.

Yi-xin CHEN,Zai-xu CHEN,Yong-sheng LIU,Shuai YANG,Hao-jie GUO,Jin-san JIA. Structural optimization design of excavator bucket based on improved depth surrogate model[J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(11): 3485-3497.

图/表 32

图1

图2

表1

表2

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

图13

表3

表4

图14

图15

图16

表5

图17

图18

图19

表6

表7

图20

表8

表9

表10

图21

图22

参考文献 28

[1]	王同建, 杨书伟, 谭晓丹, 等. 基于DEM-MBD联合仿真的液压挖掘机作业性能分析[J]. 吉林大学学报: 工学版, 2022, 52(4): 811-818.
	Wang Tong-jian, Yang Shu-wei, Tan Xiao-dan, et al. Performance analysis of hydraulic excavator based on DEM-MBD co-simulation[J]. Journal of Jilin University (Engineering and Technology Edition), 2022, 52(4): 811-818.
[2]	蔡敢为, 黄一洋, 田军伟, 等. 一种新型正铲液压挖掘机工作机构的研究[J]. 机械工程学报, 2021, 57(13): 132-143.
	Cai Gan-wei, Huang Yi-yang, Tian Jun-wei, et al. Research on a new working mechanism of face-shovel hydraulic excavator[J]. Journal of Mechanical Engineering, 2021, 57(13): 132-143.
[3]	Palomba I, Richiedei D, Trevisani A, et al. Estimation of the digging and payload forces in excavators by means of state observers[J]. Mechanical Systems and Signal Processing, 2019, 134: No.106356.
[4]	Saldaña R A, Bustos G A, Peña L D J, et al. Structural design of an agricultural backhoe using TA, FEA, RSM and ANN[J]. Computers and Electronics in Agriculture, 2020, 172: No.105278.
[5]	Yuan Y L, Lv L Y, Wang S, et al. Multidisciplinary co-design optimization of structural and control parameters for bucket wheel reclaimer[J]. Frontiers of Mechanical Engineering, 2020, 15(3): 1-11.
[6]	Sun W, Peng X, Dou J, et al. Surrogate-based weight reduction optimization of forearm of bucket-wheel stacker reclaimer[J]. Structural and Multidisciplinary Optimization, 2020, 61(3): 1287-1301.
[7]	Shi Y P, Xia Y M, Zhang Y M, et al. Intelligent identification for working-cycle stages of excavator based on main pump pressure[J]. Automation in Construction, 2020, 109: No.102991.
[8]	Si J K, Zhao S Z, Feng H C, et al. Multi-objective optimization of surface-mounted and interior permanent magnet synchronous motor based on Taguchi method and response surface method[J]. Chinese Journal of Electrical Engineering, 2018, 4(1): 67-73.
[9]	Forrester I A, Keane J A. Recent advances in surrogate-based optimization[J]. Progress in Aerospace Sciences, 2008, 45(1): 50-79.
[10]	Libano F, Rech P, Tambarar L, et al. On the reliability of linear regression and pattern recognition feedforward artificial neural networks in FPGAs[J]. IEEE Transactions on Nuclear Science, 2018, 65(1): 288-295.
[11]	Li F, Zurada J M, Liu Y, et al. Input layer regularization of multilayer feedforward neural networks[J]. IEEE Access, 2017, 5: 10979-10985.
[12]	单德山, 张潇, 顾晓宇, 等. 基于多层感知深度学习的大跨度斜拉桥索力调整[J]. 桥梁建设, 2021, 51(1): 14-20.
	Shan De-shan, Zhang Xiao, Gu Xiao-yu, et al. Cable force adjustment for long-span cable-stayed bridge based on multilayer perceptron deep learning[J]. Bridge Construction, 2021, 51(1): 14-20.
[13]	林景亮, 黄运保, 李海艳, 等. 基于深度代理模型的叉车臂架液压系统设计优化[J]. 中国机械工程, 2022, 33(3): 290-298.
	Lin Jing-liang, Huang Yun-bao, Li Hai-yan, et al. Design optimization for hydraulic systems of forklift boom based on deep surrogate model[J]. China Mechanical Engineering, 2022, 33(3): 290-298.
[14]	晏福, 徐建中, 李奉书. 混沌灰狼优化算法训练多层感知器[J].电子与信息学报, 2019, 41(4): 872-879.
	Yan Fu, Xu Jian-zhong, Li Feng-shu. Training multi-layer perceptrons using chaos grey wolf optimizer[J]. Journal of Electronics & Information Technology, 2019, 41(4): 872-879.
