Fault diagnosis in industrial chemical processes using interpretable patterns based on Logical Analysis of Data

作者：

Highlights：

• Logical Analysis of Data is applied as a pattern-based diagnostic method.

• It is used to detect and analyze faults in the industrial chemical processes.

• It relies on extracting a set of strong and interpretable patterns.

• LAD's patterns can help the user to relate the fault to its causes.

• LAD shows a performance that is comparable to the common accurate methods.

摘要

•Logical Analysis of Data is applied as a pattern-based diagnostic method.•It is used to detect and analyze faults in the industrial chemical processes.•It relies on extracting a set of strong and interpretable patterns.•LAD's patterns can help the user to relate the fault to its causes.•LAD shows a performance that is comparable to the common accurate methods.

论文关键词：Fault detection and diagnosis,Industrial chemical processes,Tennessee Eastman Process,Logical analysis of data,Machine learning and pattern recognition,Black liquor recovery boilers

论文评审过程：Received 2 June 2017, Revised 16 November 2017, Accepted 17 November 2017, Available online 23 November 2017, Version of Record 1 December 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.11.045