摘要
【目的】针对传统铸件分拣平台运动自由度受限及调姿精度不足的问题,设计一种多维振动调姿平台并提出相应的优化算法。【方法】首先,研发了一种具有拨动机构的创新结构,并引入多维振动激励模块以实现铸件位姿调整;其次,提出了一种自适应动态调整权重的改进粒子群优化(Improved Particle Swarm Optimization, IMPSO)算法,通过引入自适应干扰力和动态权重机制,提升了算法的收敛性能;再次,建立了以位置和姿态偏差为目标的适应度函数,通过Matlab软件仿真对比了IMPSO算法与粒子群优化(Particle Swarm Optimization, PSO)、遗传粒子群优化(Genetic Algorithm-Particle Swarm Optimization, GA-PSO)、混沌粒子群优化(Chaotic Particle Swarm Optimization, CPSO)算法的性能;最后,搭建了样机试验平台进行实测验证。【结果】仿真结果表明,相比于传统算法,IMPSO算法的迭代收敛速度提升了10%~27%;样机试验显示,平台对铸件的位置与姿态调整最大偏差分别控制在0.396 74 cm与0.244 13°以内,能够满足复杂工业场景下的高精度柔性分拣需求。
[Objective]To address the issues of limited degrees of freedom and insufficient accuracy in traditional casting sorting platforms,a multi-dimensional vibration posture adjustment platform and an associated optimization algorithm are proposed.[Methods]Firstly,an innovative toggle mechanism and a multi-dimensional vibration excitation module were developed to achieve flexible posture adjustment.Secondly,an improved particle swarm optimization(IMPSO)algorithm was proposed,incorporating adaptive interference forces and dynamic weight adjustment to enhance global search capabilities.Then,a fitness fuction targeting position and posture deviations was established,comparative simulations between IMPSO,particle swarm optimization(PSO),genetic algorithm-particle swarm optimization(GA-PSO),and chaotic particle swarm optimization(CPSO)were performed using Matlab.Finally,a physical prototype was constructed to verify the performance through experimental tests.[Results]The results show that the IMPSO algorithm improves the convergence speed by 10%to 27%compared with other algorithms.Prototype tests demonstrate that the maximum position and orientation deviations are controlled within 0.35 cm and 0.25°,respectively.This study provides a theoretical and technical reference for high-precision flexible sorting in complex industrial scenarios.
作者
李康康
王智慧
朱梓豪
王成军
尹豪宇
郑艳
LI Kangkang;WANG Zhihui;ZHU Zihao;WANG Chengjun;YIN Haoyu;ZHENG Yan(School of Artificial Intelligence,Anhui University of Science&Technology,Hefei 231131,China;School of Mechanical and Electrical Engineering,Huainan Normal University,Huainan 232038,China)
出处
《机械强度》
北大核心
2026年第3期68-76,共9页
Journal of Mechanical Strength
基金
安徽省自然科学基金项目(K2230074)
国家级大学生创新创业训练项目(202410361096,S202410361196)
安徽省教育厅重点研究项目(2024AH051746)。
关键词
多维振动
拨动机构
自适应权重
调姿
分拣
Multi-dimensional vibration
Toggle mechanism
Adaptive weight
Pose adjustment
Sorting
