Adaptive Wind Turbine Maintenance Scheduling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current condition-based maintenance systems for wind turbines and similar complex systems face inefficiencies due to pre-defined hard-coded alarm limits that are not adaptable to part-load operations, leading to high false alarm rates and reduced sensitivity to deviations, resulting in extended downtime and increased costs.

Innovation Solution

A method that involves obtaining physical data from measuring instruments, calculating optimal service intervals to minimize downtime, and establishing a condition-based maintenance plan that considers failure modes and their probabilities, allowing for the gradual introduction of condition monitoring to improve system availability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If pre-defined hard-coded alarm limits are used for condition-based maintenance, then the maintenance trigger is established, but the sensitivity to deviations is reduced and false alarm rate increases

Engineering Contradiction:
Improvemaintenance trigger reliabilityVSAvoiddeviation detection sensitivity
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The alarm limits are transformed from static pre-defined values to dynamic adaptive limits that automatically adjust based on the specific turbine's operational characteristics and historical data. The system continuously learns and adapts to each turbine's unique behavior patterns, enabling precise detection of actual deviations while avoiding false alarms from normal operational variations.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the parameters used for alarm limit determination from fixed universal values to variable values specific to each turbine's operational context. By analyzing historical operational data and identifying turbine-specific normal ranges, the system dynamically adjusts alarm thresholds to match actual operational conditions, thereby improving both reliability and sensitivity.

Inventive Principle:
Principle #35Parameter changes

2Ease of manufacture

If pre-defined hard-coded alarm limits are used, then the monitoring system can be implemented, but the false alarm rate increases due to scattering of measurement values and individual turbine parameters

Engineering Contradiction:
Improvemonitoring system implementationVSAvoidalarm accuracy
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

The monitoring system performs self-adjustment by automatically learning each turbine's operational characteristics from its own historical data. The system independently determines turbine-specific alarm limits without requiring manual calibration or external intervention, adapting to each turbine's unique measurement value scattering patterns and operational parameters.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system incorporates continuous feedback loops where alarm performance is monitored and used to refine future alarm limit settings. By analyzing the outcomes of previous alarms and comparing them with actual turbine conditions, the system continuously improves its alarm accuracy, reducing false alarms while maintaining reliable fault detection.

Inventive Principle:
Principle #23Feedback

3Reliability

If condition monitoring is introduced for all failure modes, then system availability improves, but the technical complexity and costs increase

Engineering Contradiction:
Improvesystem availabilityVSAvoidmonitoring system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

Instead of applying uniform condition monitoring to all failure modes across all turbines, the system implements targeted monitoring focused on each turbine's specific critical failure modes. By identifying and prioritizing the most impactful failure modes for each individual turbine based on operational data and risk assessment, the system achieves significant availability improvements while minimizing unnecessary monitoring complexity.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The monitoring system is segmented into modular components that can be selectively activated based on turbine-specific needs. Rather than a monolithic complex system monitoring everything, the solution divides monitoring into discrete, manageable modules focused on specific failure modes, allowing incremental implementation and reduced overall complexity.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP2837984B1Process to optimize the maintenance of technical systems
Publication Date: 2019.05.22 UPTIME HLDG
  • EP2837984B1 patent drawingFigure 1
  • EP2837984B1 patent drawingFigure 2a
  • EP2837984B1 patent drawingFigure 2b

AI summary

A process to establish optimised preventive (condition based) maintenance of wind parks is described. The activities aim at preventing faults due to all those failure modes, which are limiting the availability and lifetime of wind turbines acc. to expert reasoning.