Correlation Analysis And Feedforward Control Research Of Engine Assembly Process And Performance Data

Engine assembly is a crucial part of the engine manufacturing process.A large number of sensors and electronic testing equipment are arranged in the engine assembly line to collect various process parameters and engine quality performance during assembly and quality inspection.In the context of industrial big data,making full use of these production data and extracting the effective value of the engine production assembly process can effectively improve production efficiency and engine downline pass rate.The research content of this paper is on the basis of the engine assembly line process data and engine performance test data.Firstly,the related system is obtained and the correlation model is established through the research on the relationship between engine assembly line technology and performance.The Pearson correlation coefficient method is used to analyze and calculate the main detection items such as exhaust pressure and cylinder torque to judge the influence degree of each station process parameter on engine quality and performance.The significance test and the data fitting of least squares are used to find the key influencing factors.Secondly,prediction of engine performance using ElasticNET elastic network algorithm and artificial neural network algorithm is based on key influencing factor data and engine quality detection data.And the model result was assessed by mean square error and interpretation variance.Through the adjustment of engine assembly process parameters to ensure the controllability of engine performance,and achieve feedforward control of engine performance.Then,the Spark framework-based big data analysis method is used to design the production data acquisition and storage scheme in the engine assembly process and realize the distributed storage of data.Big data parallel analysis and forecasting system in industrial production was designed.Finally,through the requirements analysis,software architecture design,software function design,database design,software detailed design and software testing,the engine performance prediction and visualization software system design,development and application are realized.