关键词: 汽车轻量化；大数据；工艺参数优化；预测性质量控制；多目标优化

Abstract: The aim of this paper is to propose an intelligent process parameter tuning solution based on big data and machine learning in order to meet the demands of optimization of process parameters and quality control in lightweight automotive manufacturing. The data collection platform covers the whole lightweight manufacturing process and gathers all sorts of data like material properties, process parameters, production equipment status, and the state of finished products. Neural networks technology will be applied to model the non-linear connections between process parameters and product quality which can automatically adjust the process parameters. Besides, a predictive quality control system has been created with the help of time series analysis and sequential anomaly detection that detects possible quality problems at an early stage when they arise during production. The findings of real-life production experiments indicate that this strategy effectively improves the quality of lightweight automotive components produced through it. The number of defects was reduced by 60.3%, the productivity gained by 12.6%, and energy usage decreased by 18.7% without compromising the quality of goods. It offers a useful example to promote the intelligent growth of the Chinese auto industry.

Key words: automotive lightweight; big data; optimization of process parameters; predictive quality control; multi objective optimization