Research On Quality Optimization Of SMT Process In D Company Based On Bayesian Network

In recent years,China’s electronic information industry has made remarkable progress.The high and new technology represented by the information industry has driven the vigorous development of the whole industry.Industrial upgrading,quality improvement and technoiogical breakthrough have become the top priority of the whole industry,As the key manufacturing technology of PCB,the basic component of electronic information products,the quality control and optimization of SMT process has also attracted much attention because of its important role in the manufacturing of core components of products.Company D focuses on the fields of power distribution,communication and fire protection,and aims to provide professional and reliable online and offline integrated services.Therefore.the company invests a lot of costs every year to improve product quality and functional reliability,and sets up an independent team to be responsible for offline delivery,accident tracking and quality optimization.This thesis focuses on the manufacturing process of a certain type of distribution power detection products of D company.Firstly,starting with the fault data in the actual use of the products,this thesis reversely locates the key link of product production and manufacturing,namely SMT process,as the key research object of this thesis,and analyzes and designs the quality optimization scheme to improve the reliability of the manufacturing process,Finally,the purpose of process quality optimization of enterprise core manufacturing technology is achieved.The main research contents of this thesis are as follows:Using FMECA product fault mode analysis method,this thesis analyzes the fault data of distributed electric communication products in actual use,locates SMT,the key manufacturing process of the core parts in the product system,and puts forward the demonstration method to determine this process as the research object of this thesis.The fuzzy analytic hierarchy process model of quality information is constructed to preliminarily clean the redundant quality data in SMT process and form the data samples which can be used in the research method.The SMT process quality traceability model was obtained by using the bayesian network theory and the data samples formed by the above hierarchical analysis.The manufacturing process quality was further traced to obtain the key processes and influencing factors of the process,as well as the quality improvement points of other important processes.A quality prediction model based on BP neural network was constructed for the characteristic quality parameters of key processes,and a multi-objective optimization function was constructed for the output quality indexes.The improved particle swarm optimization algorithm was used to search and solve the characteristic quality parameters,and finally the optimization parameter solution set of key processes was obtained.Based on the system analysis method,the quality control of components,which is an important process quality improvement point,is proposed based on system control and supplier comprehensive evaluation management.To sum up,the SMT process quality optimization program is finally formed,which provides reliable ideas for enterprises to improve process quality control,reduce costs and improve efficiency.