Research On The Technology Of Independent Variable Pitch Control For Wind Turbine

In response to the great challenge of the energy crisis and climatechange, our contry energetically develop renewable energy. Wind powerhas the low cost of development, pollution-free to the environment andinexhaustible, leading the exploitation and utilization of renewable energyin the whole society, becoming one of the most size and application valuein the all new renewable energy. Control technology is the key to thewind turbine system, for independent variable pitch control is remarkablein terms of reducing the blade fatigue stress and improving the quality ofelectrical energy, therefore, researching on the technology of independentvariable pitch control has important theoretical value and practicalsignificance to realize localization of wind power.Firstly, according to the aerodynamic characteristic and the principleof variable pitch control，analyzing the effect acting on the wind turbinethat generated by turbulence, wind shear effect and shadow effect. Withthe method of combining analysis modeling and experiment data,establishing the complex, changeable, nonlinear, strong coupling windturbine control system which can be used to describe the dynamicbehavior of the wind turbine，this model not only has the generality butalso is suitable for control purpose. In addition, it lays the foundation forresearching independent pitch control in the subsequent chapters byestablishing the time-invariant model of wind turbine based on thecoordinate transformation.Secondly, in view of the uneven force, unbalanced load problem inthe collective variable pitch, this topic starting with weight coefficientand load feedback, researching azimuth weight coefficient basedindependent variable pitch control strategy and load feedback basedindependent variable pitch control strategy. Optimizeing PID parametersthrough particle swarm optimization in the MATLAB, the simulationresults show that the independent variable pitch control has the obviouseffect in terms of stabilizing output power and reducing the unbalancedload compared to the collective variable pitch control. And through the comparison and analysis, the load feedback based independent variablepitch control strategy has stronger response to the changing wind, so itscontrol effect is superior to the azimuth weight coefficient controlstrategy, but the cost of the latter is low and easy to implement, it is easierto get promotion in the field of practical engineering application.Finally, for support vector machine (SVM) control algorithm canautomatically design model complexity, and has the very stronggeneralization properties and learning ability, In combination with weightcoefficient, this topic proposed SVM weight coefficient based onindependent variable pitch control strategy. The nonlinear systemidentification of wind turbine in independent variable pitch can befinished by support vector regression (SVR), and using the incrementalalgorithm and reductive algorithm to realize online SVM weightcoefficient based on independent variable pitch control strategy, makingthe controller using online learning mechanism revise the system modelparameters in real time and fine tuning weight coefficient of each blade.The simulation results show that wind turbine can capture maximumenergy and improve power transient response well during switchingcontroller, proving the SVM weight coefficient control strategy has goodabilities of real-time and robustness.