工艺产品加工质量监测优化

Optimization of Process Quality Monitoring

质量监测可以有效地提高产品质量和生产效率。在复杂产品的生产过程当中,多个质量特性之间相互作用,共同对产品的生产质量产生影响。由于质量特性的数量较多、有些特性的关系是耦合的,因此准确诊断出异常变量是研究的难点。为了高效、准确地诊断出异常变量,提高产品的质量和生产效率,提出了基于改进网格优化的principal component analysis(PCA)-support vector machines(SVM)多元控制图均值偏移诊断模型。在模型训练之前,使用主元分析(PCA)算法对数据进行预处理,降低数据维数和提取数据特征信息;再用改进网格算法对支持向量机(SVM)的参数进行优化,最终得到优化的SVM模型。仿真结果表明,采用的方法与传统方法相比,训练时间更短,且拥有更高的分类准确率。

Quality monitoring can effectively improve product quality and production efficiency. In the production process of complex products, the interaction of multiple quality characteristics together affect the quality of production.The large number of quality characteristics, some characteristics of the relationship is coupled, so accurate diagnosis of abnormal variables is difficult. In order to efficiently and accurately diagnose the abnormal variables proposes and improve product quality and production efficiency a model of monitoring the mean shift of multivariate control charts based on the improved grid optimization principal component analysis （PCA）-support vector machines （SVM）.Before training the model, principal component analysis （PCA） algorithm is used to process the data to reduce data dimension and extract data feature information. Then, using the modified grid algorithm to optimize the parameters of support vector machine （SVM）. Finally, the optimized SVM model is attained. The simulation results show that compared with the traditional methods, the training time is shorter, and has a higher classification accuracy.