|
|
|
| Modeling and parameter optimization for vibrating screens based on AFSA-SimpleMKL |
| LI Zhan-fu, TONG Xin |
| College of Mechanical Engineering and Automation, Huaqiao University, Xiamen 361021, China |
|
|
|
Abstract In view of the poor screening efficiency of vibrating screens and the incomplete screening theory, a comprehensive mathematical model is need to be established to guide the design of vibrating screens. Screening simulation experiment based on the Discrete Element Method was applied to solve problems like the complexity of screening process and the difficulty to obtain the screening data, and its validity was verified by an experimental prototype with adjustable parameters. In principle, the mathematical relationship between screening efficiency and parameters was a complex non-linear problem, instead of being limited to low precision as the traditional regression algorithm was, the nonlinear regression model of vibration screen with designing the sample space of operation parameters and screen configurations based on Simple Multiple Kernel Learning was introduced. Considering multi-extremum, large-scale, and non-differentiable of this computational model, the artificial fish-swarm algorithm with strong robustness and global convergence was applied to parameters optimization. Finally, the optimal vibration parameters were as follows: vibration amplitude was 2.5 mm, frequency was 22 Hz, vibration direction angle was 50°, screen panel square hole dimensions was 0.9 mm, wire diameter was 0.45 mm, inclination was 21.6°. In summary, this methodology could be applied to the research of vibrating screen. Additionally, authors of the scheme are confident that the results will be useful to improve the design and manufacture of vibrating screen.
|
|
Received: 27 October 2015
Published: 28 April 2016
|
|
|
基于AFSA-SimpleMKL对振动筛建模及筛机优化
针对目前振动筛筛分性能差及筛分理论不完善,亟待建立筛机参数与筛分效率间综合数学模型来指导振动筛的设计.基于离散单元法(Discrete Element Method,DEM)的筛分仿真实验解决筛分过程的复杂性和筛分数据难获得等问题,用可调参数的振动筛对仿真实验进行验证.筛分效率与筛分参数之间的数学关系是一个复杂的非线性问题,由于传统的回归算法对筛分数学模型预测精度低,利用能有效解决小样本问题和基于统计学理论的简单多核支持向量机(Simple Multiple Kernel Learning,SimpleMKL)对仿真实验获得的数据建立回归模型.但其模型是多极值且不可微分的多参数大规模计算问题,借用鲁棒性强和全局收敛性好的人工鱼群优化算法(Artificial Fish Swarm Algorithm,AFSA)对由SimpleMKL建立的筛分回归模型进行参数寻优,得出筛机振动和结构参数:振幅为2.5 mm,振动频率为22 Hz,振动方向角为50°,筛孔大小为0.9 mm,筛丝直径为0.4 mm,筛面倾角为21.6°.提高了振动筛的筛分效率,为振动筛的设计和制造提供了新思路.
关键词：
离散单元法,
简单多核支持向量机,
人工鱼群算法,
参数优化,
建模,
筛分效率
|
|
| [1] 刘光焕,童昕,辛成涛.振动筛筛分过程的数值模拟及进展[J].金属矿山,2008(9):104-110. LIU Guang-huan,TONG Xin,XIN Chen-tao.Numerical simulation of vibration screening process and its process[J].Metal Mine,2008(9):104-110.[2] CLEARY P W,SAWLEY M.DEM modelling of industrial granular flows:3D case studies and the effect of particle shape on hopper discharge[J].Applied Mathematical Modelling,2002,26(2):89-111.[3] LI J,WEBB C,PANDIELLA S S,et al.Discrete particle motion on sieves:a numerical study using the DEM simulation[J].Powder Technology,2003,133(1/3):190-202.[4] JIAO H G,MA J,ZHAO Y M.Study on the numerical simulation of batch sieving process[J].Journal of Coal Science & Engineering,2006,12(2):80-83.[5] CLEARY P W,SINNOTT M D,MORRISON R D.Separation performance of double deck banana screens:part 1:flow and separation for different accelerations[J].Minerals Engineering,2009,22 (14):1218-1229.[6] CLEARY P W,SINNOTT M D,MORRISON R D.Separation performance of double deck banana screens:part 2:quantitative predictions[J].Minerals Engineering,2009,22 (14):1230-1244.[7] CHEN Y H,TONG X.Modeling screening efficiency with vibrational parameters based on DEM 3D simulation[J].Mining Science and Technology,2010,20(4):615-620.[8] CHEN Y H,TONG X.Application of the DEM to screening process:a 3D simulation[J].Mining Science and Technology,2009,19(4):493-497.[9] 李菊,赵德安,沈惠平,等.基于输入优选的三维并联振动筛输出特性分析[J].中国机械工程,2014,25(14):1852-1857. LI Ju,ZHAO De-an,SHEN Hui-ping,et al.Analysis on output characteristics of three-dimensional parallel kinematics vibrating screen based on input optimization[J].China Mechanical Engineering,2014,25(14):1852-1857.[10] LI Z F,TONG X.Modeling and parameter optimization for the design of vibrating screens[J].Minerals Engineering,2015,83:149-155.[11] VAPNIK V.The nature of statistical learning theory[M].New York:Springer,1995:84-90.[12] VAPNIK V N.An overview of statistical learning theory[J].IEEE Trans Neural Networks,1999,10 (5):988-999.[13] 李元诚,方廷健,于尔铿.短期负荷预测的支持向量机方法[J].中国电机工程学报,2003,23(6):55-59. LI Yuan-cheng,FANG Ting-jian,YU Er-keng.Study of support vector machines for short-term load forecasting[J].Proceedings of the CSEE,2003,23(6):55-59.[14] 申中杰,陈雪峰,何正嘉,等.基于相对特征和多变量支持向量机的滚动轴承剩余寿命预测[J].机械工程学报,2013,49(2):183-189. SHEN Zhong-jie,CHEN Xue-feng,HE Zheng-jia,et al.Remaining life predictions of rolling bearing based on relative features and multivariable support vector machine[J].Chinese Journal of Mechanical Engineering,2013,49(2):183-189.[15] 陆凤仪,王 爽,徐格宁,等.构建起重机载荷谱v-SVRM预测模型的改进方法[J].工程设计学报,2015,22 (5):412-419. LU Feng-yi,WANG Shuang,XU Ge-ning,et al.Improved modeling method for prediction model of crane load spectrum based on support vector regression machine[J].Chinese Journal of Engineering Design,2015,22 (5):412-419.[16] RAKOTOMAMONJY A,BACH F R,CANU S,et al.SimpleMKL[J].Journal of Machine Learing Reseach,2008,9(11):2491-2521.[17] 李晓磊,邵之江,钱积新.一种基于动物自治体的寻优模式:鱼群算法[J].系统工程理论与实践,2002,22(11):32-38. LI Xiao-lei,SHAO Zhi-jiang,QIAN Ji-xin.An optimizing method based on autonomous animats:fish-swarm algorithm[J].Systems Engineering—Theory & Practice,2002,22(11):32-38.[18] CLEARY P W.DEM prediction of industrial and geophysical particle flows[J].Particuology,2010,8(2):106-118. |
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
| |
Shared |
|
|
|
|
| |
Discussed |
|
|
|
|
