| 引用本文: | 许寅,李佳旭,王颖,李晨,和敬涵.考虑光伏出力不确定性的园区配电网日前运行计划[J].电力自动化设备,2020,40(5): |
| XU Yin,LI Jiaxu,WANG Ying,LI Chen,HE Jinghan.Day-ahead operation plan for campus distribution network considering uncertainty of photovoltaic output[J].Electric Power Automation Equipment,2020,40(5): |
|
| 摘要: |
| 随着分布式光伏大量接入配电网，其出力的不确定性成为影响配电网安全运行的重要因素，在制定日前运行计划时需要予以考虑。提出一种以最小化购电费用和电压偏移期望为目标的园区配电网日前运行机会约束规划模型。使用高斯混合模型描述光伏出力的不确定性，并采用样本均值近似方法将表述电压越限概率的机会约束转化为确定性约束，从而将机会约束规划模型转化为确定性的混合整数线性规划模型。通过求解该模型得到最优的配电网日前运行计划。以IEEE 13节点标准算例和北京某高校校园配电网为例，验证了所提模型和方法的有效性。 |
| 关键词: 配电网 日前运行计划 不确定性 高斯混合模型 样本均值近似 |
| DOI：10.16081/j.epae.202003004 |
| 分类号:TM615 |
| 基金项目:中央高校基本科研业务费专项资金资助项目(2018-JBZ004) |
|
| Day-ahead operation plan for campus distribution network considering uncertainty of photovoltaic output |
|
XU Yin, LI Jiaxu, WANG Ying, LI Chen, HE Jinghan
|
|
School of Electrical Engineering, Beijing Jiaotong University, Beijing 100044, China
|
| Abstract: |
| Along with the access of massive distributed photovoltaic into distribution network, the uncertainty of its output have become an important factor affecting the safe and operation of distribution network, which should be considered in making day-ahead operation plan. A chance-constrained programming model for day-ahead operation plan of campus distribution network is proposed, which takes the minimum expectation of power purchase cost and voltage offset as its objective. GMM(Gaussian Mixture Model) is used to describe the uncertainty of photovoltaic output, and SAA(Sample Average Approximation) method is adopted to transform the chance constraints which express the probability of voltage over-limit into deterministic constraints, then the chance-constraint programming model is transformed into a deterministic mixed-integer linear programming model. The optimal day-ahead operation plan of distribution network can be obtained by solving the model. The standard IEEE 13-bus system and the campus distribution network of a university in Beijing are taken as examples, and the effectiveness of the proposed model and method is verified. |
| Key words: distribution network day-ahead operation plan uncertainty Gaussian mixture model sample average approximation |
