具有混沌特性的风电功率预测研究

Research on Wind Power Forecasting with Chaotic Characteristics

风能作为一种可再生的清洁能源,逐渐成为人类社会可持续发展的首选能源,风速的不确定性和随机性,使得风力发电具有不可调度性,因此准确的分析风电特性和预测风电功率对电力系统的发展至关重要.本文首先基于混沌理论对风电功率数据进行相空间重构,使用自相关法及CAO方法分别确定合适的时间延迟和恰当的嵌入维数,从而达到提高相空间重构质量及其预测精度的研究目的;其次依据指数判断重构的相空间是否具有混沌特性,若具有混沌特性,则建立混沌局域预测模型,对风电序列作混沌特性分析,最后对风电数据运用加权一阶局域预测模型进行短期预测.

As a kind of renewable clean energy, wind energy has gradually become the first choice energy for the sustainable development of human society. Wind power generation has a non-schedulable nature due to the uncertainty and randomicity of the wind speed. Hence, it is imperative to analyze the characteristics of wind pow- er and predict wind power for power system operators. Firstly, the phase space reconstruction of wind power data is carried out based on chaos theory, and the autocorrelation method and the CAO method are applied to deter- mine the appropriate time delay and embedding dimension, respectively, so as to improve the reconstruction quality of phase space and its prediction precision. Secondly, whether the reconstructed phase space has chaotic characteristics is determined according to the Lyapunov exponents. If it has chaos characteristic, then the chaotic local forecast model is established and the chaotic characteristics of the wind power sequence are analyzed. Final- ly, the wind power is forecasted by using the weighted first-order local prediction method.