基于人工免疫的异步电机故障检测与诊断

Detection and Diagnosis of Induction Motor Faults Based on Artificial Immunity

将反面选择算法引入异步电机故障检测与诊断,使用实数组成特征向量表示电机的工作状态。针对电机故障诊断的特点,提出一个改进的反面选择算法,改进检测器形成和训练方法。从感应电机的实测数据中提取各种特征样本,使用这些样本对算法进行的仿真结果,表明这一算法具有较高的故障检出率。

Introduces negative selection algorithm for detection and diagnosis of induction motor faults. Describes the work state of induction motor by eigenvectors consisted of real numbers. Aiming at the feature of induction motor faults, proposes a novel negative selection algorithm. Improves the generation and training methods of the detectors. Collects the whole of stylebooks from experiment data. The simulation result using measuring data indicates that the detection ratio using this algorithm is superior.