摘要
为了提高动态过程质量异常模式识别的精度,提出一种基于主元分析的多特征融合方法.首先提取出样本数据的统计特征和几何特征;接着将混合的多种特征进行PCA处理,提取出主元特征向量;然后利用粒子群算法寻找SVM分类器的最优参数;最后,通过仿真实验与其他识别方法进行对比,实验结果表明:本文提出的多特征PCA融合方法具有较高的识别精度,为质量异常模式识别研究提供了新的方法.
Improving the recognition accuracy of quality abnormal patterns in dynamic process is quite important to realize real-time monitoring and diagnosis for automatics manufacturing. A novel multi-feature fusion method based on PCA was proposed. First, statistic features and shape features were extracted from sample data. Then, mixed multi-feature was extracted with principal component analysis method. After that, particle swarm optimization was applied to find the optimal parameters of SVM. At last, the method proposed in this paper was compared with other models with simulation experiment. Simulation results show that the proposed algorithm has very high recognition accuracy. It is significant for quality monitoring and diagnosis in manufacture dynamic process.
出处
《河南大学学报（自然科学版）》
CAS
北大核心
2014年第6期731-737,共7页
Journal of Henan University:Natural Science
基金
国家自然科学基金项目(71272207
61271146)
