Study On Stream Of Variation Modeling And Robust Control Method For Complex Thin-walled Parts In The Investment Casting Process

Investment casting(IC),as a very important integral forming method,is widely used in producing key parts of aviation,national defense,shipbuilding and automobile industry,especially in the core component of aero engine such as casings and turbine blades.With the continuous development of these industries,the deformation and dimension accuracy control of those complex thin-walled investment casting parts has been placed higher demands.Thus,studies on dimensional variation control and robust control method play a crucial role in improving the precision of the investment casting parts and raising the level of the industry in China.The IC process is a complex process with many procedures and process parameters.Its major stages consist of the production of wax parts,the production of the ceramic shells and the metal casting.In these major stages,many smaller processes are involved,each of which has its own complexities and poses a potential and uncertain threat on the accuracy of final part dimensions.However,current studies just focus on the dimensional deformation of individual stages;it is difficult to predict the dimensional variations of the final casting part according to the dimensions of die.Therefore,not only individual stages but also the whole IC process should be taken account.Stream of Variation(SoV),a mathematical analysis method coming from the traditional machining field,takes the generation,propagation and accumulation of the dimensional variations of the manufacturing in different stages into account and describes them concisely so that the engineers are able to propose quality control methods to improve the dimensional accuracy of final casting parts.Considering of the characteristics of IC process,SoV is one of appropriate applications.SoV is employed in this dissertation.The multi-source and heterogeneous variations are analyzed on the basis of the mechanism of variation generation and test experiments in every stage so that the main variation sources are confirmed and the quantitative relationship between the variation sources and the dimensional variations are represented.Then the propagation and accumulation of the dimensional variation in IC process are interpreted and a SoV model is built up.Besides,a robust control method based on the SoV model is proposed according to the multiple input and multiple output characteristics of IC process.Finally,the SoV model and the robust control method are applied in the production of a complex thin-walled part,and the dimensional accuracy of final casting parts are improved from CT6 to CT3 tolerance grade.The main contents are as follows:(1)Research on the relationship between the multi-source and heterogeneous variations and dimensional variationsIn this dissertation,the mechanism of variation generation is illuminated,and the multi-source and heterogeneous variations are analyzed on account of test experiments in every stage.The main random factors in the multi-source and heterogeneous variations are classified,and the main sources of variation in the whole IC process are defined,including the variations of holding pressure,holding time of pressure,storage temperature,storage time,the thickness of shell,dewaxing time,roasting temperature,holding time of temperature,casting temperature and so on.Finally,the mathematical representation between the multi-source and heterogeneous variations and dimensional variations are derived according to the definitions of the variation source vector and the dimensional change rate as well as the quantitative relationship between the dimensional change rate and process parameter variations,which lays a foundation for the SoV modeling.(2)SoV Modeling for the IC processIn view of the propagation and accumulation of the dimensional variation in the IC process,the mathematical conditions of SoV modeling in the whole IC process are established based on the definitions of state vector and dimensional change rate center.Then,a SoV model is put forward on the basis of the mathematical derivation of the variation propagation and accumulation in the whole IC process,and the quantitative calculation of variations in the IC process is realized.Finally,the proposed method is used to predict the dimensional accuracy of the complex tin-walled parts in the IC process.The result is that the variation value and mean value calculated through SoV model is smaller than that of the actual casting parts with slight differences 0.2237 mm and 0.0296 mm,respectively,which proves the validity of the model.(3)Robust control based on the SoV model in the IC processIn view of the characteristics of long process flow,various process parameters and interference between each response in the IC process,the relationship between the dimensional variations and the slight fluctuations of process parameters is studied and a robust control method based on SoV model of whole IC process is proposed in this dissertation.According to this method,the stages and process parameters with high contribution to the dimensional variation of final casting are controlled,and the robustness of each dimension is weighed by the robust optimization of range for multiple responses based on desirability function.Then,the process parameters in the wax injection process and wax storage process are calculated through this method,and the variations for the maximum outer ring diameter and wall thickness was reduced 0.6mm and 0.14 mm,respectively,which realizes the robust optimization of multi responses in the IC process under the condition of a smaller optimization scale.(4)The application of variation stream model and the summary and analysis of the deformation law in the IC processUnder the ideal condition of a controllable direction of process parameter variations,the mean dimensional value and the dimensional variation are adjusted according to the SoV model.Besides,combined with computer aided technology and SoV model,a novel method is presented for the design and modification of die.A new die with complex thin-walled characteristics are designed and modified,by which some new casting parts are produced.As a result,the maximum difference between the dimensional mean values and the nominal values is less than 0.55 mm,which proved the validity of the method.Obviously,it is very significant for the design of die in the engineering production process.The end of this dissertation,on the basis of the experiments of investment casting of complex thin-walled parts,summarizes and analyzes the rule of dimension change in the whole IC process,the influences of thin-walled and variable cross section features as well as the interactions between different objects at different stages on the dimension changes.In consideration of the main stages of the producing for complex thin-walled parts in the IC process,including the construction of wax parts,the construction of shells and the casting process,this dissertation makes exploratory researches on the propagation and accumulation of the dimensional variation in the IC process on account of the deformation mechanism analysis and some test experiments in every stage.Then,the SoV model is proposed,by which the quantitative calculation of variations in the IC process is realized.Finally,a novel method for the design and modification of die and a robust control method are raised on the basis of the SoV model.All in all,this dissertation provides a new idea and new method for the precision control of complex thin-walled parts in the IC process.