摘要
针对磨齿机在磨削加工时,电主轴存在热致误差等问题,提出一种基于思维进化算法(MEA)优化BP神经网络建立磨齿机电主轴热误差预测模型的方法。通过测量磨齿机电主轴在加工过程中的温升与位移情况,利用思维进化算法优化BP神经网络算法在MATLAB软件中建立预测模型,并与未经过算法优化的BP神经网络建立的模型进行了对比。在电主轴X向热误差预测实验中,未经过算法优化的BP模型最低补偿率为84.85%,而经过思维进化算法优化BP模型最低补偿率为95.29%。结果表明:经过思维进化算法优化BP神经网络建立的热误差模型,在拟合和预测精度上要优于未经过算法优化的BP神经网络热误差模型。
To improve the problem of heat induced error of motorized spindle in grinding machine,a method based on mind evolution algorithm（MEA） to optimize BP neural network to establish the thermal error compensation model of grinding machine motorized spindle was proposed. By measuring temperature and displacement of the grinding machine motorized spindle in the process of machining,the MEA was used to optimize the BP neural network algorithm to build a predictive model in MATLAB software,and the model was compared with the BP neural network which has not been optimized. In the X direction of motorized spindle thermal error prediction experiment,the minimum compensation rate of the BP model without the algorithm optimization was 84. 85%,and the minimum compensation rate of the optimized BP model was 95. 29%. The results show that the thermal error model of BP neural network which optimized by the mind evolution algorithm is superior to the BP neural network thermal error model without algorithm optimization in the fitting and forecasting accuracy.
出处
《组合机床与自动化加工技术》
北大核心
2017年第6期1-4,共4页
Modular Machine Tool & Automatic Manufacturing Technique
基金
国家自然科学基金项目(51635003)
关键词
热误差
电主轴
思维进化算法
BP神经网络
thermal error
motorized spindle
mind evolutionary algorithm
BP network model
