(电网技术)基于马尔科夫量的光伏发电量预测.pdfVIP

第35 卷 第1 期 电 网 技 术 Vol. 35 No. 1 2011 年1 月 Power System Technology Jan. 2011 文章编号：1000-3673 （2011 ）01-0152-06 中图分类号：TM 615 文献标志码：A 学科代码：470·4051 基于马尔可夫链的光伏发电系统 输出功率短期预测方法 丁明，徐宁舟 （教育部光伏系统工程研究中心(合肥工业大学) ，安徽省 合肥市 230009 ） A Method to Forecast Short-Term Output Power of Photovoltaic Generation System Based on Markov Chain DING Ming, XU Ningzhou (Research Center for Photovoltaic System Engineering (Hefei University of Technology), Ministry of Education, Hefei 230009, Anhui Province, China) ABSTRACT: A method to forecast output power of 输出功率的波动性对电力系统运行与控制的影响 photovoltaic (PV) plant is proposed. Based on Markov chain 日益引起关注，如何更加准确地预测光伏电站输 utilizing the statistics of historical data of PV plant, a model of 出功率，进而采取有效的应对措施是学术界和工 PV plant is built and the output power of the PV plant can be 程界关注的焦点[1] 。对预测光伏电站输出功率的研 directly forecasted. The feasibility of the proposed 究和应用可归纳为 2 类：一类方法是基于太阳能 mathematical model is verified by theoretical derivation. Taking the PV plant in PV Centre of the Ministry of Education 辐射强度的预测模型。首先建立太阳辐射模型， (MOE) of the People’s Republic of China for example, the 根据当地的气候历史数据得到太阳辐射的预测 modeling of the PV plant is performed and the output power 值，然后再对逆变过程进行建模，最终得到光伏 forecasting proves the effectiveness of the proposed method. 系统输出功率预测值，如文献[2-6]。这类方法依 More accurate forecasting results are obtained by adjusting the 赖于详细的气象数据，要求预测结果越精确，模 parameters of the built model. 型就越复杂，所需的历史气象数据量和数据类型 KEY WORDS: photovoltai