基于改进马尔科夫链的风电功率时间序列模型

Wind Power Time Series Model Based on Improved Markov Chain

模拟风电功率时间序列在风电并网系统的规划和评估研究中具有重要意义,针对原始马尔科夫链在风电功率建模上无法保留其自相关性的不足,构建了一种基于改进马尔科夫链的风电功率时间序列模型。首先分析了风电功率的季节特性、日特性和波动特性;然后将风电功率数据按照不同月份及时段进行了细致划分,生成相应的状态转移概率矩阵;最后,对风电功率波动量的概率分布进行拟合,并叠加波动量,建立了基于改进马尔科夫链的风电功率时间序列模型。实例分析表明,本文所建新模型生成的风电功率序列能够保留历史序列自相关性,同时在一般统计参数、概率密度分布和自相关性三方面的准确性也优于已有模型。

The simulation of wind power time series is of great significance in the planning and evaluation of wind power grid-connected systems. In order to solve the problem that the original Markov chain can not keep its autocorrelation in wind power modeling,this paper constructs a wind power time series model based on improved Markov chain. This paper firstly analyzes the seasonal characteristics,daily characteristics and fluctuation characteristics of wind power; and then subdivides the wind power data according to different months and time periods to generate the corresponding state transition probability matrix. Finally,this paper fits the probability distribution of wind power fluctuation and increase the amount of fluctuation to establish the wind power time series model based on improved Markov chain. The case analysis shows that the wind power series generated by the proposed model is superior to the existing model in the aspects of general statistical parameters,probability density distribution and autocorrelation,while preserving the historical sequence autocorrelation.