论文部分内容阅读
提出一种基于GA-PLS和AdaBoost的液压系统故障诊断方法。该方法用遗传算法与偏最小二乘法相结合(Genetic algorithm-partial least squares,GA-PLS)的算法对初始特征向量进行筛选,提取出与故障信息相关程度高的特征向量,把该特征向量作为输入,运用AdaBoost(Adaptive Boost)方法建立分类器,以识别液压系统的工作状态和故障类型。对实验数据分析的结果说明,该方法能准确地选择出特征向量,并有效地应用于液压系统的故障诊断。
A fault diagnosis method of hydraulic system based on GA-PLS and AdaBoost is proposed. In this method, genetic algorithm and partial least squares (GA-PLS) algorithm is used to filter the initial eigenvectors, and the eigenvectors with high degree of correlation with the fault information are extracted. The eigenvectors are regarded as Input, using AdaBoost (Adaptive Boost) method to establish a classifier to identify the working status of hydraulic systems and fault types. The experimental results show that this method can accurately select the eigenvectors and can be effectively applied to the fault diagnosis of hydraulic system.