Multi-Objective Optimization And Decision Making For Selective Assembly Of Complex Mechanical Products Based On Improved NSGA-Ⅲ

The "14th Five-Year Plan" period,a new round of scientific and technological revolution and industrial technology change on the machinery products manufacturing industry,the whole process and high-quality development of higher requirements,and strive to promote the high-end manufacturing machinery products,intelligent and green development,and promote industrial optimization and upgrading.Complex mechanical products are the main production products of machinery manufacturing industry,the quality of which is mainly determined by the processing accuracy and assembly process.Processing accuracy due to technical,cost and other reasons is difficult to improve the case,the need to upgrade and optimize the assembly process,based on this,carried out a study on the selection of complex mechanical products assembly optimization and decision-making,the main research content is as follows:(1)Based on the in-depth understanding of the principle of mechanical product selective assembly,the characteristics and problems of the current complex mechanical product selective assembly are analyzed;based on the analysis of the assembly structure of complex mechanical products and the calculation method of dimensional chain,the dimensional chain constraints and feature size association constraint matrix of complex mechanical products are established,and the association relationship between feature size and feature size,feature size and quality characteristics is identified;based on the assembly success rate The mathematical evaluation model of complex mechanical product selective assembly is established based on the assembly success rate and Taguchi quality loss function;based on the in-depth understanding of multi-objective optimization theory and solution method,multiple optimization objectives and multi-objective optimization mathematical model of complex mechanical product selective assembly are established using the constructed selective assembly evaluation mathematical model,and for the problem that the optimization algorithm is easy to fall into local optimum when the number of optimization objectives is large,the NSGA-Ⅲ algorithm is selected and improved.The NSGA-Ⅲ algorithm is selected and the coding method,crossover and variation strategy of the algorithm are improved.(2)The piston-cylinder assembly case was selected,and the component feature size data were collected using simulation,and the selective assembly optimization model and the improved NSGA-Ⅲ algorithm were applied to solve the problem.The performance of the algorithms was investigated for different feature size distributions and different component batches.The results show that the proposed algorithm is less sensitive to the feature size distribution and achieves the highest quality solution set in the multi-objective optimization problem of case selective assembly for different lot sizes,which verifies the feasibility and practicality of the proposed algorithm.(3)Based on the subjective experience of the engineering decision maker and the objective information entropy of the solution set data,the VIKOR method is used to filter and rank the Pareto optimal solutions obtained by the improved NSGA-Ⅲ algorithm to decide the optimal choice of selective assembly solution,and the effect of the change in the value of the coefficients of the VIKOR decision mechanism on the decision results is further investigated.The results show that the resulting selective assembly solution takes into account both the group benefit value and the individual regret value,which verifies the feasibility and stability of the VIKOR method in the decision making process of the set of selective assembly solutions.The research confirms the feasibility and advantages of the proposed method in multi-objective optimization and decision making for complex mechanical product selective assembly,which can improve the assembly efficiency and assembly quality,reduce the assembly cost,provide a decision basis for complex mechanical product selective assembly scheme selection,and also has some significance to enhance the actual decision making ability of engineers and further accelerate the optimization and upgrading of mechanical product manufacturing industry.