Research On Wind Turbine Load Control Based On Individual Pitch Control

Take the project supported by National Natural Science Foundation of China (WindTurbine Loads Control Strategy Considering the Nonlinear Uncertainties Dynamics) asbackground, the large scale double fed wind turbine is taken as the research object in thisthesis. For complex operating environment characteristic of large wind turbines, as well asthe dual demands of wind turbine power output stability and reduce the load of keycomponents. Therefore in the dissertation the proposed scheme is to design H2/H∞drivetrain torsional load controller and model-free adaptive rotor unbalance load controller toensure output power while reducing the dynamic load of the key unit of wind turbine (rotorand drive train).Through analyzing dynamic load on rotor and drive train of wind turbines, dynamicload control strategies is proposed by adjusting the pitch angle to reduce dynamic load,which bases on electric pitch actuator, and ultimately achieving the goal of load reduce.The design of controller is based on the characteristics of system and controller, therandomly uncertain external disturbances and the requirement of power and load reduce ofwind generation system.For the complex drive train structure, strongly nonlinear, uncertainty andmulti-parameter changes, this paper presents a multi-objective H2/H∞drive train loadcontroller based on Kalman state estimation. The controller design is optimize performanceand robust stability. H2/H∞controller has a very significant inhibition for unit loadfluctuations root-mean-square response and peak response while considering the H2/H∞performance.For the nonlinear relationship among wind speed, pitch angle and unbalance load,imprecise modeling problem. The model-free adaptive control strategy is proposed basedon rotor unbalance load. This control strategy does not rely on accurate mathematicalmodel of wind turbine, while solving the problem of nonlinear coupling and time-delaymeasurements of three blade pitch angle coursed by rotor dynamic loads. Finally, for the proposed methods, the simulations and experiments are conducted,which solving the nonlinear uncertain problem. The control strategy can ensure outputpower stable and achieve the reduce of key components load.