Research On Residual Stress Regulation And Welding Sequence Optimization Method Of Multi-pass Welding Of Subway Frame

Bogie frame is a key bearing and load transmission component of vehicles, and it plays an important role on traffic safety. Usually, a subway bogie frame is a complex welded structure made of several parts using multi-pass welding. The welding process is inevitable due to the residual stress and deformation, has an affect on the size stability and fatigue life of the structure. Therefore, the establishment of the accurate welding residual stress and deformation numerical simulation techniques can be used to analyze the evolution and regulation methods of stress and deformation, and further study on the optimal design of welding sequence. It will promote the development of precision welding forming technology and improve the quality and reliability of service security, which will have important academic significance and application value.A welding simulation error quantitative assessment method based on weld bead contour shape was proposed to search for heat source model parameters of welding numerical simulation fast and accurately. Image enhancement processing of tested welded joint photo and simulated temperature field was conducted by pseudo color method and difference image method. A function was constructed to evaluate contour shape error between simulated and tested weld beads quantitatively. Based on the error, an inversion method was established using pattern search method for solving the heat source function parameters. Using this method, the value of the single double ellipsoid heat source function and the Gaussian+double ellipsoid combined heat source function parameters of the T-joint were optimized. Results show that combined heat source function is more suitable than a double ellipsoid heat source function for T-joint welding temperature simulation. The minimum error between simulated and tested weld bead contour shape is17.8%when simulated using the combined heat source function. The accuracy was improved by11.5%compared to the single heat source function.In order to study regulation mechanism and factors of welding residual stress distribution by the multi-pass welding, Q345C plates surfacing weld experiments were conducted with coverage were0%and50%respectively. Temperature cycling curve and residual stress were measured near the first weld toe. Based on the measurement results, three-dimensional transient thermal elastic-plastic finite element method was used to simulate and analyze the temperature, stress and strain fields during flat surfacing welding process. Study results have shown that the residual stress will decrease with the elastic strain decrease while plastic strain can remain constant. For multi-pass welding, the elastic compressive strain generated by the following weld bead offsets the initial elastic tensile strain result in the decrease of the initial residual stress. The conditions of reducing the initial residual stress near by the yield strength were studied by analyzing the relationship between welding peak temperature and the residual stress. The welding conditions to reduce the residual stress of the first weld toe to be compressive stress were studied and the weld heat input estimation formula was established.According to conditions of the first weld toe generated compressive residual stress, considering the number of weld beads, the heat input and the welding sequence of the weld beads, design principles of capping multi-pass welding to regulate residual stress was proposed. It is applied to capping welding process design of tube-sheet welded joint of the frame bracket. The welding deformation of the optimization process obtained decreases7%compared to the original process and the residual stress at all the weld toes turned into compressive stress. It has achieved the purpose of residual stress regulation.In this paper, considering the relationship and constraint of weld position and weld sequence during the welding assembly process, the welding sequence optimization problem was studied with the constraint of change times of weld positions in a complex welded structure. An advanced genetic algorithm optimization method with weld position constraint gene repair operation is proposed. A multi-parameter cascade encoding method is presented to reflect the weld position number, welded joint group number and weld seam number. Based on gene rank adjustment, the constraint optimal problem was solved. Gene repair operator can exclude the unusable welding sequence, so that a more satisfactory initial population becomes possible. The proposed method has been applied for welding sequence optimization in a bogie frame welding assembly process. In order to minimize the deformation after welding, a constrained multi-parametric optimization model was established, combined with welding distortion simulation to optimize the welding sequence. With the least change times of welding position, the optimized welding sequence of the girth weld seams of the bogie frame was obtained. The maximum vertical deformation difference using the optimized welding sequence is2.6mm. The error is18%compared with experimental results.