Research On Expert System For Laser Welding Pool Morphology Based On Approximate Energy Attenuation Keyhole Model

Welding simulation is an important component of modern intelligent laser welding manufacturing and can be used to predict the evolution of the physical dynamics of the laser welding process and to assist in the development and optimisation of the laser welding process.When modelling laser welding simulations,the heat source model is fundamental and directly determines the validity of the simulation results.Existing heat source modelling studies include consideration of flow field models and geometry distribution models.Flow field models involve fluid flow,mass migration,phase change and other mechanisms within the melt pool,making such models computationally intensive and complex,and the technology is focused on only a few institutions;geometry models are simple and popular,but lack guidance on the mechanisms of the laser physical process;in response to the above problems,this study starts by considering the keyhole effect,establishing a link between the internal physical mechanisms of the melt pool and the welding process parameters.The establishment of a mathematical model based on approximate energy decay keyhole and the design of the corresponding expert module,aimed at building a heat source model based on the physical mechanism of the laser welding heating principle,building process parameters and heat source model correlation.To achieve these objectives the following research elements are carried out:Based on the formation of the keyhole energy absorption mechanism and the principle of energy conservation within the melt pool,the establishment of an approximate energy decay of the keyhole mathematical model,in order to describe the energy decay characteristics and the shielding effect of the plasma cloud on the laser beam,the concept of the decay function is proposed,and the use of a single factor control variable method using the keyhole mathematical model for simulation calculations to study the parameters in the decay function and the law between the change in the shape of the melt pool.On this basis,the design of 304 stainless steel laser welding test,using a single-factor controlled variable method were set up laser power,welding speed,the amount of defocusing three process parameters,and to carry out controlled simulations,compared with the melt pool cross-sectional shape to verify the accuracy of the model,study the influence of process parameters on the shape of the melt pool law.Based on the expert system design rules,the framework and process of the laser welding pool morphology expert system is designed,and the model is further improved by using intelligent calibration with machine learning and empirical reasoning,and the user interface of the expert system is designed.Using the aforementioned keyhole model and parameter adjustment rules for engineering applications,for the aluminum alloy laser welding thermal cracking phenomenon for temperature field simulation,analysis of the lap structure and the impact of welding process parameters on the temperature field.The reliability of the model was verified using an expert system for engineering applications.