Research On Multi-Station Welding Process Planning Method Of Automobile Front Subframe For Digital Twin

As one of the core links of automobile manufacturing system,welding process planning is a key step to ensure welding quality and guide production.The planning results directly affects the man hour,logistics,cost and production capacity of the production line.With the transformation of China’s manufacturing industry to intelligence,in the whole product life cycle process,the traditional two-dimensional process modeling and planning methods,which can not effectively realize the consistency,sharing and integration of data and information,are not compatible with the full three-dimensional digital design in the upstream and the advanced manufacturing execution process in the downstream,and restrict the development of intelligent manufacturing.In view of the problems existing in the current welding process planning,such as non intuitive information expression,low information utilization and integration degree and low intelligent level,this paper takes the front subframe of a vehicle model as the research object,introduces the digital twin technology into the process planning,proposes the welding process planning system of the front subframe for digital twin.In addition,the digital twin welding process model construction,process knowledge database construction,welding unit division method and multi-station welding sequence planning method are emphatically studied.It includes the following aspects:(1)In order to solve the problems and defects existing in the current automobile front subframe welding process planning system and help enterprises realize intelligent transformation,digital twin technology is introduced into the welding process planning,and a welding process planning system of the front subframe for digital twin is established.(2)Based on MBD technology,a welding process model for digital twin is established,which includes welding process information model,welding process planning model and welding process simulation model.The model realizes the reliable storage and effective transmission of data in the stage of product design and process planning,and meets the needs of simulation.The welding process knowledge database is established to realize the integration,accumulation and reuse of process knowledge,and provide support for the selection of welding process parameters.(3)A welding element division method based on entropy weight TOPSIS method and genetic algorithm is proposed.Firstly,the quantity range of basic parts is determined according to the optimal unit division formula;Secondly,taking the quality,volume,surface area,connection quantity,weld quantity and weld volume of parts as evaluation indexes,the basic parts are determined by entropy weight TOPSIS method;Then,taking the element stability,element balance and element complexity as the optimization objectives,the best element division scheme with different number of elements is obtained by genetic algorithm,and the best element division scheme is determined through comparative analysis to realize the scientific and reasonable division of welding elements.(4)Considering the welding residual stress and deformation caused by welding thermal cycle,the welding finite element simulation method based on inherent strain method is used to analyze the influence of welding sequence of automobile front subframe on welding residual stress and deformation.The results show that welding sequence is the factor affecting welding residual stress and welding deformation.Based on the secondary development technology of MSC Marc’s,the finite element simulation analysis module of welding residual stress and welding deformation is developed by using Python to realize the setting of welding process and the reading of post-processing file information.(5)Aiming at the integrated optimization of welding sequence planning and the balance of two kinds of welding production lines,a multi-station welding sequence planning method is proposed.Firstly,considering the factors such as welding sequence,station task balance and welding quality,the fitness functions of the two kinds of optimization problems are established respectively with the evaluation indexes of welding time,station number,production beat,line smoothing coefficient,element aggregation,welding residual stress and deformation;Then,a certain number of welding sequence schemes are obtained by using the Improved Elephant Herding Optimization Based Hyper-Heuristic Algorithm(IEHO-HH);Finally,the welding residual stress and welding deformation corresponding to these schemes are obtained by finite element simulation,the final fitness value is calculated,and the optimal welding sequence planning scheme is selected.By comparing the process planning schemes of automobile front subframe,it is found that the scheme solved by the multistation welding sequence planning method proposed in this paper has been optimized in many aspects,such as total welding time,line smoothing coefficient and so on.The clamping process and the welding process are simulated to verify the interference between parts and the interference between welding gun and parts.Using plant simulation to carry out welding production line simulation,the results show that utilization rate of each workstation has exceeded 80%,and the production capacity of the production line has been improved,which further verify the rationality and feasibility of the method.(6)The welding process planning systemfor digital twin is developed based on the secondary development technology of Teamcenter and NX,which realizes the auxiliary selection of welding process parameters,the intelligent division of welding units and the intelligent planning of welding sequence.