Research On Process And Path Planning Of Wire Arc Additive Remanufacturing

With the development and upgrading of the domestic manufacturing industry,remanufacturing has received extensive attention due to its very important social and economic significance,and various remanufacturing technologies have been developed.Remanufacturing technology based on WAAM(Wire Arc Additive Manufacture)has great technical potential and application prospects due to its low cost,high processing efficiency and high mechanical properties.With the goal of improving the accuracy of forming,this thesis studies the following issues around the path planning of arc additive remanufacturing:A SVM(Support Vector Machine)regression prediction model for welding bead width and height is established.Firstly,the number of combinations of required process experiments is effectively reduced through orthogonal experiment design.And then,the hyperparameters of SVM regression prediction model are optimized by using CSA(Coupled Simulated Annealing)algorithm,so that SVM regression prediction model can be trained with small samples to get better prediction accuracy and get better prediction results than BP neural network.The bead size obtained through process parameter prediction provides basic parameters for path planning.A continuous path planning algorithm for complex cross-section contour areas is implemented.The continuous path can reduce the number of arc starting and extinguishing,and the coincidence of the arc extinguishing point and the starting point can also reduce the concavity of the arc extinguishing point which can improve the forming accuracy.Based on this,a continuous path planning compound filling algorithm for convex polygonal areas is first proposed,and the negative impact caused by too many corners between weld beads on forming is determined by determining the optimal scan direction.Then approximate convex decomposition is used to decompose the complex non-connected concave polygonal cross-sectional contour area to obtain convex polygon sub-regions.Compound filling is used in the sub-regions.Finally,the route between the sub-regions is optimized by an improved ant colony algorithm that can reduce the length of the idle route.In this way,continuous path planning for complex cross-section contour areas is achieved.The effectiveness and feasibility of the algorithm are verified through specific cases and experiments.Numerical simulations of different paths were performed.Through numerical simulation,the temperature field,residual stress and deformation rules of different paths in the arc additive remanufacturing process were studied,which provided a reference basis for the path selection during the remanufacturing process.