Loading menu…

Remaining useful Life Prediction System for Industrial Power Converters Integrated into Enterprise SCADA

N.V. Bunin
N.V. Bunin * § Postgraduate Student Postgraduate Student
§ The Russian Presidential Academy of National Economy...

Send Message

To: Author

Remaining useful Life Prediction System for Industrial Power Converters Integrated into Enterprise SCADA

Article Fingerprint

ReserarchID

S7490

Remaining useful Life Prediction System for Industrial Power Converters Integrated into Enterprise SCADA Banner

Article in Press

Article

Abstract

In the context of industrial digital transformation, transitioning from scheduled preventive maintenance to condition-based maintenance is a key driver for enhancing production reliability. Industrial variable frequency drives (VFDs) / power converters (PCs), as the backbone of adjustable-speed drives for technological installations, are subject to stringent operating reliability requirements. Sudden failures of power components lead to cascading outages and severe financial losses in continuous manufacturing plants. An analysis of industrial practice reveals that enterprises frequently lack regulations for the maintenance of power electronics by operating personnel, specialized servicing competencies, and dedicated monitoring tools at operator workstations. This precludes the possibility of preemptive decommissioning of equipment (in particular, VFDs) for scheduled repair and incurs unjustified expenditures on periodic calendar-based servicing irrespective of actual equipment wear. Aim. Practical implementation of a two-tier decision support system (DSS) integrated into the existing automated dispatch control system for enterprise power supply (ADCS) for continuous condition monitoring and Remaining Useful Life (RUL) estimation of power components in variable speed drives within continuous chemical manufacturing. Methods. The first tier of the system implements a computationally lightweight algorithm for continuous calculation of the Asset Health Index (AHI) of power converters based on Thomas L. Saaty’s Analytic Hierarchy Process (AHP). The weighting factors of telemetry parameters (stator output current, heatsink temperature, output frequency, DC-bus voltage, and DC-bus voltage ripples) are derived via an expert-mathematical approach. The second tier utilizes a hybrid prognostic model combining the physics of failure of semiconductor devices with deep machine learning algorithms (CNN-LSTM). Results. The developed software is integrated into the ADCS (on the «CK-11» platform) of a chemical plant to monitor ABB ACS550-01-038A-4 power converters driving mission-critical pumping and blower equipment in sulfuric acid manufacturing. Data exchange is established through an existing programmable logic controller (PLC) of the process control system via the OPC UA protocol (with SCADA acting as an OPC client). Computational results are dynamically displayed in the dispatcher’s interface. Upon triggering a warning or alarm notification, the DSS automatically initiates a maintenance request in the enterprise outage and maintenance schedule management system. Conclusion. Practical commissioning of the two-tier DSS provides the foundation for a full transition from scheduled preventive maintenance to condition-based maintenance of variable-speed drive equipment (VFDs). Continued pilot operation will validate economic benefits and facilitate scaling to other technological workshops of the enterprise.

This is the heading

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

High-quality academic research articles on global topics and journals.

Remaining useful Life Prediction System for Industrial Power Converters Integrated into Enterprise SCADA

Nikita Bunin
Nikita Bunin
Vladislav Maslov
Vladislav Maslov