Wind Turbine Sensor Failure Estimation via Luenberger Observer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern wind turbines face challenges in maintaining control and reducing loads when sensor failures occur, particularly for individual pitch control systems, as they rely on complete data from bending moment sensors for optimal blade angle adjustment.
Innovation Solution
A method and device that utilize a reduced Luenberger observer to estimate missing sensor variables, allowing the IPC control to continue functioning by adapting a model based on available sensor signals and angle of attack data, ensuring the safety and efficiency of wind turbine operations even in the event of sensor failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a sensor fails in an IPC-controlled wind turbine, then the control system loses complete data for optimal blade angle adjustment, but installing redundant sensors increases system complexity and cost
Solution Approach 1:
The patent creates a virtual copy of the failed sensor signal through mathematical modeling. A reduced-order model replicates the behavior of the non-functional bending moment sensor by using data from operational sensors and aerodynamic relationships, providing an estimated signal that allows the IPC control to continue functioning without physical redundancy
Solution Approach 2:
The patent introduces an intermediary computational layer between the physical sensors and the control system. This intermediary model processes available sensor data and generates estimated values for missing signals, acting as a mediator that bridges the gap caused by sensor failure and maintains control system operation
2Ease of operation
If complete sensor data is available, then optimal individual pitch control can be achieved, but sensor failure creates data gaps that prevent proper control
Solution Approach 1:
The patent transforms the control approach by changing from direct use of physical sensor parameters to using modeled estimated parameters. When sensors fail, the system switches to using modeled bending moment estimates derived from aerodynamic parameters and operational sensor data, maintaining control optimization despite information loss
Solution Approach 2:
The patent replaces the mechanical sensor measurement system with a computational modeling system. Instead of relying on physical bending moment sensors, the system uses aerodynamic models and available sensor data to compute equivalent bending moment values, substituting mechanical measurement with computational estimation
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for determining an estimated value (S1) for at least one measured variable (M1, M2, M3) of a wind turbine (WKA), wherein the measured variable represents in particular a bending torque on a blade root of a rotor blade of the wind turbine from a rotor plane. The method comprises a step of reading in (300) the measured variable and an angle of attack (u) for the rotor blade. The method further comprises a step of adapting (302) a model, which simulates an estimated value of the measured variable on the basis of a modeling instruction for a relationship between the measured variable and the angle of attack, wherein the model is adapted using the read-in measured variable and the read-in angle of attack. Finally, the method comprises a step of providing (304) the estimated value for the at least one measured variable using the model.