Optimization Of Locating Point Of Flexible Thin Plate Fixture Based On Improved Gravitational Search Algorithm

In the development of the modern automobile industry,the body engineering gradually takes the lead in the design field,and the body determines the safety,appearance,stability and wind noise.In order to achieve the fluency of the design,the body adopts a large number of flexible thin plate parts for assembly,but the rigidity of the flexible thin plate parts is poor,and a reasonable fixture positioning scheme for the thin plate parts needs to be designed.The traditional design method is dependent on the experience of skilled workers.To carry out the positioning,its positioning design level depends on the individual subjective,and the design process is not standardized.With the increasing requirements on the assembly accuracy,the traditional positioning design can no longer meet the manufacturing requirements.The main work of this paper is as follows.1.Analyze the characteristics,requirements and errors of the thin plate.Study the positioning principle of "N-2-1" and determine the position combination of the main plane positioning point as an optimization variable.The analysis of the error sources of the thin plate shows that the flexible deformation and fixture errors of the thin plate affect the final assembly quality of the product.Therefore,the objective function of optimizing the deformation of the key points of the thin plate is established for the flexible deformation of the thin plate.For the fixture positioning error,based on the influence coefficient method and the Monte Carlo method,the objective function with the key measurement point deviation as the optimization object is constructed.2.The gravitational search algorithm and the particle swarm algorithm are analyzed,and the improved gravitational search algorithm optimization(IGSA)is proposed.Two optimization measures are proposed for the problems existing in the gravitational search algorithm's practical application: the global memory capacity of the introduced particle swarm algorithm.The elite particles were retained.Five uni-peak and five multi-peak test functions in the classic Benchmark were used to perform numerical experiments on the improved gravitational search algorithm,the standard gravitation search algorithm,and the particle swarm algorithm.The average and maximum values of the experimental results were selected.The value and the minimum value,the test results compare the search accuracy and convergence rate of different algorithms for different test functions,and verify the optimization performance of the improved gravitation search algorithm.3 Using MATLAB to write optimization algorithms.The cyclical finite element analysis software ABAQUS is used to achieve the automatic optimization of fixture positioning points.Taking the key measurement point deformation and deviation as the objective function,a comparative analysis of multiple sets of simulation tests is performed on a simple plane thin plate.The maximum,minimum,and average values in the optimization results and convergence algebra are selected and applied in the deviation analysis using ABAQUS generation.Sensitivity matrix,and the mean,maximum value and variance of the deviation values of each point in Monte Carlo simulation were applied in MATLAB to test the optimization performance of the improved gravity search algorithm for fixture positioning points.Finally,the body side surrounding sheet is taken as an example to test the universality of the improved gravity search algorithm.