[15]	Moayedi H, Abdullahi M M, Nguyen H, et al. Comparison of dragonfly algorithm and harris hawks optimization evolutionary data mining techniques for the assessment of bearing capacity of footings over two-layer foundation soils[J]. Engineering with Computers, 2019, 37(1): 1-11.
[16]	Pandey C A, Tikkiwal A V. Stance detection using improved whale optimization algorithm[J]. Complex Intelligent Systems, 2021, 7(3): 1-24.
[17]	Xue Y, Tong Y L, Ferrante N. An ensemble of differential evolution and Adam for training feed-forward neural networks[J]. Information Sciences, 2022, 608: 453-471.
[18]	Wang L, Liu Z C. Data-driven product design evaluation method based on multi-stage artificial neural network[J]. Applied Soft Computing Journal, 2021, 103: No.107117.
[19]	南敬昌, 杜有益, 王明寰, 等. 深度学习架构神经网络对超宽带天线建模优化[J]. 激光与光电子学进展, 2022, 59(13): 362-368.
	Jing-chang Nan, Du You-yi, Wang Ming-huan, et al. Deep learning architecture and neural network optimization of ultra-wideband antenna modeling[J]. Laser & Optoelectronics Progress, 2022, 59(13): 362-368.
[20]	孔翔. 大型矿用挖掘机铲斗结构优化设计[D]. 大连: 大连理工大学机械工程学院,2018.
	Kong Xiang. Structure optimization design of large mining excavator bucket[D]. Dalian: School of Mechanical Engineering, Dalian University of Technology, 2018.
[21]	雷睿. 液压挖掘机铲斗挖掘效率优化设计研究[D]. 杭州: 浙江大学机械工程学院,2018.
	Lei Rui. Optimal design of excavation efficiency of hydrualic excavator bucket[D]. Hangzhou: College of Mechanical Engineering, Zhejiang University, 2018.
[22]	陈一馨, 刘永生, 吕彭民. 挖掘机载荷谱试验的作业介质及作业类型[J]. 筑路机械与施工机械化, 2018, 5(12): 117-122.
	Chen Yi-xin, Liu Yong-sheng, Lv Peng-min, et al. Operation medium and type of load spectrum test of excavator[J]. Road Machinery & Construction Mechanization, 2018, 5(12): 117-122.
[23]	刘坤宇, 苏宏杰, 李飞宇, 等. 基于响应曲面法的土壤离散元模型的参数标定研究[J]. 中国农机化学报, 2021, 42(9): 143-149.
	Liu Kun-yu, Su Hong-jie, Li Fei-yu, et al. Research on parameter calibration of soil discrete element model based on response surface method[J]. Journal of Chinese Agricultural Mechanization, 2021, 42(9): 143-149.
[24]	顿国强, 陈海涛, 纪文义. 基于EDEM仿真与SolidWorks Simulation的凿式深松铲有限元分析[J]. 河南农业大学学报, 2017, 51(5): 678-682.
	Guo-qiang Dun, Chen Hai-tao, Ji Wen-yi. Finite element analysis of chisel-type deep shovel based on EDEM and Solid Works Simulation[J]. Journal of Henan Agricultural University, 2017, 51(5): 678-682.
[25]	Coetzee C. Review: calibration of the discrete element method[J]. Powder Technology, 2017, 310: 104-142.
[26]	张锐, 韩佃雷, 吉巧丽, 等. 离散元模拟中沙土参数标定方法研究[J]. 农业机械学报, 2017, 48(3): 49-56.
	Zhang Rui, Han Dian-lei, Ji Qiao-li, et al. Calibration methods of sandy soil parameters in simulation of discrete element method[J]. Transactions of the Chinese Society for Agricultural Machinery, 2017, 48(3): 49-56.
[27]	Maasa A L, Hannun A Y, Ng A Y. Rectifier n onlinearities improve neural network acoustic models[C]∥Proceedings of the 30th International Conference on Machine Learning. Atlanta, USA: PMLR, 2013: 456-462.
[28]	Li L S, Kevin J, Giulia D, et al. Hyperband: a novel bandit-based approach to hyperparameter optimization[J]. Journal of Machine Learning Research, 2018, 18(1): 6765-6816.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

材料参数	密度/（kg·m^-3）	泊松比	剪切模量/Pa
铲斗	7 801	0.29	2.09×10¹¹
Ⅱ类土	1 800	0.25	1×10⁷

接触参数	恢复系数	静摩擦因数	动摩擦因数
Ⅱ类土-Ⅱ类土	0.35	0.8	0.1
Ⅱ类土-铲斗	0.5	0.7	0.3

尺寸变量	R₁/mm	L/mm	B/mm	D₁/mm	A/（°）	R₂/mm	D₂/mm
原始尺寸	487	376	1 042	9	4.73	1 800	15
上边界	537	406	1 142	11	10	1 950	18
下边界	437	346	942	7	0	1 650	12

超参数	下界	上界
学习率α	1×10^-6	1×10^-3
网络层数k	3	5
隐藏层神经元数量n^（^k^）	10	100

评估方法	铲斗质量	最大挖掘载荷	挖掘质量	挖掘能耗
R²	0.999	0.942	0.933	0.984
RMSE	1.241	1 732.905	21.034	1 890.9
MAE	2.803	4 975.78	40.900	4 007.9
MRAE	0.0027	0.052 9	0.040	0.018