A Generalized Rough Set Modeling Method for Welding Process

Modeling is essential, significant and difficult for the quality and shaping control of arc welding process. A generalized rough set based modeling method was brought forward and a dynamic predictive model for pulsed gas tungsten arc welding (GTAW) was obtained by this modeling method. The results show that this modeling method can well acquire knowledge in welding and satisfy the real life application. In addition, the results of comparison between classic rough set model and back-propagation neural network model respectively are also satisfying.     

粗糙集与神经网络在发动机故障诊断中的融合应用

电喷发动机运行的状态信息众多而复杂,故障与状态信息的关系模糊而不确定,如何从复杂的多元信息中获取有用部分并加以利用是电喷发动机故障诊断的关键.本文应用粗糙集理论对冗余信息进行约简,得到更为简明的诊断规则,将约简结果与神经网络相结合,建立了故障诊断系统.网络的训练对比结果表明,粗糙集理论的约简处理简化了神经网络结构,提高了网络的训练效率;通过实例验证了粗糙集理论与神经网络相结合进行电喷发动机故障诊断的可行性.     

粗糙集与神经网络在发动机故障诊断中的融合应用

电喷发动机运行的状态信息众多而复杂,故障与状态信息的关系模糊而不确定,如何从复杂的多元信息中获取有用部分并加以利用是电喷发动机故障诊断的关键.本文应用粗糙集理论对冗余信息进行约简,得到更为简明的诊断规则,将约简结果与神经网络相结合,建立了故障诊断系统.网络的训练对比结果表明,粗糙集理论的约简处理简化了神经网络结构,提高了网络的训练效率;通过实例验证了粗糙集理论与神经网络相结合进行电喷发动机故障诊断的可行性.     

点焊镀锌钢板时球形电极表面熔敷TiC涂层对电极失效的分析

利用SEM、EDX和XRD等方法分析了在点焊镀锌钢板时球形电极表面熔敷TiC涂层对电极失效的影响。结果表明:点焊镀锌钢板时,球形电极的失效机制,主要是电极和镀锌板之间局部焊接的断裂发生在电极表面而导致的电极磨损,以及电极和镀锌板表面的锌之间的合金化。表面涂敷TiC的CuCrZr电极的寿命(1200点)是CuCrZr电极寿命(500点)的2.4倍,表面处理能提高电极寿命的主要原因是,在点焊镀锌钢板时表面涂敷的TiC层能阻碍电极和镀锌板之间的局部焊接和阻碍电极和镀锌板表面的锌之间的合金化。     

Numerical Simulation of Projection Welding Processes for Door Hinge of Automobile Based on Coupled Fields Analysis

Projection welding is a variation of electric resistance welding with the dynamic changes of the flow paths for heat and electrical properties with changing temperature caused by the large plastic deformation collapse of projection. As the joint type between the auto door hinge and the inner plate, projection welding may bring welding distortions and would affect the assembly quality of auto body. A comprehensive electric-thermal-mechanical numerical simulation was performed to quantitatively simulate the processes of projection welding by using a coupled finite element method. The mechanism of projection collapse and the formation process of nugget were discussed and good conclusions have been achieved comparing with the test results.     

小净距隧道开挖围岩压力的混沌神经网络模型预测

隧道围岩压力是分析围岩稳定性的关键因素。由于围岩变形受众多因素影响,且各因素之间存在强烈的非线形关系,因此,难以进行有效的预测。提出基于混沌神经网络模型的方法,分析混沌与神经网络相结合预测手段的可行性,对围岩随时间变化的压力进行仿真计算。对该理论的建立及预测方法进行系统分析,为该领域的研究提供有效的技术方法。结果表明,此方法预测精度高,能够满足工程及控制要求。     

车联网中驾驶员反应时间实时估计方法

结合车辆非线性分段制动特性和车辆状态数据对驾驶员反应时间进行实时估 计,采用特征数据拟合出初速度与减速度上升时间关系模型,通过BP神经网络推算减速 度上升过程中减速度随时间非线性变化关键参数,并在此基础上基于车辆动力学运动规 律求解车辆分段制动位移,通过实际位移与理论位移等价关系实现驾驶员反应时间实时 估计,最后利用Newell 跟驰模型场景对本文提出方法进行仿真验证.结果表明：不同初速 度条件下,反应时间估计值平均误差不超过0.022 s;不同反应时间条件下,反应时间估计 值平均误差不超过0.013 s.本文提出的反应时间估计方法不用增加通信负担和额外设备, 能够达到较高精度,具备更优的适应性、可用性及工程价值.     

