Wind Turbine Predictive Monitoring via SCADA Temperature Differentials

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional maintenance methods for power generating plants, such as wind turbines, rely on preventive maintenance, which can lead to unnecessary servicing and unpredictable component failures, resulting in increased costs and downtime.

Innovation Solution

A predictive monitoring method using SCADA data to calculate differential temperature values, extract features like averages and standard deviations, and set threshold alerts for wind turbine components, allowing for early identification of potential failures and minimizing downtime.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If preventive maintenance is applied periodically regardless of component state, then maintenance simplicity and ease of implementation are improved, but unnecessary servicing costs increase and component failure prediction capability deteriorates

Engineering Contradiction:
Improvemaintenance simplicityVSAvoidcomponent failure prediction capability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent transforms fixed periodic maintenance parameters into dynamic parameters based on actual component condition. Temperature differential values are continuously monitored and compared against dynamically calculated thresholds (mean + 3 standard deviations), allowing maintenance timing to adapt to actual component state rather than following a rigid schedule. This resolves the contradiction by maintaining operational simplicity while dramatically improving failure prediction capability.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system implements continuous feedback loops where temperature data from multiple turbines is collected, processed to calculate differentials and statistical parameters, and used to adjust maintenance decisions in real-time. The threshold calculation incorporates ongoing statistical analysis of temperature patterns, creating a self-adjusting maintenance system that improves reliability without sacrificing operational simplicity.

Inventive Principle:
Principle #23Feedback

2Device complexity

If preventive maintenance is applied periodically, then implementation complexity is reduced, but total maintenance cost increases due to unnecessary servicing

Engineering Contradiction:
Improvemaintenance system complexityVSAvoidtotal maintenance cost
Core Design Contradiction:
Device complexityVSLoss of energy

Solution Approach 1:

The patent changes maintenance frequency from a fixed parameter to a dynamic parameter determined by statistical analysis of temperature differentials. By calculating thresholds based on mean + 3 standard deviations of historical data, the system automatically adjusts maintenance timing to match actual component needs, eliminating unnecessary servicing and reducing total maintenance costs while maintaining manageable system complexity.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system applies monitoring and analysis resources selectively based on component condition rather than uniformly across all components regardless of state. By focusing detailed statistical analysis only on components showing abnormal temperature differentials, the system reduces overall maintenance costs while maintaining effective oversight, avoiding the excessive action of servicing all components on every scheduled interval.

Inventive Principle:
Principle #16Partial or excessive action

3Speed

If monitoring time period is shortened to detect failures earlier, then failure detection speed is improved, but data accuracy and reliability deteriorate due to insufficient data samples

Engineering Contradiction:
Improvefailure detection speedVSAvoidtemperature differential accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent performs preliminary calculations of mean and standard deviation thresholds using historical temperature data collected over extended periods. These pre-established statistical parameters serve as reference benchmarks that enable rapid real-time comparison with current temperature differentials. This preliminary preparation allows the system to achieve both fast failure detection and high measurement precision by comparing against pre-validated statistical thresholds rather than requiring extensive real-time data accumulation.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system resolves the time-accuracy tradeoff by introducing a statistical dimension to the monitoring process. Instead of relying solely on the duration of monitoring, the patent uses statistical parameters (mean and standard deviation calculated from historical data) to enhance the information content of each measurement. This dimensional transformation allows rapid detection while maintaining precision by leveraging accumulated historical knowledge rather than requiring prolonged real-time observation.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP4127465B1Method for predictive monitoring of the condition of wind turbines
Publication Date: 2024.02.07 FLUENCE ENERGY LLC
  • EP4127465B1 patent drawingFigure 1
  • EP4127465B1 patent drawingFigure 2~3
  • EP4127465B1 patent drawingFigure 4

AI summary

The invention relates to a method for predictive monitoring of the condition of wind turbines, the method comprising the steps of selecting at least one wind turbine inside a wind farm and at least one component of the wind turbine; acquiring SCADA data comprising operational data of the wind farm during a preselected time period, wherein the SCADA data comprises temperature values of the at least one component of the wind turbine during the preselected time period; processing SCADA data comprising calculating differential data, wherein the differential data is a difference between the temperature values of the selected turbine component of the selected wind turbine and an average temperature of the selected wind turbine component in at least two wind turbines in the wind farm; defining a monitoring time period to monitor the component; extracting features, wherein the extraction of the features comprises calculating at least one predetermined statistic of the differential data during the monitoring time period, and saving the predetermined statistic as a monitoring feature; testing if at least one monitoring feature exceeds a threshold value.