摘要
针对柔性臂坐标测量机的热变形误差问题,为了进一步提高其测量精度,分析了自身发热导致的热误差对测量机精度的影响,布置了相应的温度传感器进行监测.利用BP神经网络建立热变形误差补偿模型,通过模拟退火算法优化权值,加快了收敛速度并解决了神经网络易陷入极小值缺点.通过实验获得样本训练所建模型,进行测量误差补偿验证,结果表明SA-BP模型补偿后误差的平均值相比BP模型减小了0.0129 mm,标准差减小了0.019 mm,验证了所提算法的有效性.
For the thermal deformation error of the flexible arm coordinate measuring machine, to further improve its measurement accuracy, the influence of thermal error caused by self-heating on the accuracy of the measuring machine was analyzed, and the corresponding temperature sensor was arranged for monitoring. The BP neural network was used to establish the thermal deformation error compensation model, whose weight is optimized by the simulated annealing algorithm to accelerate the convergence speed and solve the shortcomings of being easy to fall into the minimum value for the neural network. Through the experiment,the model built by the sample training was obtained, and the measurement error compensation verification was carried out. The results show that the average error based on the SA-BP compensation model is reduced by 0.0129 mm and the standard deviation is reduced by 0.019 mm, compared with the BP compensation model, which verifies the effectiveness of the proposed algorithm.
作者
朱嘉齐
李勇
冯旭刚
ZHU Jiaqi;LI Yong;FENG Xugang(School of Electrical and Information Engineering,Anhui University of Technology,Ma’anshan,Anhui 243032,China;Thermoelectric Plant,Ma’anshan Iron and Steel Company Limited,Ma’anshan,Anhui 243032,China)
出处
《宜宾学院学报》
2019年第12期16-19,64,共5页
Journal of Yibin University
基金
安徽省自然科学基金项目（1908085ME134）
安徽省重点研究与开发计划项目（1804a09020094）
安徽省高校自然科学研究重点项目（KJ2018A0054,KJ2018A0060）
关键词
神经网络
模拟退火算法
柔性臂坐标测量机
热变形
neural network
simulated annealing algorithm
flexible arm coordinate measuring machine
thermal deformation
