Analysis Of Product Multi-parameter Coupling Change Propagation Mechanism Based On Complex Network

In the design process of complex mechanical products,changes are inevitable.However,traditional design methods may result in extensive parts redesign when changes occur.This can significantly impact product design cycles and design costs.Therefore,exploring the multiparameter coupling change propagation mechanism based on the characteristics of product parameters and their relationships is an essential problem to be solved.Properly addressing this problem is crucial to perfecting the theory of product change design and enhancing the market competitiveness of enterprises.Through a review of relevant literature,this paper provides an in-depth study and analysis of the current status of domestic and international research on parameter change propagation.Moreover,this paper focuses on the core problem of multi-parameter coupling change propagation in the complex mechanical product design process.Based on the parameter coupling change propagation network model and the basic model,this paper conducts relevant basic research in four aspects: analyzing product parameter association characteristics,finding a solution for parameter change propagation cost,analyzing parameter coupling association and decoupling,and optimizing parameter coupling change path.Finally,this paper develops a multi-parameter coupling change propagation and path preference system based on the research findings to realize the path preference of parameter coupling change.The main research contents of this article are as follows:A basic model of the propagation of changes in multiple parameters of a product was established.To begin with,this study analyzes the fundamental characteristics of product design in the state of parameter coupling change and establishes a product affiliation association model based on the product’s multi-layer hierarchy.Secondly,a complex network model of the part is constructed by integrating various part attributes,which are then extended to derive the complex network model of multi-parameter association with association characteristics.Furthermore,the study examines the coherence and logic of the product base information and analyzes subsequent key technologies.Finally,the study formulates a general analysis approach for the multi-parameter coupled change propagation mechanism of the product based on complex network theory.A solution method is proposed for parameter association importance,based on the importance of each part.Firstly,the part complex network model is used to evaluate the functional attributes and fault sensitivity of each part as the main indicators for its importance.Secondly,the functional weights of the parts are divided into two aspects: self-functional weights and network-functional weights,based on the functional association between the parts and the product as a whole.Quantification methods based on the multi-layer hierarchical analysis model and the improved HITS algorithm are proposed for both characteristics.Thirdly,the fault risk value of the part is defined accurately and scientifically based on triangular fuzzy theory,and the fault sensitivity weight of the part is quantified accordingly.Finally,an algorithm that integrates conflict coefficients and trustworthiness is proposed based on improved evidence theory.A product part trustworthiness model is established,which fuses the functional and faultsensitivity weights of the part to solve for its trustworthiness as a key part.This model is used to perform quantitative analysis of the importance of the parameter associationBuilding upon the basic model of parameter-coupled change propagation and the parameter complex network with community characteristics,the multi-parameter coupled change propagation mechanism is analyzed,focusing on the following aspects:A method is proposed for solving the propagation cost of parameter changes based on a combination of mechanical and network properties.Based on the impact of parameter changes on products,this paper proposes three evaluation indexes for parameter change propagation cost: change propagation index,cost coefficient,and correlation propagation intensity.Corresponding quantitative methods are proposed as follows: Firstly,the existing eigenvector centrality algorithm is improved to calculate the change propagation index of parameters by solving the parent and child eigenvector centrality of different and similar communities.Secondly,the expert evaluation method is used to establish the cost coefficient to represent the change cost of parameters.Thirdly,the correlation degree between parameters is extracted based on different forms of product design formula,and a quantitative index of correlation strength is established.Finally,these three variables are integrated to determine the quantitative method of parameter change propagation cost.A method is proposed for analyzing and decoupling parameter coupling correlations based on an inverse model for propagating parameter changes.In the first step,potential conflicts that may arise from propagating parameters are identified and classified into three types based on various combinations of parameter associations,and criteria for judging the corresponding types are provided.In the second step,four models for coupled parameter changes are constructed based on the inverse model of the basic model of change propagation.In the final step,a method for identifying complex network coupling models and a decoupling strategy based on the objective of minimizing the cost of propagating changes are proposed.A parameter coupling change path optimization method based on a hybrid elite backtracking-simulated annealing algorithm is proposed.Firstly,the existing backtracking algorithm is improved by incorporating the inner loop algorithm of the simulated annealing algorithm,resulting in the proposed hybrid elite backtracking-simulated annealing algorithm.Secondly,a comprehensive evaluation index is introduced for this algorithm,taking into account network complexity,association importance,and parameter change propagation cost.Finally,during the optimization of coupled change paths,the hybrid algorithm accepts other potential propagation parameters as alternative parameters with a certain probability,enters them into the elite library,and when the current path cannot fully absorb the change impact,returns part of the change impact to the alternative parameter nodes for path reconstruction.Thus,the preferential selection of the parameter coupling change path is completed.Based on the aforementioned content,a system for multi-parameter coupled change propagation and path selection was developed on the Solid Works platform.The development tool used was Visual Studio,and the system was programmed in C#.The backend database was implemented using Access.The multi-parameter coupled change propagation path optimization was successfully implemented,and the effectiveness of the research method was verified by running example scenarios.