CAJ | 학술논문

基于收缩跟踪区间的最大功率点跟踪控制能够改善湍流风速条件下大转动惯量风力机的风能捕获效率。但是，该方法仅依据平均风速优化设定收缩跟踪区间，忽略了湍流强度、风力机的某些气动、结构参数(如最佳叶尖速比、转动惯量等)等其它影响因素。考虑到跟踪区间优化设定与多种因素存在难以解析描述的复杂关系，提出了一种运用径向基函数神经网络优化跟踪区间的最大功率点跟踪控制。该改进方法以平均风速和湍流强度作为神经网络的输入变量，以具体风力机仿真数据作为训练样本，以补偿系数作为神经网络的输出变量。从而使得跟踪区间的优化设定不仅能够考虑变化的风速条件，而且能同时反映具体风力机的气动、结构设计。最后，对模拟风速序列进行了仿真计算与比较分析，验证了该方法的有效性和优越性。
기우수축근종구간적최대공솔점근종공제능구개선단류풍속조건하대전동관량풍력궤적풍능포획효솔。단시，해방법부의거평균풍속우화설정수축근종구간，홀략료단류강도、풍력궤적모사기동、결구삼수(여최가협첨속비、전동관량등)등기타영향인소。고필도근종구간우화설정여다충인소존재난이해석묘술적복잡관계，제출료일충운용경향기함수신경망락우화근종구간적최대공솔점근종공제。해개진방법이평균풍속화단류강도작위신경망락적수입변량，이구체풍력궤방진수거작위훈련양본，이보상계수작위신경망락적수출변량。종이사득근종구간적우화설정불부능구고필변화적풍속조건，이차능동시반영구체풍력궤적기동、결구설계。최후，대모의풍속서렬진행료방진계산여비교분석，험증료해방법적유효성화우월성。
The maximum power point tracking (MPPT) control based on reducing tracking range can improve the wind energy capture efficiency of wind turbines with high rotor inertia under turbulent conditions. However this method makes optimal setting of the reduction of tracking range only according to mean wind speed and other impacting factors such as turbulence intensity and several aerodynamic and structural parameters of wind turbine, for example the optimum tip speed ratio, rotational inertia and so on, are neglected. Considering that there exist complex relationships between the setting of tracking range and various factors, which are hard to be analytically described, an MPPT control method, in which the radial basis function neural network (RBFNN) is utilized to optimize the tracking range, is proposed. In the proposed improved control method the mean wind speed and turbulence intensity are taken as the input variable of neural network, and the simulated data of specific wind turbine is taken as training samples and the compensation coefficient as the output variable of neural network, thus an optimization setting of tracking range can be realized, in which both varying wind speed and the aerodynamic and structural parameters of wind turbine can be considered simultaneously. Finally, simulation calculation and comparative analysis on generated wind speed sequence are performed and the effectiveness and the advantages of the proposed control method are validated.