Large Rotating Machinery Fault Diagnosis and Knowledge Rules Acquiring Based on Improved RIPPER
暂无分享,去创建一个
The data of fault monitoring for large rotating machine are large and noisy, there are some relationships between properties and property values, which are coincide with the rules of data mining technology. The data mining technology was studied in order to obtain the laws and classify the faults. The improved RIPPER (Repeated Incremental Pruning to Produce Error Reduction) data mining rule learning algorithm was studied for large rotating machine, the rules set files were obtained by analyzing the fault samples and updated in time. The extracted knowledge rules could also be used as the real time diagnosis of common faults.
[1] Andreas D. Lattner,et al. Experimental Comparison of Symbolic Learning Programs For The Classification of Gene Network Topo , 2003 .
[2] Johannes Fürnkranz,et al. Incremental Reduced Error Pruning , 1994, ICML.
[3] Padhraic Smyth,et al. Knowledge Discovery and Data Mining: Towards a Unifying Framework , 1996, KDD.