The Study Of Plate-like Riser Optimization And Defects Prediction By Simulation Approach

There are many plate-like structures in different kinds of castings.Casting defects are the main factor that affects the quality of castings,and the main way to eliminate defects is to apply risers.However,the incorrect design of the riser size and its position can lead to defects generated in the casting or reduction of the productivity.Therefore,the main purpose of this thesis is to have an in-depth investigation of riser designation and defect prediction.Based on the ProCAST casting process simulation software,compared with the traditional riser module design method,the new method in this experiment is based on the algorithm of riser natural growth,combined with the rapid optimization of snake algorithm.Regard to the slab of casting 40 mm in thickness,the simulation results show that after 5 to 6 computations,the optimal solution can be obtained,and the riser volume is reduced by more than 46% and the yield of the process is increased by about 20%.The relationship between the volume of casting and riser can be expressed as VC =(2.4 ~ 2.6)VR.The casting experiment shows that the optimization design of riser can be casted without defects,the volume of which is 48% of the module method,the process yield can be increased by 22%,while shortening the solidification time of 550 s.In this thesis,the relevance between Niyama criterion and the modulus method are studied,and the Niyama criterion is obtained by using the modulus to explain the Niyama criterion.The Niyama-modulus method is also applied to predict the defects of a plate casting with its riser diameter of 105 mm.According to the results of the actual pouring test,casting defects are found in the predicted defective parts of the Niyama-modulus.Through the actual production of shovel plate at a company in Shenyang,the growth algorithm and Niyama-modulus method are employed to design the riser and predict the defect.According to the actual production of castings,and through the X-ray detection of the internal defects,the process yield is increased by about 12%,which meets the design requirements.It has been proved that the applicable and reliable of growth algorithm and Niyama-modulus method in actual production.