引入模型定阶的ARMA模型在风力发电系统风速仿真中的应用

Application of ARMA Model and Identification of Model′s Order to Simulation of Wind

对于风力发电系统来说,研究的基础就是能否仿真出与自然界比较一致的风速序列。在已知VonKarman功率谱密度函数的情况下,选择利用自回归滑动平均(ARMA)模型来仿真风速。首先进行模型定阶工作,这样得到的ARMA模型阶数更为符合Von Karman谱。在得到了满足可逆性和稳定性条件的模型系数后,建立起短期风速模型。通过对ARMA模型仿真结果的分析,并与已有的风速模型进行比对,结果表明,该模型能够产生满足研究需要的风速序列。

For wind power generation systems,the basis of the study is whether the simulation of wind speed records is relatively close to the natural.Auto-regressive moving-average（ARMA） model is used to simulate the wind speed based on Von Karman power spectral density function.Firstly the ARMA model′s order is identified.Thereby the model will accord better with the Von Karman spectral.Then,the short-term wind speed model is established after getting coefficients which satisfy the reversibility and stability.The analysis of ARMA model simulation results and the contrast with the other model indicate that the ARMA model can satisfy the need of research in simulating wind speed records.