基于气象背景选取邻近点的风电功率爬坡事件预测方法  被引量:13

Prediction Method of Wind Power Ramp Events by Using Meteorological Conditions to Select Adjacent Points

在线阅读下载全文

作  者:欧阳庭辉 查晓明[1] 秦亮[1] 熊一[1] 夏添[1] 黄鹤鸣[1] 

机构地区:[1]武汉大学电气工程学院,湖北省武汉市430072

出  处:《电网技术》2015年第11期3266-3272,共7页Power System Technology

基  金:国家重点基础研究发展计划项目(973项目)(2012CB215101)~~

摘  要:考虑到风电功率爬坡事件在大规模、高集中度的风电发展模型下危害较大,为降低爬坡危害,维护电网的正常运行,提出了基于气象背景选取邻近点的爬坡事件多变量预测方法。首先,为提高风电功率预测的可靠性,充分考虑多气象因子的影响,通过对气象数据进行相空间重构建立了含多变量的基本预测模型。其次,考虑爬坡事件与气象背景间的联系,基于相关性分析确定不同气象背景下邻近点的选取准则。结合支持向量机分类模型,通过训练给出过去不同气象条件对应的邻近点选取机制,并指导选取预测所需的邻近点。最后,结合风电功率预测结果和爬坡事件检测方法,对实例数据进行爬坡仿真和预测,验证所提方法的有效性和可行性。With wind power developing in large-scale and high concentration in China, the harm brought by wind power ramp events is serious. In order to reduce this harm and keep safe operation of power system, this paper proposes a multi-variable prediction method of wind power ramps using meteorological condition to select adjacent points. The method makes use of meteorological data to improve performance of prediction. Firstly, it reconstructs phase space of meteorological factors and wind power series, and establishes basic multi-variable prediction model. Then, considering relationship between meteorological conditions and ramp events, the principle of selecting adjacent points under different meteorological conditions is given through correlation analysis. For purpose of guiding selection of adjacent points of future predicted points, this paper utilizes support vector machine(SVM) classification model to train adjacent point selection mechanism on different meteorological conditions. Finally, combining method of wind power prediction and ramp detection, simulation and prediction of wind power ramp events on example data are completed. Through assessment analysis of the results, the proposed method is verified to be feasible and effective.

关 键 词:爬坡事件 多变量预测模型 邻近点选取 支持向量机 

分 类 号:TM715[电气工程—电力系统及自动化]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象