Welding process factor control
For a long time, WAAM has been positioned for low-cost, high-efficiency near-net forming of large-sized metal parts. This does not require high absolute precision in forming, but extremely high stability in forming. Unlike lasers and electron beams, WAAM has a large molten pool and is in a weakly constrained state. Its ability to suppress complex disturbances is low, and its forming stability is insufficient. The unstable factors of WAAM forming unstable, including the trajectory factors such as the starting and ending arc point, large curvature, intersection, and other factors, as well as the deposition height, interlayer temperature and heat dissipation factors. The morphology of the previous surfacing deposit layer based on TIG exhibits specific spatiotemporal and spatial discontinuous "genetic" characteristics. The surface morphology of the deposited deposit layer has a greater impact on the morphology of the next layer, so these disturbance factors will As the size and complexity of parts become more prominent, it seriously restricts the development of WAAM technology to industrial automation applications. Therefore, a prerequisite for WAAM's mature application is that it still has stable repeatability under complex disturbance conditions.
For the cross structure, Ding et al. Of Wollongong University proposed a right-angle lap forming process, which can not only avoid the uplift problem at the intersection point, but also reduce the stress concentration problem at the intersection point. For corners and large curvature structures, Geng et al. Of Northwestern Polytechnical University studied the limit geometric parameters. Both corners with too small angles and arcs with too small radius of curvature will produce obvious forming topography errors. Beijing University of Technology, Li et al. Studied the influence of mechanical platform acceleration and deceleration at large curvature structures on forming morphology, and proposed an adaptive process parameter control algorithm. On the one hand, the welding rate decreases with the increase of the corner curvature The mechanical system can be smoothly commutated. On the other hand, the wire feed rate and the welding gun rate are also reduced, so that the shape of the corner is uniform. 75% reduction.
In the final analysis, one of the key reasons for the insufficient stability of WAAM forming is that the existing WAAM technology is mostly a simple transplant from traditional arc welding to additive manufacturing, and traditional arc heat sources are usually in the "simple track, stable thermal diffusion, strong constraints "Molten pool" under the working conditions of heat and mass transport to obtain a metallurgical combination, its heat and mass transfer has a deep coupling, so it is difficult to dynamically and freely match to adapt to the "complex trajectory, dynamic heat Disturbance conditions such as diffusion and weakly constrained molten pool.
Since Cranfield University of the UK has carried out a lot of pioneering work in the field of WAAM in the 1990s, domestic and foreign scientific research institutions have carried out extensive research on WAAM's forming process, stress deformation, organizational performance, automatic control, etc., and achieved remarkable results. In recent years, with the establishment of companies such as Norwegian NTi and Australian AML, WAAM has gradually moved from the experimental research stage to the large-scale commercial application stage.
Additive forming process control
In order to realize the manufacture of complex parts, existing research usually starts with the forming process of basic structures such as single-layer single pass, single-layer multiple pass, multi-layer single pass. For single-layer and single-pass structures, statistical methods are usually used to establish mathematical models of typical process parameters such as welding rate, wire feed rate, voltage, current, etc. and forming morphology to guide the selection of process parameters. For example, Xiong et al. Of Harbin Institute of Technology established a mathematical model for the additive manufacturing process of GMAW based on neural networks and quadratic regression methods, respectively. Beijing University of Technology Li et al. Established a relational model of the cold metal transition gas shielded welding (CMT) and Tandem-GMAW additive manufacturing process based on the universal rotation combination method. For a single-layer multi-pass structure, the core issue is to determine the overlap distance between adjacent weld passes to ensure a high flatness on the upper surface. Cao et al. Of the Academy of Armored Force Engineering proposed that when the ratio of the center distance between adjacent weld beads to the width of a single weld bead is 63.66%, a single-layer multi-pass structure can achieve a higher surface smoothness.
For the multi-layer single-channel structure, the most extensive research in recent years, many thin-walled structural parts are directly composed of the multi-layer single-channel structure. The difficulty is that as the deposition height increases, the form of thermal diffusion gradually changes from heat conduction to the substrate to thermal convection and heat radiation to the surrounding air. The rate of thermal diffusion becomes slower and the problem of heat accumulation becomes prominent, which makes it difficult to control the stability of the molten pool. Extremely high. Xiong et al. Of Harbin Institute of Technology pointed out that under the condition of large welding current, although the metal molten pool can be stable at the first layer, it cannot be stable at the higher layer and flows, so the problem can only be avoided by reducing the welding current , Which greatly limits the improvement of deposition efficiency. Wu et al. Of Wollongong University studied the effect of heat accumulation on the WAAM of titanium alloy, and found that with the increase of the deposition height, due to the effect of heat accumulation, the temperature between the layers became higher, which caused serious material oxidation problems; The layer height gradually decreases until it reaches a thermal equilibrium state. In particular, the change in layer height also has a certain effect on the arc shape and metal transition behavior. In response to this problem, Wang et al. Of Southern Methodist University proposed a variable process parameter control scheme. From layer 1 to layer 40, the welding current was reduced from 140 A to 100 A, and then kept constant. It was balanced by reducing the heat input. The declining heat spreads to obtain a more uniform shape. However, the control scheme of this variable process parameter relies heavily on empirical adjustment, and the reduction of heat input will also reduce the deposition efficiency, because the heat transfer and mass transfer of the traditional arc heat source are deeply coupled. Northwestern Polytechnical University Geng et al. Raised variable process parameter control from experience to theoretical level, and established a thermal model of the WAAM process based on some assumptions. As the deposition height increases, the interlayer temperature and heat input are controlled in real time to ensure that each deposition layer has Consistent heat dissipation boundary conditions, and thus obtain a consistent shape. Beijing University of Technology Li and others have also proposed a shaping control method based on external thermoelectric auxiliary refrigeration to enhance the heat diffusion ability. The flow occurs, so that the upper and lower floors maintain a more consistent shape. This control scheme does not need to change the process parameters and does not reduce the deposition efficiency, but it requires additional external refrigeration equipment and increases the system complexity.
The above-mentioned forming process control for arc additive manufacturing when there is disturbance is essentially feed-forward control, that is, the process parameter adjustment is based on the disturbance mathematical model obtained in advance, and the response sensitivity is very high. However, the effect of feedforward control is limited by the modeling accuracy of the disturbance. Large modeling errors may cause cumulative errors perpendicular to the growth direction, which directly affects the stability of the arc and even prevents subsequent additive manufacturing from continuing. get on. To this end, some studies have introduced on-line monitoring and feedback control of the forming dimensions of each layer by introducing a sensing system, without relying on accurate modeling of disturbances.
Harong Institute of Technology Xiong et al. Have established a GMAW additive manufacturing system with dual vision sensing system. The CCD behind the welding torch monitors the width characteristics of the tail of the liquid layer of the accumulation layer. Distance, and designed a neural unit self-learning controller to adjust the welding rate in real time, so that the accuracy of the final formed parts is better than 0.5 mm. Tufts University Doumanidis and Kwak have established a dual input and output closed-loop control system, that is, two sets of structured light sensors are used to monitor the cladding layer morphology, and an infrared camera is used to detect the surface temperature of the formed part. The wire feed rate is used as the control variable, and the cladding stack height and layer width are used as the controlled variables to achieve real-time closed-loop control of the forming size. Due to the characteristics of high heat input, large heat source radius, and short flow in the molten pool, WAAM determines that its forming size has low response sensitivity to process parameter adjustment, which adds a lot of difficulty to real-time feedback control. In addition, the additional visual inspection system greatly increases the complexity and cost of the system, and there may be problems such as physical interference and visual blind areas, which limits its scope of application.