A part of FEV Group
Advanced prediction strategies for component and system level failure/faults
Used solutions
-
Cloud & Data Solutions
Find the latest events here
A part of FEV Group
Introduction
At FEV.io, we develop innovative solutions by combining mathematics, AI, and engineering analytics. Our expertise includes advanced prediction strategies, such as Prognostics and Asset Health Analytics (PAHA), aimed at optimizing performance, ensuring safety, and extending the lifespan of components and systems across industries.
Background
We collaborated with a global OEM to deploy a comprehensive PAHA Framework for predictive maintenance and health analytics. This project targeted system-level health monitoring, fault detection, and predictive maintenance for complex assets, including EV motors, aircraft engines, IC engines, and heavy machinery components.
Approach
1. Component-to-Fleet Health Analytics:
- Designed a scalable framework for health analytics at component, system, and fleet levels.
- Developed real-time health estimation and predictive maintenance applications for dynamic operations.
- Developed critical event detection mechanisms for issues like short circuits and thermal runaway.
- Enhanced operational visibility through interpretative engineering dashboards.
2.Advanced Prediction and Maintenance Strategies:
- Implemented white-box algorithms for fault detection, leveraging modular approaches for additional assets.
- Developed hybrid models integrating physics-based and AI-driven analytics for accurate RUL estimation.
- Delivered predictive maintenance solutions across mechanical, electrical, and thermal domains.
3. Engineering Visualization:
- Created dynamic, interactive dashboards to monitor system health and track degradation trends.
- Enabled real-time visualization of critical data, including thermal anomalies, short circuits, and performance deviations.
- Simplified complex data insights into actionable metrics for better decision-making.
4. Industry-Specific Adaptations:
- Tailored the PAHA framework to diverse use cases, including heavy vehicles, aircraft engines, and EV motors.
- Ensured seamless integration across various asset types, from powertrains to bearings and gearboxes.
Impact
This structured approach delivered measurable benefits:
- Achieved a 90% reduction in false alarms, significantly improving reliability.
- Realized over 12% predictive maintenance savings, reducing downtime and operational costs.
- Enhanced system-wide safety and performance by enabling early detection of faults and accurate RUL predictions.
The PAHA framework set a new benchmark for predictive maintenance, offering scalable, adaptable solutions that extend asset lifespans and optimize operational efficiency across industries.
