Industrial Multivariate Explorer: A solution with Matlab for time series mining

Abstract

The study introduces the development of a Matlab application named Industrial Multivariate Explorer (IME), designed to explore and extract information from time series data in chemical and thermo-energy processes. The tool allows visualization of up to eight normalized signals, selection of specific time windows with cursors, and calculation of descriptive statistics for static modeling. Results can be saved in Excel files, facilitating the analysis of quasi-stationary states in industrial processes. The application was tested with real data from a 250 MW thermal power plant and simulated data, demonstrating its utility for process monitoring and diagnostics.