基于支持向量机的控制图异常模式参数估计

Parameters estimation for abnormal patterns of control charts based on support vector machine

针对控制图5种异常模式的6个参数,提出基于一对一算法的多类分类支持向量机的控制图异常模式下的参数估计方法。在模型构造中采用遗传算法优化支持向量机的参数。仿真实验结果表明,该方法结构简单,收敛速度快,具有较高的识别精度,适合于控制图异常模式的参数估计。

To properly estimate six parameters of five abnormal patterns of control chart, parameters estimation method for abnormal patterns of the control chart for one-against-one algorithm multi-class classification support vector machine was presented. In the modeling of structure, the genetic algorithm was used to optimize the parameters of SVM. The simulation results show that the proposed method is simple, of quick convergence and higher recognition rate, and applicable in the abnormal patterns parameters estimation of control charts.