Authors: A. Makshanov, A. Musayev, D. Grigoriev
Title of host publication: Proceedings - 2024 International Conference on Industrial Engineering, Applications and Manufacturing, ICIEAM 2024
Abstract: This study addresses a comprehensive set of issues related to the construction and use of sequential algorithms for detecting spontaneous changes (discrepancies) in the probabilistic characteristics of multidimensional time series. The research is motivated by the challenges of mathematical support for decision-making processes based on multichannel monitoring of large systems and is dedicated to the analysis of the spatio-temporal dynamics of multidimensional time series measurements. As an alternative to traditional approaches, new technologies for anlyzing inter-channel connections are proposed. Dimensionality reduction technologies are used, based on the representation of data matrices in the first singular basis and multiple regression in the space of projections. The application of the developed approach is demonstrated in the task of analyzing the characteristics of turbulent flow based on pressure deviation measurements at various points in the volume.
Title of host publication: Proceedings - 2024 International Conference on Industrial Engineering, Applications and Manufacturing, ICIEAM 2024
Abstract: This study addresses a comprehensive set of issues related to the construction and use of sequential algorithms for detecting spontaneous changes (discrepancies) in the probabilistic characteristics of multidimensional time series. The research is motivated by the challenges of mathematical support for decision-making processes based on multichannel monitoring of large systems and is dedicated to the analysis of the spatio-temporal dynamics of multidimensional time series measurements. As an alternative to traditional approaches, new technologies for anlyzing inter-channel connections are proposed. Dimensionality reduction technologies are used, based on the representation of data matrices in the first singular basis and multiple regression in the space of projections. The application of the developed approach is demonstrated in the task of analyzing the characteristics of turbulent flow based on pressure deviation measurements at various points in the volume.