窄间隙GMAW焊接过程高性能控制方法与应用研究

With the high-speed development of modern manufacturing, it is urgent to research and develop high-efficiency high-performance welding technologies and equipments for thick/super thick plates, which can realize high welding quality for large welded structures in the ultra-high-speed, supercritical extreme service environment. Narrow gap gas metal arc welding (NG-GMAW) is known for saving time, power, labor, material and so on, which has the most promising. NG-GMAW welding process essentially is an extremely complicated nonlinear system with time-varying, uncertain, strong-coupling characteristics. High-quality appearance of weld is difficult to be aquired by using traditional control approaches. Therefore, novel theories and methods are needed to solve the problem. Starting with exploring the relationship between welding quality and welding parameters, new data is acquired by orthogonal experiments and analyzed by the grey relational analysis theory. The relevance vector machine method is used to model welding quality and parameters based on the data-driven theory. Then, control and filtering algorithms with high precision and fast response are proposed to realize the closed-loop control of welding quality, including the multiple input multiple output adaptive fuzzy control and cubature quadrature Kalman filter for electric feedback information. Finally, the hybrid multi-objective optimization method is proposed to establish the comprehensive performance evaluation system of NG-GMAW welding quality, efficiency and cost. This proposal is the frontier research involved in multi-disciplines of control, materials, electrical and so on. This research can not only break up the key bottleneck in the NG-GMAW welding technique and equipment, but also significantly promotes the theory and application of the relevant disciplines.

伴随现代制造业的高速发展，特别是超高速、超临界等极端服役环境下的重大工程建设与日剧增，研发厚/超厚板材高效高性能焊接技术与装备极为迫切。窄间隙熔化极气保焊（NG-GMAW）以省时/电/力/料而著称，极具应用前景。NG-GMAW焊接过程本质上是一类异常复杂的非线性、时变、不确定、多参数强耦合系统，现有方法难以实现焊缝的高质成形，亟需新的理论和方法予以突破。本项目将从探究焊接质量和工艺参数关系入手，由正交试验获得完备数据并灰关联分析；基于数据驱动理论和相关向量机方法，建立焊接质量等的预测模型；研究高精度、快响应焊接电反馈参数容积积分卡尔曼滤波算法和MIMO自适应模糊控制策略，实现焊接质量的闭环控制；最后，基于混合多目标优化方法建立焊接质量、效率和成本综合评价体系。本项目属控制、材料、电工等多学科交叉前沿研究，有望突破NG-GMAW焊接技术及装备产业化关键瓶颈，促进相关学科理论与应用的发展。

量大面广的焊接过程普遍存在能/材耗巨大、污染严重、可靠性低等问题，是制约我国高端装备制造业和经济可持续发展的重要瓶颈。本项目拟针对高效节能的窄间隙GMAW 焊接过程展开研究，这不仅是焊接技术领域的时代主题和研究热点，也是我国经济发展国防建设以及高端装备制造业崛起的重大需求。.（1）窄间隙焊接过程非线性建模与仿真分析。详细分析了窄间隙焊接机理，建立了基于电弧摆动和电弧旋转的窄间隙焊接过程动态仿真模型，从工艺参数对热输入影响的角度切入，研究了其对焊接质量的影响以及侧壁熔合不良、焊缝超高、凸度过大、咬边等缺陷产生的原因，为窄间隙焊接工艺参数优化提供了理论依据。.（2）功率变换系统及控制策略研究。分析了VIENNA整流器输入电流过零点畸变机理及电位振荡与零序分量之间的数学关系，提出一种修正注入零序分量载波调制策略，解决了电流过零点畸变问题，同时实现VIENNA整流器中点电位的平衡和纹波振荡的抑制；为解决常规模糊控制器存在稳态误差的缺点，设计了一种弧焊电压自适应模糊控制器，动态响应快、稳态误差小。.（3）焊接过程信号处理与焊接质量评价方法研究。焊接过程强烈的高频干扰和随机扰动，影响了反馈信息的真实性，为此提出一种适合于焊接电流的平滑窗口宽度可调的自适应卡尔曼滤波方法，鲁棒性强；焊接性能的综合评价是优化焊接工艺和指导焊接过程控制的重要保障，为此研究了电弧电压和焊接电流概率密度分布与焊接稳定性的关系，提出了基于U-I二值图像的弧焊稳定性判定方法，可定量分析焊接过程的稳定性。.（4）大功率电源高抗扰组网群控系统研究与实现。提出了基于三层网络结构的焊接电源群控系统方案，设计了基于B/S架构的群控管理系统；研究了基于多路Wi-Fi的动态组网方法，通过自组网算法，构建了多信道多路径的Wi-Fi网络，保障了大功率逆变电源系统的集群可靠管理。