基于制造数据驱动的质量波动溯源分析研究
Research on Quality Fluctuation Traceability Analysis Based on Manufacturing Data Driven

为研究中国中车青岛四方机床加工过程中的加工质量波动与加工制造数据之间的关联性及问题溯源，通过传感器实时采集机床加工过程的加工制造数据参数，利用三坐标测量机检测加工尺寸精度(即加工质量)，并通过MES系统导出数据。经过初步分析，认为温度是影响加工质量，造成加工质量波动的关键因素，通过采集轴承座和光栅尺位置处的温度数据，利用Matlab对温度数据和与之对应的加工质量数据进行关联性的回归拟合，利用神经网络进行优化得出最优模型，从而对优化加工质量的参数调整提供理论依据。同时对一日内的温度变化作出变化趋势的分析拟合，对一日内不同时间段的加工质量波动的溯源及优化作出指导。
To study the correlation between the processing quality trends and processing parameters during the machining process of CRRC Qingdao Sifang Machine Tool Co., Ltd., we collected real-time processing parameters through sensors during the machining process, used a three-coordinate measuring machine to detect the processing dimension accuracy (i.e., processing quality), and exported the data through the MES system. After a preliminary analysis, temperature was considered a critical factor that affected the processing quality. We collected temperature data from the bearing block and the grating ruler position, used Matlab to perform regression fitting on the correlation between the temperature data and the corresponding processing quality data, and optimized the data with neural networks to obtain an optimal model, providing a theoretical basis for parameter adjustments to optimize processing quality. At the same time, we conducted an analysis and fitting of the temperature change trends within a day, providing guidance for optimizing the processing quality during different time periods of the day.