Parameter Optimization of Vibratory Sorter using Neural Network and Particle Swarm Optimization
-
摘要: 针对废旧铅酸蓄电池回收预处理工艺中极群分离问题设计了振动式分选机,并应用椭圆基神经网络近似模型和粒子群算法研究了振动式分选机参数优化问题。首先,以降低质量、降低应力和提高第3阶固有频率为目标,利用灵敏度分析获得影响程度最高的结构参数;其次,采用具有最高拟合度的椭圆基神经网络构建试验设计样本的最佳近似模型,并采用优化效果更佳的粒子群算法进行数值优化;最后,对优化前后的结构进行可靠性分析。结果表明:优化后结构质量降低了9.7%,最大等效应力降低了36.7%,第3阶固有频率提高了12.9%,结构可靠度提高了5.9%。Abstract: A vibratory sorter was designed for the separation of polar groups in the recycling pretreatment process of used lead-acid batteries, and the ellipsoid basis function(EBF)neural network approximation model and particle swarm optimization(PSO)algorithm were used to optimized structural parameters of the vibratory sorter. Firstly, to reduce the mass, reduce the stress and increase the third-order natural frequency, the sensitivity analysis is done to obtain the structural parameters with the highest degree of influence. Secondly, the optimal approximation model of the experimental design sample is constructed by using EBF neural networks with the highest fitting degree, and PSO algorithm with better optimization effect is adopted for numerical optimization. Finally, the reliability analysis was performed on the structures before and after the optimization. The results show that the quality of the optimized values is reduced by 9.7%, the maximum equivalent stress is reduced by 36.7%, the third-order natural frequency is increased by 12.9%, and the structural reliability is increased by 5.9%.
-
表 1 材料属性设置
材料 密度/(kg·m-3) 弹性模量/Pa 泊松比 Q235 7 850 2.0×1011 0.3 橡胶 1 112 2.6×106 0.499 表 2 前6阶固有频率和振型
阶数 固有频率/Hz 模态振型描述 1 5.95 入口沿YZ平面左右摆动 2 6.98 整体沿XZ平面上下平动 3 20.097 出口沿YZ平面左右摆动 4 23.843 整体沿XY平面上下摆动 5 37.235 整体沿YZ平面上下摆动 6 45.269 整体沿Z轴扭动 表 3 设计变量的初始值和范围
变量 参数定义 初始值/mm 取值范围/mm t1 出入口上板厚度 5 3≤t1≤7 t2 侧方连接板厚度 10 4≤t2≤12 t3 弹簧处连接板厚度 6 4≤t3≤12 t4 振动电机上方加强板厚度 15 5≤t4≤18 t5 下主梁厚度 5 3≤t5≤7 t6 上主梁厚度 4 3≤t6≤7 t7 调距横梁厚度 15 5≤t7≤18 表 4 设计变量的灵敏度
变量 灵敏度 质量M 最大等效应力S 第3阶固有频率F3 t1 0.0043 -0.064 3 -0.007 6 t2 0.1089 -0.103 6 -0.043 8 t3 0.06 -1.373 3 0.077 6 t4 0.0187 -0.041 1 0.020 3 t5 0.0488 -0.123 3 0.130 1 t6 0.0365 -0.411 5 -0.018 9 t7 0.0807 -0.097 -0.017 表 5 近似模型拟合效果评估
类型 决策系数 平均误差/10-2 最大误差/10-2 M S F3 F2 M S F3 F2 M S F3 F2 RSM 1 0.949 0.937 0.999 0.001 6.03 4.275 0.149 0.008 20.58 15.065 0.51 Kriging 0.949 0.95 0.938 0.949 3.97 5.26 3.761 3.742 21.11 24.98 23.89 17.968 RBF 0.999 0.933 0.952 0.998 0.662 6.96 3.769 0.831 3.467 22.549 12.684 3.92 EBF 1 0.984 0.972 0.999 0.104 3.22 2.662 0.87 0.349 10.55 11.176 3.683 表 6 基于EBF近似模型的数值优化对比
参数 初始值 PSO优化 NSGA-II优化 x1/mm 10 4 4.011 7 x2/mm 6 9.151 5 9.327 x3/mm 5 6.286 4 6.318 5 x4/mm 4 5.959 7 5.927 6 x5/mm 15 5 5.052 8 M/kg 173.54 156.61 156.93 S/MPa 125.99 76.807 76.648 F3/Hz 20.097 22.616 22.611 表 7 优化值相对误差分析
目标 M/kg S/MPa F3/Hz 优化前初始值 173.54 125.99 20.097 有限元计算值 156.77 79.757 22.628 PSO算法优化值 156.61 76.807 22.616 计算相对误差/% 0.1 3.7 0.05 -