基于FNN的环境库兹尼茨曲线预测分析

运用模糊神经网络(FNN)理论，通过建立环境和经济发展协调度预测模型，对库兹尼茨曲线进行了定量研究，从而为区域经济决策提供依据，以成都市为例，对其环境与经济发展的协调度进行了预测．根据预测结果对成都市经济发展与环境的关系进行的分析表明，其环境库兹尼茨曲线的峰值特征不明显，表明它的环境质量和经济发展的关系趋于协调。     

基于神经网络的船体焊接规范自动预测方法研究

针对目前船体焊接规范参数设计的特点,设计了船体焊接规范自动决策的人工神经网络模型．通过实际焊接工艺数据对网络模型进行训练,把以规则显示表达的工艺设计知识转化为隐式网络权值,应用网络推理机快速、准确地预测船体焊接规范参数．实验表明所建神经网络模型具有较好的自学习和预测能力,为船体焊接工艺参数优选提供了一条新途径．     

Relationship between quality and attributes of elliptical spot welds of high strength steel

More high strength steel(HSS) sheets are being applied in the automobile industry to improve safety and fuel efficiency.When HSS is spot welded, the strength of the welded-joint is usually determined by the attributes of the weld nugget.However, it is diffcult to obtain the circular nugget and to establish the relationship between the weld attributes and welding quality because of a large number of variables and experimental uncertainty.In the paper, the numerical analysis model of tensile-shear strength test for HSS spot welds was established.And the critical ellipticity of the elliptical spot weld was determined through finite element analysis.Quantitative relationship between elliptical weld attributes and weld strength was investigated statistically using the Latin hypercube and regression analysis method.The sheet thickness, nugget ellipticity, hardness radio and yield strength have the most influential attributes in affecting welding quality.The research results can be used as a guideline for HSS resistance spot welding design.     

点焊电极点焊镀锌钢板的失效分析

本文对Ｃｒ—Ｃｒ—Ｃｕ合金材料制造的电极在点焊镀锌钢板时的失效形式进行了分析．结果表明：Ｃｒ—Ｚｒ－Ｃｕ合金材料制造的电极在点焊镀锌钢板时的失效形式主要是电极头部的塑性变形、电极端面的合金化以及再结晶和粘附。点焊镀锌钢板时电极失效的主要原因是Ｃｒ—Ｚｒ—Ｃｕ名金材料的软化温度低，点焊温度下的硬度、强度低和端面的合金化。     

基于Marc和遗传算法的焊接工艺参数优化

焊接变形的大小主要受焊接速度、焊接电流、电压、送丝速度、焊接顺序等工艺参数的影响.根据焊接热加工特点和实际需要,针对设计变量数目繁多、类型多样的焊接工艺,采用遗传算法与有限元分析相结合的方式,对重点参数进行了优化.以二维平板对接焊的模型为例,以工件最小焊接变形量为优化目标,经过遗传算法进化后,结合有限元方法进行每组参数组合的焊接过程热力耦合分析,找出焊接能量和焊接速度的最优组合,为焊接工艺的优化和制定提供科学的评价和预测.     

Comparison of Three Control Methods in Penetration Control of Pulsed Gas Tungsten Arc Welding

An artificial neural network model for backside bead width was established and three control meth-ods——PID. fuzzy and neuron were designed, simulated and tested. The test results of bead-on-plate weld of GTAW indicate that the artificial neural network (ANN) modeling and learning control method have more advan-tages than the conventional method. They show that the ANN modeling and learning control method is an effective approach to real time control of welding dynamics and ideal quality.     

