A parallel computing method for fault diagnosis based on Hadoop

To improve the efficiency and the speed of fault diagnosis for multimode process under big data environment, a new parallel computation fault diagnosis approach, called GFCM-VMD-HM, is proposed for multimodal TE process based on Hadoop platform. Firstly, the global fuzzy c-means (GFCM) clustering algorithm is applied to distinguish the operation modes. Then, Variational Mode Decomposition(VMD) algorithm is introduced for the sake of filtering. After that, the data are encoded string files and put to Hadoop Distributed File System (HDFS). Finally, character statistics program is written to output multiple files under MapReduce data frame to complete the fault feature extraction and diagnosis. An experimental study is carried out to evaluate the performance of the proposed algorithm. The results obtained show that the method saves fault diagnosis time by means of parallel computing.