Wind Turbine Online Models for Real-Time Multivariable Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional wind turbine control models lack the fidelity needed for accurate real-time operation due to limited computational capabilities, relying on simplified structural dynamics and aerodynamics that result in inaccurate performance and limited energy production.
Innovation Solution
Development of high-fidelity wind turbine control models that incorporate rigorous mechanical and aerodynamic equations, including flexible blade representations, accurate aerodynamic forces, and dynamic wake effects, while utilizing advanced algorithms to compute model Jacobians for real-time operation on limited computational platforms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If simplified quasi-steady aerodynamics models with lookup tables are used in conventional control models, then the computational complexity is reduced and real-time operation is enabled, but the model fidelity and predictive accuracy deteriorate
Solution Approach 1:
The patent segments the aerodynamic model into discrete blade elements along the span, with each element having its own independent force and moment calculations. This segmentation allows the complex aerodynamic behavior to be broken down into manageable computational units that can be processed in real-time while maintaining overall model fidelity.
Solution Approach 2:
The patent transitions from static lookup tables to a dynamic aerodynamic model that continuously calculates forces and moments based on current operating conditions. The model dynamically updates blade element velocities, aerodynamic coefficients, and structural responses in real-time, enabling accurate prediction of transient behavior while maintaining computational efficiency through optimized calculation sequences.
2Measurement precision
If more degrees of freedom are included in the turbine model to improve predictive accuracy of transient behavior, then the model fidelity improves, but the computational cost increases making real-time operation infeasible
Solution Approach 1:
The patent applies partial action by selectively modeling the most critical degrees of freedom for control applications. Rather than including all possible structural modes, the model focuses on the first few bending and torsional modes of the tower and blades that have the greatest impact on control performance, achieving sufficient accuracy without excessive computational cost.
Solution Approach 2:
The patent changes the parameter representation from high-fidelity simulation parameters to control-optimized parameters. The model uses reduced-order structural parameters that capture essential dynamics while being computationally efficient for real-time operation. Aerodynamic parameters are continuously updated based on operating conditions rather than pre-computed for all possible states.
3Ease of operation
If conventional simplified models are used for control applications, then real-time operation is enabled, but the control performance and energy production are limited due to inaccuracy
Solution Approach 1:
The patent implements continuous feedback by using the high-fidelity aerodynamic model to predict future states and comparing these predictions with actual sensor measurements. The model continuously adjusts its internal state estimates based on feedback from tower accelerometers, blade root strain gauges, and other sensors, enabling accurate real-time control that maximizes energy production while maintaining operational constraints.
Data Source
AI summary
A system for computing wind turbine estimated operational parameters and/or control commands, includes sensors monitoring the wind turbine, a control processor implementing a model performing a linearization evaluation to obtain a structural component dynamic behavior, a fluid component dynamic behavior, and/or a combined structural and fluid component dynamic behavior of wind turbine operation, and a module performing a calculation utilizing the linearization evaluation of the structural component dynamic behavior, the fluid component dynamic behavior, and/or the combined structural and fluid component dynamic behavior. The module being at least one of an estimation module and a multivariable control module. The estimation module generating signal estimates of turbine or fluid states. The multivariable control module determining actuator commands that include wind turbine commands that maintain operation of the wind turbine at a predetermined setting in real time. A method and a non-transitory medium are also disclosed.