一种基于神经网络的馈线自适应继电保护技术

针对常规保护的自适应能力及检测某些复杂故障方面存在一定局限性,利用前馈神经网络,导出了常规保护与神经网络参数间的显式关系,证明了神经网络继电保护可实现常规馈线保护的功能。     

铁路货车用高强度耐候钢Q450NQR1焊接工艺

通过选择焊接材料及焊接工艺参数,对新型耐候钢Q450NQR1进行了焊接工艺、焊接接头的显微组织和力学性能等研究．试验结果表明：用的S-8018焊条及BI-1550-NQ-A焊丝焊接Q450NQR1耐候钢时接头各项力学性能满足规定的技术要求,焊缝金属的显微组织基本相同,均以针状铁素体为主＋少量的先共析铁素体,过热区组织均为粒状贝氏体,焊接热影响区正火及不完全正火区的组织均为铁素体＋珠光体;焊接线能量变化对焊接接头焊缝及热影响区显微组织有一定的影响,但影响不大．     

高速铁路模型箱梁静力非线性和动力损伤关系研究

桥梁结构在荷载作用下的静力非线性与结构动力特性之间的关系是结构损伤识别领域的重要问题。基于高速铁路模型箱梁的重复荷载试验,采用平面非线性有限元分析方法对静力非线性和动力损伤之间的关系进行了研究。通过定义混凝土等效应力－等效塑性应变曲线来定义其弹塑性行为。对数值模型加载到各级荷载后卸载进行动力特性分析,即卸载后计算自振频率和振型。计算结果表明：采用平面非线性有限元分析方法可以比较准确地对箱梁模型进行非线性分析,除了初始模型刚度存在一定误差外,结构的骨架曲线的特征值、加卸载刚度和残余位移吻合较好。在进行非线性分析后进行动力特性是分析可行。竖向频率值的变化能够反映静力非线性发生后结构的损伤出现和损伤的发展规律,这为结构动力损伤识别提供了有效的途径。     

High Efficiency and Quality of Multi-Pass Tandem Gas Metal Arc Welding for Thick Al 5083 Alloy Plates

Multi-pass tandem gas metal arc welding(TGMAW) for Al 5083 alloy plates with 30 mm thickness is carried out to study its high efficiency and high quality. The welding process is evaluated by high-speed photographs and electrical signals. The deposition rate and welding time are calculated to characterize the welding efficiency. The bead formation, porosity, microstructure and mechanical properties are tested and analyzed. The results indicate that though arc deflection occurs due to the electromagnetic interference between two arcs during TGMAW process, it does not affect the welding process stability significantly. The deposition rate and welding time of TGMAW process are about 15 g/min larger and about 16.7% less than those of gas metal arc welding(GMAW)process respectively, accompanied by the reduction of heat input. The TGMAW welded joint has less porosity and more refined microstructure to contribute to the improvement of mechanical properties(micro-hardness, tensile strength and ductility). It can be concluded that TGMAW process yields excellent performance of welded joint with improved welding efficiency, which makes it extremely practical during welding of thick plates.     

神经网络在车辆半主动控制中的应用

针对车辆半主动悬挂模型的非线性特性，研究了采用神经网络的自适应控制在车辆悬挂半主动控制中的应用，设计了采用前馈神经网络的辩识器和控制器，对控制器网络采用的学习规则进行了分析。仿真计算表明，采用神经网络自适应控制方法能有效改善车辆的平稳性，采用的神经网络辩识器和控制器的改进BP算法是有效的。     

钢轨加压铝热焊的研究

为了提高钢轨常规铝热焊的接头质量,作者对钢轨加压铝热焊工艺进行了试验研究.结果表明,这种工艺使铸态焊缝变窄,母材产生一定塑性变形,接头韧性大幅度提高,是既能保证接头质量又便于现场生产使用的一种工艺.     

双丝焊送丝机嵌入式通信系统前期研究

双丝MAG(熔化极活性气体保护焊)焊接的应用越来越广泛.在造船行业中,由于船台焊接作业特点决定了电焊机设备普遍使用几十米的加长控制电缆.利用嵌入式系统,研究焊接电源和送丝机之间控制信号通信系统,实现减少控制电缆的芯数,既可减少生产成本,又可减少断线故障率,提高生产质量和效率、节能降耗,具有重要的现实意义,且对于多丝焊接...