Study On Semi-solid Slurry Preparation Device And Technology Of High-strength Aluminum Alloy Based On Enthalpy Compensation Method

Semi-solid forming technology is a near net forming technology that has developed rapidly in the decades.It combines the advantages of solid forming and liquid forming to obtain high product performance at low production costs.The semi solid formed products are widely applied in automotive,aerospace and communication industries.The currently existed technologies of making semi-solid slurries mainly have limitations such as only applicable for a few alloy materials,non-uniform distribution of temperature in a slurry and slow process of optimizing parameters.This project is to develop a novel technology to make h igh quality semi-solid slurry based on so the called Enthalpy Compensation Method.A series of studies on 7075 high-strength aluminum alloy materials have been carried out using the developed technology.The detailed results are presented as follows:The finite element analysis software was used to simulate the influence of the key process parameters on the slurry temperature distribution,thus assisting the design of experimental equipment and experimental procedure.The simulation results show that the thinner the bottom of the crucible the more u niform the slurry temperature distribution.Appropriately increasing the distance between the bottom of the crucible and the bottom of the induction coil can make the slurry temperature distribution more uniform.As the heating time increases while other parameters keep constant,the maximum temperature difference within the slurry initially decreases and then increases.It is important to detect the inflection point of the maximum temperature difference changes with time and to determine the appropriate heating time at different process parameters for uniform slurry temperature distribution.According to the characteristics of Enthalpy Compensation Method and the analysis of numerical simulation results,an automatic device of making semi-solid slurries has been designed.The process parameters of initial crucible temperature,melt weight,heating power and heating time were selected for experimental studies.The influence of single parameter on the radial,axial and overall temperature distribution within the semi-solid slurry was investigated.The experimental results show that the Enthalpy Compensation Method can be successfully applied to effectively obtain a uniformly distributed temperature within semi-solid slurry,to reduce the maximum temperature difference and to make high quality semi-solid slurries of 7075 aluminum alloy with solid fraction of 40% to 50%.Initial crucible initial temperature,melt weight,heating power and heating time all have great influence on the average slurry temperature,while heating time is the most important parameter for the slurry maximum temperature difference.With increase of the heating time,the average temperature of the slurry initially decreases and then increases.With increase of the heating power,the initial crucible temperature and the melt mass,the average temperature of the slurry gradually increases.Furthermore,with increase of the heating time,the overall temperature difference of the slurry initially decreases and then increases slowly.The higher the heating power,the earlier the inflection point of the slurry temperature difference change with the heating time is reached.A medium heating power(5-10 k W)can be used to obtain desirable average temperature and small temperature difference of the slurry.Concept of machine learning in terms of a neural network prediction model has been established in this study.Changes of the slurry temperature distribution within a semi-solid slurry made by the Enthalpy Compensation Method was accurately predicted.Two prediction models,namely the average temperature model and the maximum temperature difference model,were established.The two prediction models can give both determination coefficients of greater than 0.99.The average absolute error is 0.16 ? by the average temperature prediction model and the average absolute error is 0.52 ? by the maximum temperature difference prediction model.