Research On Turbulent Wind Field Simulation And Load Characteristics Of Wind Turbines

The external environment that large-scale wind turbine undergoes is complicated, establishing the reasonable and accurate turbulent wind field model, aerodynamic model and structure dynamic model of wind turbine can not only obtain accurate structure dynamic response characteristics of wind turbine, but also be the precondition for wind turbine designing. Rotational effects of the rotor should be taken into account for the modeling turbulent wind field and calculating aerodynamic loads since large-scale periodic rotation of the blades in wind fileds. Stochastic load fluctuations and response of wind turbine actuated by rotational turbulent wind velocity need to be analysed within the frequency domain. The research of dynamic responsed characteristics and load characteristics of wind turbines within complicated external environments and variable influencing factors plays a crucial role for improving sturcture of wind turbine, optimazing control strategy, and insuring operation under safe conditons. This task is financially supported by the National High-Tech Research and Development Program of China(863Program)"Research and development of collective control systems of wind farm and photo voltaic plants (Grant No.2011AA05A104)". The main works are as follows.(1)3-D turbulent wind field model is established, the improved Von Karman spectrum is deduced in detail, Calculating models of the tower shadow and wake effects which effect3-D turbulent wind field are summarized. The Blade Element Momentum (BEM) of wind turbine is introduced systematically, and the deficiencies are analysed. In order to considering the3-D effect of turbine rotor aerodynamics, the blade tip and hub loss corrections, the thrust coefficient correction, the angle of attack correction and wind shear velocity correction are introduced. Several structure modeling methods are summarized and advantages and disadvantages of three modeling methods are compared.(2) Accurate analytical expression of auto power spectral density(APSD) and cross power spectral density(CPSD) of the rotational Fourier spectrum are derived, the physical nature of rotational Fourier spectrum is analyzed, which is summation of numerous source spectra those have been multiplied by corresponding weighting coeffients after translation with integer multiple rotational frequencies, The weighting coefficient is the nth mode after translation, which is also Fourier expression coefficient of coherence function. The APSD of rotational Fourier spectrum contains both amplitude-frequency characteristics and phase-frequency characteristics. Rotational effects of rotor are mainly reflected by the coherence function and delay parameter, wherein the coherence function is the reflection of the changes in the amplitude spectrum, the delay parameter reflects the phase change. The selection of the coherence function is of vital influence to the rotational Fourier spectrum distribution. The revision of traditional Davenport’s exponential coherence function can eliminate the dips at integer multiple rotational frequencies of rotational Fourier spectrum. Wind shear can change the source value of spectrum and the selection rotating sampling points by changing the magnitude of the average wind velocities in the vertical plane of rotor.(3) Turbulent wind velocity time series of different points on the tower and rotor blades of wind turbine are simulated based on spectrum representation method, and spectrum representation formula which introduces phase lag factor is proposed. The simulated wind velocity time histories of multi-time scale rotational sampling points and still sampling points are compared. The fluctuating frequencis of wind velocity time histories at the calculated points of the rotating blades are significantly higher than those at the corresponding points of the stationary blades. The change laws of the calculated spectrum and the target spectrum are of high degree uniformity, which proved the theories and algorithms used in this paper can simulate the wind velocity time histories on the rotating rotor blades more accurately and efficiently.(4) The dynamic load spectrum actuated by turbulent wind is derived in the frequency domain and the physical nature of rotational modal is revealed. The spectrum of flapping motion amplitudes, load spectra of rotational blades as well as load spectra of flapping-blade rotors with single blade, two blades and three blades are obtained. When the system with Gaussian random turbulent wind input and response of wind turbine is a linear system, all of the statistical characteristics of the system can obtained by spectral analysis. Also fatigue loads can be estimated using spectral moments. It is most common to estimate fatigue damage rates by rainflow counting method.(5) The extreme loads and fatigue loads of wind turbine experienced all of external environment conditions and states of wind turbine are calculated, and dynamic response and fatigue load characteristics of wind turbine operated under important designed load cases are also analysed. The values of blade tip deflection, out of plane bending at blade root, thrust at hub center, tower top deflection and bending moments at tower flanges of wind turbine operated with rated wind velocity are larger than those with cut out wind velocity when it is operated with normal power produciton. It is inverse that the higher wind velocity result in greater corresponding dynamic response. The responses of components produce saltation and oscillation when wind turbine emergency stop with faults. Resonance phenomenon occures at situation of parked or idling, which results in resonant problem of wind turbine is one of main factors for turbine safe at high wind velocity. Equivalent fatigue loads of wind turbine can be calculated by rain flow counting method and the equivalent load method through transforming the random cyclic loads into equivalent fatigue loads, and then fatigue load spectra can be established, finally fatigue life of wind turbine is obtained.