等离子切割多参数智能决策与控制  

Intelligent Parameter Selection and Control for Plasma-torch Cutting

作　　者：孙容磊[1] 朱李瑾[2] 万兴奖[3] 

机构地区：[1]华中科技大学机械学院,武汉430074 [2]中国地质大学图书馆,武汉430074 [3]湖北省机电研究设计院,武汉430070

出　　处：《组合机床与自动化加工技术》2005年第6期50-51,共2页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家973项目(2003CB716207);国家自然科学基金项目(50390064)

摘　　要：等离子体切割是一种经济、高效的金属切割工艺。等离子切割工艺参数中,一些参数相互耦合,共同影响切割质量与切割效率,因此,切割参数的选择是一个关键而困难的问题。论文介绍了智能数控等离子体切割机的体系结构和主要功能模块,采用神经网络实现了等离子体切割工艺参数的优化选择,采用电压测量方式实时检测割炬头与被切割板材的距离并自动调节割炬头高度,以提高切割质量和效率。系统具有自动编程、切割仿真与切割轨迹实时跟踪显示等功能。Plasma-torch cutting is a kind of economic and high-efficient machining process. Cutting quality and productivity are affected by many parameters, some of which are coupled each other. The coupled parameters increase the difficulty of selection of cutting conditions. This paper presents an intelligent methodology of parameter selection and control for plasma-torch cutting. The optimal cutting parameters are determined through a trained neural network. Distance between the cutting head and metal sheet is measured on-line and is adjusted to meet the optimal cutting parameter so as to improve the cutting quality and productivity. Additionally the system supports auto-programming, cutting simulation, and real-time display of cutting locus.

关 键 词：智能控制 数控技术 神经网络 等离子体切割 

分 类 号：TP273.5[自动化与计算机技术—检测技术与自动化装置]