基于人工神经网络的微磨料水射流铣削深度控制的研究  

Research on control of milling depth of micro abrasive water jet based on artificial neural networks

作　　者：杨振[1] 雷玉勇[1] 邹虎[1] 吕华[1] 杜文正[1] 

机构地区：[1]西华大学机械工程与自动化学院,四川成都610039

出　　处：《矿山机械》2010年第22期24-27,共4页Mining & Processing Equipment

基　　金：国家自然科学基金项目(50844033);四川省教育厅项目(09ZZ030/09ZB085)

摘　　要：基于人工神经网络理论,建立微磨料水射流铣削加工的BP神经网络模型。该模型建立了微磨料水射流铣削工艺过程中铣削深度、靶距、磨料粒度、磨料流量与横移速度等参数之间的映射关系。通过控制铣削横移速度和磨料流量实现所需的铣削深度。微磨料采用步进电动机驱动螺杆主动输送,通过控制电动机的转速达到磨料精确馈送的目的。试验表明,用BP神经网络计算的铣削横移速度和磨料流量来铣削工件,可以使实际铣削深度与给定铣削深度的相对误差达0.14%～3.24%。Based on the theory of artificial neural network,a BP neural network model for water jet milling with micro abrasive was established.The mapping relationship among milling depth,target distance,abrasive diameter,abrasive flow rate as well as feeding rate of the cutting head of micro abrasive water jet milling were established.The desired milling depth could be achieved by controlling the feeding rate of cutting head and abrasive flow rate.Micro abrasive was conveyed by screw auger which is driven by step motor,and the abrasive flow rate was controlled precisely by controlling the rotary speed of the motor.The abrasive flow rate and feeding rate of cutting head calculated by BP neural network could be applied to the millng process,and the relative error between the actual milling depth and desired one is from 0.14% to 3.24%.

关 键 词：微磨料水射流 BP神经网络 铣削 

分 类 号：TG54[金属学及工艺—金属切削加工及机床]