Research On Crankshaft Assembly Of Aero Piston Engine

With the improvement of assembly accuracy requirements of mechanical products and the development of parts assembly to mass production,the traditional assembly process can no longer meet the requirements of modern assembly which are high-precision and large-scale.Therefore,the establishment of an assembly optimization model and the use of optimization algorithms to solve the optimal assembly combination have become the mainstream of contemporary assembly technology.In this paper,a corresponding research on the crankshaft assembly related problems of an aero piston engine is carried out.The main research contents and results are as follows:In the assembly process of the piston engine crankshaft,the repairing method is used to adjust the axial clearance of the crankshaft.The repaired parts need to be repeatedly polished,disassembled and reassembled.The assembly qualification rate is low,the labor intensity is high,and the assembly efficiency is low.Aiming at the shortcomings of traditional repairing methods,a computer-aided assembly based on genetic algorithm is designed: real number coding is used to establish an optimization model with the number of repair parts and the range of size fluctuations as the goal.The results show that under the premise of satisfying the complete interchange within the group and ensuring that the axial clearance of the crankshaft is within the qualified range,by grouping the repair parts,the mass production and processing of repair parts can be realized,the labor intensity is reduced,and the assembly efficiency is improved.Secondly,in the process of component selection and assembly,the assembly quality included not only the assembly qualification rate,but also the assembly precision and assembly cost.In this paper,Taguchi quality loss was used to measure assembly accuracy,and a multi-objective optimization model was established based on assembly accuracy and qualified number.The classical multi-objective evolutionary algorithm NSGA-Ⅱ(Non-dominated Sorted Genetic Algorithm-Ⅱ)was used to solve the multi-objective optimization problem,and a group of better non dominated solutions were obtained.In view of the large number of duplicate individuals in the population and the deficiency of NSGA-Ⅱ algorithm in maintaining population diversity,we proposed an improved hierarchical strategy and three elite retention strategies based on the distribution function and the maximum uniformity of the population to eliminate some duplicate individuals.By introducing adaptive strategies of crossover,mutation and population renewal,we could avoid inbreeding and add new individuals in a certain proportion,which further improved the diversity of population.In terms of convergence,the nearest neighbor search was used to overcome the local search ability of NSGA-Ⅱalgorithm,and improved the evolution speed and convergence accuracy of the algorithm.Moreover,CINSGA-Ⅱ algorithm(comprehensive improvement of non-dominated sorted genetic algorithm-Ⅱ)was proposed and compared with the standard NSGA-Ⅱalgorithm.The results showed that the diversity and convergence of the optimization results obtained by CINSGA-Ⅱ algorithm were greatly improved.Finally,the performance differences of NSGA-Ⅱ and CINSGA-Ⅱ in convergence and distribution were compared by using ZDT series test functions with generation distance,spacing,anti-generation distance and distribution as evaluation indexes.The results showed that compared with NSGA-Ⅱ algorithm,CINSGA-Ⅱalgorithm has obvious advantages in convergence and distribution.