[1] 谈大伟.废旧铅酸电池回收预处理锯切系统的研究[D].武汉: 武汉科技大学, 2013 http://cdmd.cnki.com.cn/Article/CDMD-10488-1014147634.htmTan D W. The study of sawing system for used lead-acid batteries' recycling pretreatment[D]. Wuhan: Wuhan University of Science and Technology, 2013(in Chinese) http://cdmd.cnki.com.cn/Article/CDMD-10488-1014147634.htm [2] 向东, 吴育家, 杨继平, 等.印刷线路板振动拆解模型及其拆解过程分析[J].机械工程学报, 2017, 53(9):127-134 http://d.old.wanfangdata.com.cn/Periodical/jxgcxb201709015Xiang D, Wu Y J, Yang J P, et al. The disassembly model and its analysis of PCB by vibration[J]. Journal of Mechanical Engineering, 2017, 53(9):127-134(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/jxgcxb201709015 [3] Noga L B S, Skrzat A, Stachowicz F. Dynamic analysis of vibrating screener system[J]. Journal of Physics:Conference Series, 2013, 451(1):012028 [4] 柴保明, 史朋波, 张娴, 等.基于ANSYS Workbench的ZKB振动筛的模态分析与优化[J].煤炭工程, 2017, 49(9):162-165 http://d.old.wanfangdata.com.cn/Periodical/mtgc201709043Chai B M, Shi P B, Zhang X, et al. Modal analysis and optimization of ZKB vibrating screen based on ANSYS Workbench[J]. Coal Engineering, 2017, 49(9):162-165(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/mtgc201709043 [5] Baragetti S, Villa F. A dynamic optimization theoretical method for heavy loaded vibrating screens[J]. Nonlinear Dynamics, 2014, 78(1):609-627 doi: 10.1007/s11071-014-1464-4 [6] 贺孝梅, 刘初升, 张成勇.基于多频约束和解析灵敏度法的大型振动筛优化设计[J].中南大学学报, 2011, 42(3):664-670 http://d.old.wanfangdata.com.cn/Periodical/zngydxxb201103018He X M, Liu C S, Zhang C Y. Optimal design of large vibrating screen based on multiple frequencies constraints and analytical sensitivity[J]. Journal of Central South University, 2011, 42(3):664-670(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/zngydxxb201103018 [7] 贺孝梅, 刘初升, 张成勇.基于增广拉格朗日乘子法的大型振动筛动态优化设计[J].中国矿业大学学报, 2009, 38(1):80-85 http://d.old.wanfangdata.com.cn/Periodical/zgkydxxb200901016He X M, Liu C S, Zhang C Y. Dynamic optimal design of large vibrating screen based on an augmented Lagrangian method[J]. Journal of China University of Mining & Technology, 2009, 38(1):80-85(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/zgkydxxb200901016 [8] 王立军, 张传根, 丁振军.玉米收获机清选筛体结构优化[J].农业机械学报, 2016, 47(9):108-114 http://d.old.wanfangdata.com.cn/Periodical/nyjxxb201609016Wang L J, Zhang C G, Ding Z J. Structure optimization of cleaning screen for maize harvester[J]. Transactions of the Chinese Society for Agricultural Machinery, 2016, 47(9):108-114(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/nyjxxb201609016 [9] 龙腾, 刘建, Gary W G, 等.基于计算试验设计与代理模型的飞行器近似优化策略探讨[J].机械工程学报, 2016, 52(14):79-105 http://d.old.wanfangdata.com.cn/Periodical/jxgcxb201614011Long T, Liu J, Gary W G, et al. Discuss on approximate optimization strategies using design of computer experiments and metamodels for flight vehicle design[J]. Journal of Mechanical Engineering, 2016, 52(14):79-105(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/jxgcxb201614011 [10] 姚梓萌, 李言, 杨明顺, 等.