点焊过程恒流控制与熔核尺寸预测系统研究

【作者】 王茜；

【导师】 白钢；

【作者基本信息】 西北工业大学 ， 材料加工工程， 2004， 硕士

【摘要】 电阻点焊过程与质量控制一直是电阻焊领域重要的研究课题之一，而采用智能控制技术已成为其发展的必然趋势。本文基于模糊控制与人工神经网络智能技术，提出了集点焊恒流模糊控制与点焊熔核尺寸的人工神经网络预测功能于一体的点焊质量控制思想。设计了点焊恒流模糊控制器，建立了熔核尺寸预测模型，初步实现了点焊恒流模糊控制。 论文首先对控制系统的总体方案进行了原理分析，对实现模糊控制的可行性、控制参量选择的合理性、采用逐点积分方法检测并计算焊接电流有效值的原理以及实现点焊熔核尺寸神经网络预测的方法进行了论证。文中着重对模糊控制器所涉及到的输入和输出变量、模糊语言变量、论域、隶属函数、模糊控制规则确定以及模糊推理、解模糊化、模糊控制表的生成等方法进行了详细分析讨论，对模糊控制器的仿真过程及结果进行了介绍。此外，对基于人工神经网络技术的点焊熔核尺寸预测功能在单片机系统上的实现途径与相关汇编程序设计，进行了系统性论述。 其次，对控制系统的中心控制器80C196KC单片机，电流有效值的检测电路以及实现点焊循环过程与控制的其它硬件电路进行了设计分析。全面介绍了采用模块化设计方法所设计的系统主程序及设定、预压、通电、中断、模糊控制、熔核尺寸预测子程序等系统软件。 文章最后对控制系统的调试与现场考核试验方法及结果进行了分析。考核试验表明：控制系统对非正弦波点焊电流的有效值检测与日本表MM-315(精度为±3％)的检测结果有很好的一致性，平均差值为59A；点焊恒流控制的规律性正确，恒流控制的效果较好。 论文所取得的阶段性成果，对实现点焊过程与质量的智能化控制，有一定的理论与工程实用价值。更多还原

【Abstract】 Resistance spot welding process and quality control are always one of important research subjects in welding field, adopting the intelligent control technology has already become its certain trend. In this thesis, on the basis of fuzzy control and artificial neural networks, the idea of spot welding quality control, which integrates constant welding current fuzzy control with artificial neural networks forecast of spot welding nugget sizes is presented. A spot welding fuzzy controller is designed, a nugget sizes forecast model is established and the spot welding constant current fuzzy controller is initially realized in this thesis.Firstly, the general scheme of control system is made principle analysis, the feasibility of realizing fuzzy control, the principle of testing and computing the valid value of current adopting point-by-point integration and the rationality of choosing control parameter and the method of realizing spot welding nugget sizes artificial neural networks forecast are demonstrated. The input and output variables, fuzzy language variables, field, membership function that fuzzy controller deals with, determination of the fuzzy control rule, the methods of fuzzy reasoning, defuzzying and generation of the fuzzy control table are emphatically analyzed and simulation process and results of fuzzy controller are introduced. In addition, the methods of establishing spot welding nugget sizes model on the basis of ANN technology, the verifying result of established models and the way of nugget sizes forecast function realized on the single micro-computer system, relative assemble language program designing are systemically discussed.Secondly, the central controller (80C196KC single micro-computer) of the control system, the measuring circuit of current valid value and other hard-ware circuits that realize spot welding circling process and control are analyzed. The soft-wares of system main program adopting modularization designing method and subprogram of setting, prepressing, electrifying, interrupting and fuzzy control are generally introduced.Finally, the debugging and testing of control system in the spot are analyzed. The result manifests that the measuring the valid value of spot welding current of non-sine wave of the control system agrees with the one of Japan current meter MM-315( its precision is?%), the average different value is 59A. the regularity of spot welding constant current control is correct, the effect of constant current control is good.The phased achievement that the thesis makes has certain theory and project practical value for realizing the intelligentizing control of spot- welding process and quality.更多还原