Storage And Characteristic Analysis Of Monitoring Data For Spraying Production

As a popular manufacturing technology,spraying is widely used in various industries.The high temperature environment needed for spraying is easy to cause accidents,so it is necessary to monitor the spraying environment.On the other hand,because the current spraying process is complex and there are many factors affecting the qualified rate of spraying products,it is necessary to analyze the characteristics of production data.Aiming at the environmental monitoring in automobile parts spraying production,this paper uses cloud platform to store data and find out the factors affecting the product qualification rate from the stored data,and predicts the product disqualification rate under the influence of these factors.Firstly,apply for Aliyun ECS cloud platform,and build Node.js server and MongoDB non-relational database on Aliyun server for data storage and transmission.Through pymongo module in python,the data in MongoDB is invoked,and the characteristics are analyzed according to the product qualification rate.In the process of analysis,data preprocessing is carried out to deal with the missing values in the data and normalize the data.Lasso regression method was used to reduce the dimension of data features,and six main factors affecting the product qualification rate were selected from 15 data.Finally,the traditional BP neural network and the BP neural network based on genetic algorithm optimization were used to predict the average absolute percentage error,which reached 18.825% and 7.152%,respectively.It was found that genetic algorithm was used to analyze the characteristics obtained from lasso regression analysis.The optimized BP neural network model has a good effect on the prediction of product qualification rate,and provides scientific basis for decision-making and support for actual production.