利用智能停车场平抑风电预测误差的可行性研究  

Feasibility study of depressing wind power prediction error by Smart Park

作　　者：张广韬[1] 吴俊勇[1] 周彦衡[1] 苗青[1] 

机构地区：[1]北京交通大学电气学院,北京100044

出　　处：《电力系统保护与控制》2013年第17期88-94,共7页Power System Protection and Control

摘　　要：为了突破风电预测精度进一步提高的瓶颈,提出了一种利用智能停车场的集约化控制来平抑风电预测误差的方法。建立了智能停车场仿真模型,分析了充放电单元的端口功率特性,得到了电池储能与荷电状态之间的关系。基于一个实际风电场的出力及其预测数据,详细分析了智能停车场电动汽车规模、初始荷电状态分布、风电场装机容量和风电预测技术水平等因素与最终的风电预测误差平抑目标之间的关系。研究表明,将风电场与智能停车场联合协调运行,可以在很大程度上平抑风电预测误差,将不可调度的风电场变成可预测可调度的风电场,是一条经济可行的途径。For breaking the bottleneck that further improvement of the wind power prediction accuracy encounters, an idea of making use of the intensive control of the Smark Park to depress wind power prediction error is proposed. Firstly, a detailed Smark Park simulation model is established and the port power characteristics of charge/discharge unit are analyzed. The relationship between the battery storage energy and the state of charge （SOC） is obtained. Based on a real wind power output and its prediction data, this paper analyzes the relationship between the target of depressing wind power prediction error and the factors such as the size of Smart Park, initial state of charged distribution, wind power installed capacity and wind power forecasting technique level. The results show that the coordinated operation of wind farms and Smart Park can depress wind power prediction error to a great extent, and it is an economic and feasible way to turn the undispatchable wind farm into dispatchable energy sources.

关 键 词：电动汽车 荷电状态(SOC) V2G 智能停车场 风电预测 

分 类 号：TM614[电气工程—电力系统及自动化]