| 引用本文: | 周海强,赵春祝,顾婷艳,薛峰,高超,宋晓芳.基于高斯混合模型及近似线性规划的风电系统校正控制方法[J].电力自动化设备,2022,42(12): |
| ZHOU Haiqiang,ZHAO Chunzhu,GU Tingyan,XUE Feng,GAO Chao,SONG Xiaofang.Corrective control method of wind power system based on Gaussian mixture model and approximate linear programming[J].Electric Power Automation Equipment,2022,42(12): |
|
| 摘要: |
| 提出一种基于高斯混合模型及近似线性规划的风电系统校正控制方法。考虑到风电及负荷功率预测误差的非高斯特性及相关性，建立了描述风电及负荷功率联合概率分布的高斯混合模型。考虑切负荷、发电机功率调整以及统一潮流控制器等控制措施，以控制代价最小为目标函数，构建了计及电压、频率及线路潮流等静态安全约束的风电系统校正控制优化问题的数学模型。为求解含不确定变量的非线性优化问题，依据高斯混合模型的性质，在稳态运行点的邻域内将非线性约束近似转化为线性约束，用线性规划方法求解，并不断迭代直至收敛。最后，将该方法应用于修改后的IEEE 10机39节点算例系统验证其有效性，与场景法的对比结果表明该方法求解效率较高，且具有较好的经济性。 |
| 关键词: 风电 校正控制 高斯混合模型 灵敏度 线性规划 静态安全 |
| DOI：10.16081/j.epae.202205066 |
| 分类号:TM614 |
| 基金项目:国家自然科学基金委员会-国家电网公司智能电网联合基金资助项目(U2066601)；智能电网保护和运行控制国家重点实验室资助项目(SGTYHT/20?JS?221) |
|
| Corrective control method of wind power system based on Gaussian mixture model and approximate linear programming |
|
ZHOU Haiqiang1, ZHAO Chunzhu1, GU Tingyan1, XUE Feng2,3, GAO Chao1, SONG Xiaofang2,3
|
|
1.College of Energy and Electrical Engineering, Hohai University, Nanjing 211100, China;2.NARI Technology Co.,Ltd.,Nanjing 211106, China;3.State Key Laboratory of Smart Grid Protection and Control, Nanjing 211106, China
|
| Abstract: |
| A corrective control method of wind power system based on Gaussian mixture model and approximate linear programming is proposed. Considering the non-Gaussian characteristics and correlations between wind power and load power prediction errors, the Gaussian mixture model is built to describe the joint probability distribution of wind power and load power. Then, considering the control measures such as load shedding, generator power adjustment and the unified power flow controller, the minimum control cost is taken as the objective function, and the mathematic model of the corrective control optimization problem with the static security constraints of voltage, frequency and line power flow is established for the wind power system. To solve the nonlinear optimization problem with uncertain variables, the nonlinear constraints are approximately transformed into linear constraints according to the properties of Gaussian mixture model in the neighborhood of steady-state operation points. And the problem is solved by linear programming method. Iterations are repeated until reaching a convergence. Finally, the method is applied in the modified IEEE 10-machnie 39-bus system to validate its effectiveness. Results show that it has higher computation efficiency and better economic compared to the scenario-based method. |
| Key words: wind power corrective control Gaussian mixture model sensitivity linear programming static security |
