摘要
由于光伏出力和电力市场价格存在不确定性,现有研究常采用基于概率场景的随机优化方法,但准确的光伏发电和电力市场价格概率密度函数较难获取,且可能由于难以满足投标功率而受到惩罚。为此,提出了基于鲁棒模型预测控制的运行策略。以光储系统在能量和调频市场的联合收益最大为目标函数,计及光伏发电的出力约束和储能的充放电约束、功率约束、电量约束,考虑光伏出力和实时市场价格的不确定性,建立了多时步双层鲁棒优化模型。利用对偶理论将该模型转化为一个混合整数线性规划模型,并采用CPLEX求解该模型。最后,利用实际光伏和储能数据及ERCOT电价数据验证了所提运行策略的有效性。
Due to the uncertainty of the photovoltaic output and the electric market price, the existing study mainly employs the stochastic optimization method based on the probability scene. However, it is difficult to obtain accurate probability density function of photovoltaic output power and electric market price. The power producers will pay penalty due to the deviation of bidding capacity. Therefore, this paper proposes a robust model predictive control based operation strategy for photovoltaic and storage system to maximize its profile in the energy and regulation markets. The operation constraints of the photovoltaic and storage system are considered, including charge, discharge, power and capacity constraint of storage system, and output constraint of photovoltaic power. For the uncertainty of the photovoltaic output and the real-time market price, a multi-time step robust optimization model is established, which is transformed into a mixed integer linear programming model by dual theory. The transformed model can be solved efficiently by CPLEX, and the first time-step scheme in the optimization result is applied to the system. Finally, the effectiveness of the proposed method is verified by the actual photovoltaic and storage data and the ERCOT electricity price data.
作者
谢云云
谷志强
王晓丰
卓一
王珂
耿建
XIE Yunyun;GU Zhiqiang;WANG Xiaofeng;ZHUO Yi;WANG Ke;GENG Jian(School of Automation,Nanjing University of Science and Technology,Nanjing 210094,Jiangsu Province,China;Wenzhou Port Group Co.,Ltd.,Wenzhou 325000,Zhejiang Province,China;Huayun Information Technology Co.,Ltd.,Hangzhou 310008,Zhejiang Province,China;China Electric Power Research Institute,Nanjing 210003,Jiangsu Province,China)
出处
《电网技术》
EI
CSCD
北大核心
2020年第5期1758-1765,共8页
Power System Technology
基金
国家自然科学基金项目(51807181,61673213)
中央高校基本科研业务费专项资金项目(30918011330)
国家电网公司总部科技项目“虚拟电厂参与电力现货市场关键技术研究”。
关键词
光储系统
能量市场
调频市场
鲁棒模型预测控制
鲁棒优化
the photovoltaic and storage system
energy market
regulation market
robust model predictive control
robust optimization
作者简介
谢云云(1985),男,副教授,硕士生导师,研究方向为极端环境下电力系统运行与控制,E-mail:xyy_njust@163.com;谷志强(1995),男,硕士研究生,研究方向为电力市场,E-mail:543621979@qq.com;通信作者:王珂(1980),女,博士研究生,研究员级高级工程师,研究方向为源-网-荷互动运行调度与控制,E-mail:wangke@epri.sgcc.com.cn.com。
