A comparison study of basic data-driven fault diagnosis and process monitoring methods on the benchmark Tennessee Eastman process

This paper provides a comparison study on the basic data-driven methods for process monitoring and fault diagnosis (PM–FD). Based on the review of these methods and their recent developments, the original ideas, implementation conditions, off-line design and on-line computation algorithms as well as computation complexity are discussed in detail. In order to further compare their performance from the application viewpoint, an industrial benchmark of Tennessee Eastman (TE) process is utilized to illustrate the efficiencies of all the discussed methods. The study results are dedicated to provide a reference for achieving successful PM–FD on large scale industrial processes. Some important remarks are finally concluded in this paper.

► We review most popular data-driven PM–FD methods with recent developments.
► Systematic comparison and analysis among these methods are given.
► All the methods are tested on TE process.
► Based on the simulations, some important concluding remarks are presented.
► The results provide a reference on performance of PM–FD for large scale processes.