Comparative Multivariate Characterization of Sensor Signatures Across Manufacturing Conveyor Motors during Recorded Stall Conditions
Predictive maintenance of manufacturing conveyor systems requires understanding how sensor behaviour varies among individual equipment units. This study comparatively characterized the multivariate sensor signatures of three conveyor motors, designated Motor 1 (M1), Motor 3 (M3), and Motor 4 (M4), during recorded stall conditions. The analysis considered temperature (TempC), raw sensor output (Raw), x-, y-, and z-axis measurements, and a derived three-axis resultant magnitude, with available speed records examined as contextual data. After exclusion of initialization observations, 965,610 valid observations were retained for M1, 923,020 for M3, and 404,003 for M4. One-minute median aggregation produced 181, 178, and 77 analytical windows, respectively. Descriptive statistics, Kruskal–Wallis tests with epsilon-squared effect sizes, Bonferroni-adjusted Mann–Whitney U tests with rank-biserial correlations, and within-motor Pearson correlations were applied. Substantial motor-level differences were observed, particularly in the x- and y-axis measurements (ε² ≈ 0.98–0.99). Temperature also differed markedly among motors. Within each motor, temperature and Raw exhibited highly consistent inverse relationships (r = −0.9991, −0.9987, and −0.9983 for M1, M3, and M4). The z-axis was strongly associated with resultant magnitude (r ≥ 0.9998). The findings demonstrate that recorded conveyor stall conditions can exhibit distinct motor-specific multivariate signatures and provide a comparative statistical baseline for subsequent condition-monitoring and predictive-maintenance modelling. Matched normal-operation data and repeated independent fault events will be required to est
Authors
- Olusesan Orebanjo (ORCID: https://orcid.org/0009-0000-9736-9346)
- Lucky Olatunde ODEJIMI
- Olorunleke Aregbe
Institutions
- Igbinedion University (NG)
Publication Details
- Published
- 2026-11-15
- DOI
- https://doi.org/10.54117/9t9a4d67
- Primary Topic
- Machine Fault Diagnosis Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00