基于模糊关系的前混合磨料水射流切割模型建立

发布时间：2018-02-27 18:42

本文关键词： 磨料水射流表面粗糙度模糊理论 MATLAB　出处：《中国矿业大学》2015年硕士论文　论文类型：学位论文

【摘要】：磨料水射流切割技术经过30多年的发展,已日趋成熟,成为21世纪优先发展的主流切割技术。磨料水射流的切割机理开始成为各国专家和学者关注和研究的课题。本文首先对磨料水射流的切割机理和模型进行了比较系统的理论分析,详细描述了磨料粒子的受力过程和加速机理,并对切割表面余纹产生的过程及原因进行了分析研究。本文在切割理论的指导下,利用实验室的前混合磨料水射流切割设备,选择射流压力、切割速度、靶距及磨料粒径作为实验变量,设计正交实验对1060铝合金进行切割,分析研究了磨料水射流切割性能(切面粗糙度)与各参数变量之间的关系和规律。同时,在研究总结磨料水射流切割工艺特点的基础上,应用模糊推理系统建立了前混合磨料水射流切割的ANFIS模糊模型结构;本文采用高木—关野模糊系统分别建立了上部粗糙度预测模型、下部粗糙度预测模型和切割速度预测模型;利用实验研究所获得的大量样本数据,基于MATLAB编程实现自适应模糊推理模型的训练,并对模型进行了验证。通过验证模糊模型输出与实验值的对比,可以看出基于ANFIS的模糊模型结构的输出能够较好的拟合和表征实验数据的规律和趋势。
[Abstract]:After more than 30 years' development, abrasive water jet cutting technology has become more and more mature. In 21th century, the cutting mechanism of abrasive water jet has become the main cutting technology. The cutting mechanism of abrasive water jet has become a topic of concern and research by experts and scholars all over the world. Firstly, the cutting mechanism and model of abrasive water jet are compared in this paper. The theoretical analysis of the system, The mechanical process and acceleration mechanism of abrasive particles are described in detail, and the process and cause of the residual grain on the cutting surface are analyzed and studied. Under the guidance of cutting theory, this paper uses the pre-mixed abrasive water jet cutting equipment in the laboratory. Choosing jet pressure, cutting speed, target distance and abrasive particle size as experimental variables, orthogonal experiment was designed to cut 1060 aluminum alloy. The relationship between abrasive water jet cutting performance (cutting surface roughness) and various parameter variables is analyzed and studied. At the same time, the characteristics of abrasive water jet cutting technology are summarized. The ANFIS fuzzy model structure of pre-mixed abrasive water jet cutting is established by using fuzzy inference system, the upper roughness prediction model, the lower roughness prediction model and the cutting speed prediction model are established by using the high-wood and off-field fuzzy system respectively. The training of adaptive fuzzy reasoning model based on MATLAB programming is realized by using a large number of sample data obtained from experimental research, and the model is verified. The comparison between the output of fuzzy model and the experimental value is carried out. It can be seen that the output of fuzzy model structure based on ANFIS can fit and characterize the rule and trend of experimental data.
【学位授予单位】：中国矿业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TG664

【参考文献】