Research On CNC Machining Technology Of Integral Impeller Based On SA-PSO Hybrid Algorithm

Aviation and aerospace industry,as the most important component of national defense industry and manufacturing industry,determines the degree of national modernization.With the continuous improvement of market and national defense demand,its production requirements continue to develop towards high precision and high efficiency.Integral impeller,as an important part of typical turbine machinery,is widely used in aviation,water conservancy and other fields,accounting for about 30% of aero-engines.Therefore,under the synchronous action of market demand and national defense demand,it is particularly important to improve the overall processing level of impeller.But because of integral impeller blade thin,interval between blade and blade is narrow,so when processing,cutting tool rotation the space is little,easy to cause the interference of cutter and blade,and a developable ruled surface due to its original rational error exist,cannot joint surface of blade,tool side blade cutter envelope surface and non developable ruled surface,easy cause less cutting or cutting error processing is difficult.In this thesis,based on NURBS curve and surface modeling,in order to reduce surface undercutting and overcutting errors as the goal.The tool path planning of non-developable straight grain surface is deeply studied and a strategy based on SA-PSO(simulated annealing algorithm and particle swarm optimization algorithm)is proposed.The process of the tool envelope surface approaching the non-developable straight surface is transformed into the particle optimization process by MATLAB and compared with the traditional algorithm,the complicated calculation process is reduced,the calculation efficiency is improved and the machining error is reduced.In the UG post-processor for the preparation of machine tool post-processing,in VERICUT for the construction of simulation machine tool,and the planned tool path trajectory for programming,generating can be used for the actual processing of NUMERICAL control program.Through simulation and experimental processing,SA-PSO hybrid algorithm and PSO algorithm were compared and compared with the single particle swarm optimization algorithm,the processing error of the algorithm proposed in this thesis was reduced by 14.7%...