Gain-Scheduled Control of Blade Loads in a Wind Turbine-Generator System by Individual Blade Pitch Manipulation

Introduction

Floating offshore wind turbine-generator systems are expected to install in areas that have very deep waters. The stability of the power output and platform motion must be simultaneously established in these systems. However, the platform motion is induced by variations in aerodynamic loads of the wind turbine during rotation as well as variations in wind and wave. One effective approach is to develop a novel control strategy because the increase in the initial cost due to its implementation is slight as compared to development of a high-damping platform structure. Previous studies on this research field mainly focused on individual blade pitch manipulation.

Approach

The present study focuses on a gain-scheduled control approach of blade loads using individual blade pitch manipulation to reduce variations in aerodynamic loads of wind turbines during rotation as a previous stage to establish a novel control approach for floating wind turbine-generator systems. In previous works, the proportional gain to individually manipulate each blade pitch was fixed. In the present study, the gain-scheduled control approach to wind speed variations is developed because the dynamic characteristics of the blade loads depend on the wind speed. The development is conducted through a numerical analysis of a 5-MW on-shore wind turbine-generator system using the aeroelastic simulation model (FAST) and observed high wind speed data.

Main body of abstract

At high wind speeds, the rotor speed and the blade loads are controlled by the blade pitch manipulation. To maintain the rated rotor speed, the blade pitch is collectively manipulated by a proportional-integral control action. The individual blade pitch manipulation is applied to reduce the variations in the blade loads. The manipulation signal of each blade pitch is calculated from the two components of the blade loads in the d-q coordinates, which are converted from the sum of all the blade loads. The manipulation signal of each blade pitch is superposed on the collective blade pitch signal.
First, the sensitivity of the proportional gain for the blade loads is analyzed at constant wind speeds. The variations in the d-q components of the blade loads during one rotation are reduced by large proportional gain. However, the excessively large proportional gain causes unstable behavior of the blade loads; this is remarkable at high wind speeds. From this sensitivity analysis, the gain-scheduled strategy is established, in which the proportional gain is decreased with the increase in the wind speed. Consequently, the gain-scheduled strategy can increase the proportional gain at low wind speeds as compared with the conventional gain-fixed strategy. Second, the system dynamic behavior under observed high wind speed data is analyzed. To apply the gain-scheduled strategy to the dynamic operation condition, the proportional gain for the blade loads is varied to the collective blade pitch signal for the rotor speed control. The result shows that the variations in the d-q components of the blade loads during one rotation and the damage equivalent fatigue loads at the blade root are reduced as compared with those in the conventional gain-fixed strategy.


Conclusion

The present study developed a gain-scheduled control approach of blade loads using individual blade pitch manipulation to reduce imbalance in aerodynamic loads of wind turbines during rotation. Based on the sensitivity analysis using FAST, the proportional gain for the blade loads is varied to the collective blade pitch signal. The dynamic behavior analysis revealed that the variations in the blade loads during one rotation and the damage equivalent fatigue loads at the blade root are reduced as compared with those in the conventional gain-fixed strategy. This result suggests the possibility of controlling the blade loads to reduce the platform motion in floating offshore wind turbine-generator systems.


Learning objectives
The present study provides the new insights regarding the control approaches for the blade loads, especially, the effectiveness of the gain-scheduled control based on the individual blade pitch manipulation.