磨料水射流铣削加工表面粗糙度模型研究  

Study on model of surface roughness achieved by abrasive water jet milling

作　　者：万庆丰[1] 雷玉勇[1] 刘克福[1] 陈忠敏[1] 潘峥正[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》2013年第9期126-130,共5页Mining & Processing Equipment

基　　金：四川省科技厅项目资助(2011JYZ017);西华大学研究生创新基金资助(ycjj201340)

摘　　要：通过磨料水射流铣削加工试验,研究磨料水射流工艺参数对加工表面粗糙度的影响。采用方差分析和电子扫描显微镜,获得影响表面粗糙度的关键参数以及加工表面的微观结构。结果显示,水射流压力是影响表面粗糙度的最大极限参数,随着射流压力的增加,材料表面的条纹和波痕显著增加,表面粗糙度恶化。通过人工神经网络方法,建立磨料水射流铣削加工表面粗糙度的BP网络模型,利用试验数据对网络模型进行训练和验证。研究结果表明,预测模型的最大相对误差为3.56%,验证了BP网络用于表面粗糙度预测的可行性。Based on the abrasive water jet milling test, the influences of process parameters on machined surface roughness was studied. The significant parameters affecting the surface roughness and the microstructure of machined surfaces were obtained by using analysis of variance and electron scanning microscope. The investigations revealed that the water jet pressure was an utmost parameter affecting surface roughness as well as surface striation and waviness increased significantly with jet pressure raise, which led to the deterioration of the surface roughness. Based on the method of artificial neural network, a BP neural network model of surface roughness achieved by abrasive water jet milling was established, and test data was applied to train and verify the model. The results indicated that the largest relative error of prediction model was3.56%, which verified the feasibility of applying BP network to prediction of surface toughness.

关 键 词：磨料水射流 表面粗糙度 人工神经网络 方差分析 

分 类 号：TP69[自动化与计算机技术—控制理论与控制工程] TG84[自动化与计算机技术—控制科学与工程]