Blow molding machine control system performance degradation online diagnosis method and system based on operation data
By reconstructing the mold cavity pressure curve using virtual sensing technology and performing morphological quantitative analysis, the problem of insufficient perception of core process quality in the blow molding machine control system was solved, enabling early warning and accurate location of fault roots, thus improving maintenance efficiency and diagnostic accuracy.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NINGBO SHUANGDE TIANLI MASCH MFG CO LTD
- Filing Date
- 2026-04-03
- Publication Date
- 2026-06-30
AI Technical Summary
Existing technologies lack effective core process quality perception methods in blow molding machine control systems, making it difficult to identify early, gradual, and correlated performance degradation, and lacking the ability to locate the root cause of faults, thus failing to achieve low-cost, online, and accurate diagnosis.
By reconstructing the cavity pressure curve using virtual sensing technology, combining multi-source heterogeneous runtime timing data, and using a neural network model for morphological quantification analysis, early warning and accurate location of fault roots can be achieved. A data-driven and physical model-based approach is adopted to generate virtual cavity pressure timing curves, and degraded components are identified through morphological distortion indices and related data.
It achieves low-cost, high-sensitivity core process quality perception, enabling early warning and precise location of fault root causes, significantly improving maintenance efficiency and diagnostic accuracy, and forming a complete data-driven predictive maintenance closed loop.
Smart Images
Figure CN122308331A_ABST