生成风电功率时间序列场景的双向优化技术

发布时间：2018-08-02 11:20

【摘要】：用少量的代表性风电功率序列场景来准确刻画风电随机特征,对含有风电电力系统的规划和运行具有重要意义。然而,代表性风电功率序列场景的生成,目前方法难以实现从庞大的发生空间中选择有效的代表场景,场景模拟的质量有待提高。为此,提出一种纵横双向优化的方法以生成日风电功率序列场景。纵轴方向,基于历史的日风电功率序列数据,采用最优消减技术,产生每个时段的代表场景;横轴方向,采用禁忌搜索方法,有选择地连接每个时段的代表场景从而形成所需的日风电功率序列代表场景。该方法无需预先知道风电功率的解析概率分布函数,仅需基于已有的历史序列数据,通过纵横双向优化,自动生成满足风电随机概率特征的日序列代表场景。以爱尔兰风电场数据为例,对所产生的单时段代表场景,在均值、方差、偏态和峰度4个指标上具有与历史数据相近的统计特性;将这些场景应用于含有风电电力系统的多时段最优潮流问题,从稳定性和准确性两个方面,验证了所提出的双向优化算法的有效性。
[Abstract]:It is of great significance to use a small number of representative wind power sequences to accurately depict the stochastic characteristics of wind power system. However, with the generation of representative wind power sequence scene, it is difficult to select an effective representative scene from the huge occurrence space, and the quality of scene simulation needs to be improved. For this reason, a method of vertical and horizontal optimization is proposed to generate the scene of daily wind power sequence. The longitudinal axis, based on the historical data of the solar wind power sequence, uses the optimal subtractive technique to produce the representative scene of each time period, and the horizontal axis direction uses the Tabu search method. The representative scene of each period is selectively connected to form the desired solar power sequence. This method does not need to know the analytic probability distribution function of wind power in advance, but based on the existing historical sequence data, it can automatically generate daily representative scenes satisfying the stochastic probability characteristics of wind power through longitudinal and horizontal bidirectional optimization. Taking the Irish wind farm data as an example, it has the statistical characteristics similar to the historical data on the mean, variance, skewness and kurtosis of the single time representative scene. These scenarios are applied to the multi-period optimal power flow problem with wind power system. The effectiveness of the proposed bi-directional optimization algorithm is verified in terms of stability and accuracy.
【作者单位】：强电磁工程与新技术国家重点实验室(华中科技大学);辽宁省电力公司;
【基金】：中国博士后基金项目(2012M511209) 中国博士后特别资助项目(2013T60717) 国家自然科学基金项目(51277034,51377027) 国家高技术研究发展计划(863计划)(2011AA05A112)~~
【分类号】：TM614

【参考文献】