基于改进遗传算法的机器人去毛刺工艺参数优化

Optimization of Robot Deburring Process Parameters Based on Lmproved Genetic Algorithm

传统人工去毛刺方法存在工作环境恶劣、劳动强度大、质量一致性差等问题,以机器人为代表的去毛刺自动化装备得到了广泛关注。由于去毛刺工艺过程复杂,影响因素众多,难以通过理论分析推导工艺参数,根据经验制定相关参数值。针对机器人去毛刺工艺参数优化数学模型的求解,提出了一种集成模拟退火的改进遗传算法,种群选择采用Metropolis抽样,有效解决了遗传算法易局部收敛的问题,搭建了机器人去毛刺测试平台,开展了工艺参数优化验证。测试数据表明与经验方法相比,单位长度毛刺去除耗时缩短,效率明显提升。

Traditional manual deburring methods have some problems,such as bad working environment,high labor intensity,poor quality consistency,and so on. Robot as the representative of deburring automation equipment has been widely concerned. Because the deburring process is complex and there are many influencing factors,it is difficult to deduce the technological parameters through theoretical analysis and formulate the relevant parameters based on experience. An improved genetic algorithm based on synthetic simulated annealing is proposed to solve the mathematical model of process parameters optimization for robot deburring. By means of adaptive genetic operator strategy,the problem that genetic algorithm is easy to converge locally is effectively solved. A deburring test platform for robot is built and the process parameters optimization verification is carried out. The test data show that compared with the empirical method,the unit length burr removal time is shortened and the efficiency is significantly improved.