响应面法在单点增量成形质量控制多目标优化中的应用[J].机械科学与技术, 2017, 36(3):409-416 http://journals.nwpu.edu.cn/jxkxyjs/CN/abstract/abstract6967.shtmlYao Z M, Li Y, Yang M S, et al. Multi-objective optimization of forming quality in single point incremental forming via response surface methodology[J]. Mechanical Science and Technology for Aerospace Engineering, 2017, 36(3):409-416(in Chinese) http://journals.nwpu.edu.cn/jxkxyjs/CN/abstract/abstract6967.shtml [11] 金浩, 向宇, 蒋红华, 等.轿车前保险杠碰撞性能多目标优化[J].机械科学与技术, 2017, 36(6):943-949 http://journals.nwpu.edu.cn/jxkxyjs/CN/abstract/abstract6743.shtmlJin H, Xiang Y, Jiang H H, et al. Multi-objective optimization of side door bumper beam for a passenger car based on response surface method[J]. Mechanical Science and Technology for Aerospace Engineering, 2017, 36(6):943-949(in Chinese) http://journals.nwpu.edu.cn/jxkxyjs/CN/abstract/abstract6743.shtml [12] 张春宜, 宋鲁凯, 费成巍, 等.基于智能双重响应面法的涡轮叶盘可靠性灵敏度分析[J].推进技术, 2017, 38(5):1155-1164 http://d.old.wanfangdata.com.cn/Periodical/tjjs201705025Zhang C Y, Song L K, Fei C W, et al. Intelligent dual response surface method for reliability sensitivity analysis of turbine blisk[J]. Journal of Propulsion Technology, 2017, 38(5):1155-1164(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/tjjs201705025 [13] 张春宜, 宋鲁凯, 费成巍, 等.柔性机构动态可靠性分析的先进极值响应面方法[J].机械工程学报, 2017, 53(7):47-54 http://d.old.wanfangdata.com.cn/Periodical/jxgcxb201707007Zhang C Y, Song L K, Fei C W, et al. Advanced extremum response surface method for dynamic reliability analysis on flexible mechanism[J]. Journal of Mechanical Engineering, 2017, 53(7):47-54(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/jxgcxb201707007 [14] 汪怡平, 王涛, 黎帅.基于自由变形技术的汽车气动减阻优化[J].机械工程学报, 2017, 53(9):135-143 http://d.old.wanfangdata.com.cn/Periodical/jxgcxb201709016Wang Y P, Wang T, Li S. Aerodynamic drag reduction of vehicle based on free form deformation[J]. Journal of Mechanical Engineering, 2017, 53(9):135-143(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/jxgcxb201709016 [15] 陈波, 高殿荣, 杨超, 等.大功率远程射雾器结构参数多目标智能协同优化[J].机械工程学报, 2017, 53(6):166-175 http://d.old.wanfangdata.com.cn/Periodical/jxgcxb201706022Chen B, Gao D R, Yang C, et al. Multi-objective intelligent collaborative optimization of structure parameters for high-power remote sprayer[J]. Journal of Mechanical Engineering, 2017, 53(6):166-175(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/jxgcxb201706022 [16] 郭谨玮, 刘昱, 徐月云, 等.应用多目标粒子群算法的车辆传动系参数优化仿真研究[J].机械科学与技术, 2018, 37(4):629-634 http://journals.nwpu.edu.cn/jxkxyjs/CN/abstract/abstract7000.shtmlGuo J W, Liu Y, Xu Y Y, et al. Parameter optimization and simulation system of vehicle transmission system using multi-objective particle swarm optimization algorithm[J]. Mechanical Science and Technology for Aerospace Engineering, 2018, 37(4):629-634(in Chinese) http://journals.nwpu.edu.cn/jxkxyjs/CN/abstract/abstract7000.shtml