基于支持向量机的控制图在线检测和分析系统的研究

针对控制图在线检测和分析的要求，提出了系统基本框架。利用一对一算法的多类分类支持向量机进行控制图模式识别和异常模式下参数估计。在模型构造中，采用混合核函数，并利用遗传算法优化混合核函数支持向量机参数。仿真结果和实际应用表明：该方法结构简单、收敛速度快，识别准确率高，能够满足控制图在线检测和分析的需要。

Study on On-line Detection and Analysis System of Control Chart Based on Support Vector Machine

To satisfy the needs of on line detection and analysis of control chart, the general framework was presented. A method based on one against-one-algorithm multi-class classification support vector machine was proposed. In the modeling of structure, the hybrid kernel was applied, and the genetic algorithm was used to optimize the parameters of SVM. The simulation and ap plication results show that the performance of the proposed method has so many advantages such as simple structure, quick convergence and high aggregate classification rate, that it can be applied in on --ine detection and analysis of control chart.