
Wind Turbine Fault Diagnosis & SCADA Analysis
Diagnose wind turbine faults fast using SCADA data and sensor pattern analysis
What You Can Do
You can rapidly diagnose wind turbine faults by feeding Claude your SCADA data, alarm logs, and sensor readings. Claude structures the diagnostic analysis across mechanical, electrical, and hydraulic subsystems, cross-references symptoms against known failure modes, and generates prioritized hypotheses with specific next steps. This accelerates troubleshooting from hours of manual data review to minutes of focused investigation.
Features
extract anomalies from multi-parameter sensor logs and identify correlations between warning signals across subsystems
cross-reference observed symptoms against documented wind turbine fault signatures (gearbox, generator, pitch, yaw, brake systems)
prioritize diagnostic theories based on likelihood, subsystem criticality, and available evidence from your data
distinguish temporary sensor glitches from genuine component degradation using trend analysis
determine whether the fault requires immediate field technician dispatch or can be remediated remotely
rank multiple reported faults by risk, repair cost, and downtime impact across your turbine fleet
analyze current operation signatures against commissioning baseline data to quantify performance degradation
generate step-by-step verification procedures to confirm suspected faults before committing repair resources
Example Output
Example 1: Gearbox Fault Diagnosis
Input: SCADA showing 15% power loss, elevated vibration on high-speed shaft (HSS), elevated gearbox oil temperature, acoustic signature change over 48 hours.
Output:
- Primary hypothesis: Early-stage gearbox bearing wear (77% confidence)
- Supporting indicators: HSS vibration increase correlates with power loss; oil temp rise typical of bearing friction
- Secondary hypotheses: Gear tooth pitting (18% confidence), bearing cage failure (5% confidence)
- Recommended action: Schedule thermography inspection within 72 hours; increase monitoring frequency to 4-hour intervals; reduce turbine to 70% load pending field confirmation
Example 2: Intermittent Electrical Fault
Input: Random reactive power spikes, grid connection dropouts lasting 2–8 seconds, no thermal anomalies; events occur 3–4 times daily at unpredictable intervals.
Output:
- Primary hypothesis: Loose connection or capacitor bank contactors (62% confidence)
- Secondary hypothesis: Grid-side voltage sag triggering nuisance disconnects (28% confidence)
- Tertiary hypothesis: PLC firmware glitch causing false fault detection (10% confidence)
- Recommended action: Remote firmware log analysis first (low cost); coordinate with grid operator for voltage stability check; schedule electrical technician for terminal tightness inspection if remote checks clear
What's Included
- SKILL.md instruction file: complete diagnostic workflow, decision trees, and subsystem-specific fault patterns
- SCADA Data Template: structured format for organizing sensor logs, alarm timestamps, and performance metrics for analysis
- Fault Mode Reference Guide: documented failure signatures for gearbox, generator, pitch system, yaw system, brake, and electrical subsystems
- Diagnostic Decision Tree: flowchart for prioritizing hypotheses based on symptom clusters and subsystem criticality
- Field Action Checklist: verification procedures and safety steps for technician dispatch
Who It's For
- Wind turbine O&M engineers — diagnose complex multi-symptom faults without manual SCADA log review
- Predictive maintenance specialists — correlate SCADA trends with maintenance schedules and downtime forecasting
- Remote monitoring center technicians — triage alarms and prioritize field technician dispatch across turbine fleets
- Wind farm operations managers — assess fault severity and business impact to allocate maintenance resources efficiently
- OEM field service teams — support warranty troubleshooting and post-installation commissioning verification
Best For
- Diagnosing degraded turbine performance (power loss, efficiency drop, increased vibration)
- Analyzing intermittent or inconsistent fault patterns with multiple warning indicators
- Prioritizing maintenance work orders when multiple turbines report simultaneous issues
- Cross-referencing SCADA alarms against known failure modes to eliminate false positives
- Generating remote remediation recommendations before scheduling field technician visits







