基于幂律过程的风力机故障建模方法研究

针对传统风力机故障建模方法存在的适用条件不符、误差较大等问题,提出一种基于幂律过程的风力机故障建模方法.首先对风力机故障数据进行趋势检验,结合故障数据的非齐次泊松特点,建立基于幂律过程的风力机故障模型;然后分别利用最小二乘法、极大似然估计法及列文伯格-马夸尔特算法求解模型参数,并比较不同参数估计方法的优越性;最后通过求...

RESEARCH ON FAULT MODELING METHOD OF WIND TURBINE BASED ON POWER-LAW PROCESS

Aiming at the problems of the traditional wind turbine fault modeling methods,such as inconsistent application conditions and large errors,a fault modeling method is proposed based on the power-law process for building fault modeling of wind turbines fault data.At first,the trend of the fault data was tested and combined with the non-homogeneous poisson process characteristics of fault data,a wind turbine fault model based on power-law process is established.Then,the least square(LS)method,maximum likelihood estimation(MLE)and levenberg-marquardt(LM)method are used to estimate the parameters,and the advantages of different parameter estimation methods are compared.At last,the goodness-of-fit is utilized to show the effectiveness of the present method.It shows that the goodness-of-fit indexes obtained by different parameter estimation methods are all close to 1,so the model can describe the fault process of wind turbine more accurately.Besides,the lewenberg-marquardt algorithm has higher precision in solving model parameters,which are 0.2%higher than that of maximum likelihood estimation method and 4.2%higher than that of least square method.