计及风电预测不确定性的电力系统日内调度模型研究

发布时间：2018-02-02 21:31

本文关键词： 风电不确定性模糊理论多场景优化调度　出处：《华北电力大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着化石燃料的日益枯竭和环境污染问题的日益突出,以风电为主的新能源发电得到大力发展。然而风电预测功率具有不确定性,大规模风电并网给电力系统日内调度及AGC调控带来巨大压力。如何在现有预测精度下实现功率预测结果的高效应用,将不确定性信息纳入到系统优化调度过程中,已成为风电并网优化调度的研究热点。本文在分析风电功率预测误差的基础上,建立计及风电预测不确定性的电力系统经济调度模型。首先基于模糊理论建立含风电电力系统动态模糊调度模型,为更好描述风电出力特性,选用高斯函数作为风电出力的隶属函数建模。采用最大化满意度指标法将双目标优化问题转化为等价清晰优化问题求解。结合算例采用MATLAB编程验证了模型的有效性,并对比了三角形、梯形、高斯型隶属函数对计算结果的影响,分析了采用高斯型隶属函数描述风电预测功率不确定性的优势；计算结果也表明双目标优化比单目标更合理。同时,本文还构建了基于多场景的概率优化调度模型。该模型把风电功率可能的波动范围离散成多场景,在此基础上,通过离散风功率预测误差分布曲线得到场景概率,并在时间尺度上考虑相邻时间段预测误差的关联性。通过在目标函数中引入概率调整成本,使优化结果计及了因风电波动性产生的成本。对含风电场的IEEE-30节点系统进行测试分析,论证了所提模型的合理性和有效性,并对具体参数进行了分析。最后,对上述两种建模思路从建模难易程度、求解难易程度、决策难易程度、实用性展望四个方面进行了对比分析。以期对含风电的电力系统经济调度提供参考。
[Abstract]:With the increasingly depletion of fossil fuels and the increasingly prominent environmental pollution problems, wind power mainly new energy generation has been vigorously developed. However, the forecast power of wind power is uncertain. Large-scale wind power grid connection brings great pressure to in-day dispatching and AGC regulation of power system. How to realize the efficient application of power prediction results under the existing prediction accuracy. The introduction of uncertainty information into the optimal scheduling process has become a hot topic in wind power grid optimization scheduling. This paper analyzes the prediction error of wind power. The economic dispatching model of power system considering the uncertainty of wind power forecasting is established. Firstly, based on the fuzzy theory, the dynamic fuzzy dispatching model of wind power system is established to describe the wind power output characteristics better. Gao Si function is selected as the membership function of wind power generation. The maximum satisfaction index method is used to transform the two-objective optimization problem into an equivalent and clear optimization problem. The MATLAB programming is used to verify the proposed method. The validity of the model. The influence of triangular, trapezoidal and Gao Si membership functions on the calculation results is compared. The results also show that the double objective optimization is more reasonable than the single objective. At the same time, the probabilistic optimal scheduling model based on multi-scene is constructed, which discretizes the possible fluctuation range of wind power into multi-scene, and on this basis. The scene probability is obtained by the discrete wind power prediction error distribution curve, and the correlation of prediction errors in adjacent time periods is considered on the time scale. The probability adjustment cost is introduced into the objective function. The optimization results take into account the cost caused by wind power fluctuation. The IEEE-30 node system with wind farm is tested and analyzed, and the rationality and validity of the proposed model are demonstrated. The specific parameters are analyzed. Finally, the above two methods of modeling from the degree of difficulty in modeling, easy to solve, difficult to make decisions. Four aspects of practical prospect are compared and analyzed in order to provide a reference for the economic dispatch of power system with wind power.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM73

【参考文献】