A Change Point Detection Framework for Operating Regime Segmentation in Induction Motor Drives
Open Access
- 11 May 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 71043-71058
Abstract
This paper addresses the problem of Change Point Detection (CPD) in Induction Motor (IM) drive measurements by presenting a structured CPD-based framework for the identification and segmentation of transitions between operating regimes in IM drive systems. IM drive measurements are collected from an experimental test bench designed to reproduce realistic operating scenarios of the drive system. The proposed framework is based on CPD principles and enables the identification of break points associated with transitions between operating regimes. These break points are used to segment the measurement data into physically meaningful operating regimes, supporting the analysis of drive behavior and the assessment of load- and condition-related effects in the IM drive system. A structured procedure for data processing, parameter configuration, and result interpretation is defined to account for the specific characteristics of electric drive signals and their underlying physical behavior. The framework is evaluated on two independent IM drive systems, each providing distinct measurements under diverse operating conditions. The evaluation focuses on the consistency, robustness, and physical relevance of the detected transitions and the resulting operating regime segmentation. The results indicate that the framework provides a reliable and systematic methodology for CPD-based monitoring, condition assessment, and operating regime analysis in complex IM drive systems.Keywords
Funding Information
- Bundesministerium für Klimaschutz, Umwelt, Energie, Mobilität, Innovation und Technologie
- Österreichische Forschungsförderungsgesellschaft
This publication has 40 references indexed in Scilit:
- Change-point detection with deep learning: A reviewFrontiers of Engineering Management, 2025
- Unsupervised Change Point Detection and Trend Prediction for Financial Time-Series Using a New CUSUM-Based ApproachIEEE Access, 2022
- Condition monitoring and fault diagnosis of wind turbines based on structural break detection in SCADA dataRenewable Energy, 2022
- Two multivariate online change detection modelsJournal of Applied Statistics, 2020
- Review on Machine Learning Algorithm Based Fault Detection in Induction MotorsArchives of Computational Methods in Engineering, 2020
- Applications of machine learning to machine fault diagnosis: A review and roadmapMechanical Systems and Signal Processing, 2020
- Selective review of offline change point detection methodsSignal Processing, 2020
- A survey of methods for time series change point detectionKnowledge and Information Systems, 2016
- Three estimators of the Mahalanobis distance in high-dimensional dataJournal of Applied Statistics, 2012
- Sequential change‐point detection based on direct density‐ratio estimationStatistical Analysis and Data Mining: An ASA Data Science Journal, 2011