摘要
多能微网在运行过程中会受到电价以及新能源出力等多重不确定因素的影响。针对该问题,该文提出一种基于Wasserstein两阶段分布鲁棒优化的多主体多能微网合作博弈优化模型。首先,考虑多能微网和其内部产消者的互动关系,提出了“上层为多能微网,下层为产消者”的双层优化模型;其次,采用基于Wasserstein距离的模糊集分别构建了电价、多能微网新能源出力以及内部产消者光伏出力的不确定性模型;然后,在多主体多能微网之间,构建了考虑合作博弈和隐私保护的能源交互模型,并采用交替方向乘子法(ADMM)结合列与约束生成法(C&CG)对模型进行分布式求解;最后,基于包含三主体多能微网的系统进行算例分析,验证了该文所提模型和算法的有效性。
As the core of the energy system,multi-energy microgrid(MEMG)integrates distribution generation unit(such as wind turbine,photovoltaic),combing cooling,heat and power units,and energy storage to provide diverse energy supplies to customers for enhanced energy efficiency.However,the uncertainties in electricity prices and renewable generation are inevitably during the operation performance of MEMG.To tackle these uncertainties while enhancing economic benefits,this paper proposes a cooperative energy trading model for MEMGs based on Wasserstein two-stage distributionally robust optimization.First,to describe the interaction between MEMG and prosumer,a bilevel programming is established where the MEMG plays a leader and multiple prosumers are regarded as followers.Based on the electricity purchase power obtained from each prosumer,the MEMG optimizes the transactive price and send to followers.In each follower model,prosumer minimizes its operation cost under the given electric price which from MEMG.Meanwhile,the ambiguous set based on Wasserstein distance are utilized to capture the uncertainty of electricity price,renewable energy and photovoltaic in prosumer,respectively.The model of MEMG is formulated as a Wasserstein two stage distributionally optimization problem(WDRO)considering uncertain price and renewable energy,while the uncertainty of prosumers can be addressed by introducing the distributionally robust chance constraint(DRCC).Then various methods are employed to transform the WDRO and DRCC into finite linear terms,such as convex theory and physical support method,respectively.On this basis,the bilevel programming for“MEMG-prosumer”can be replaced by single-level model via KKT conditions,strong duality theory and big-M method.Additionally,an energy interaction model considering cooperative game and privacy protection has been developed among MEMGs,and the alternating direction multiplier method(ADMM)combined with column and constraint generation algorithm(C&CG)are used to solve the problem.Simulation results included three MEMGs verify the effectiveness of operation strategy proposed in this paper.The results indicate that:(1)The proposed bilevel programming can simultaneously provide optimal energy scheduling and price strategic for MEMG and prosumer.(2)The Wasserstein ambiguous set can effectively capture the price,renewable output and prosumer uncertainty,making it more comprehensive and less conservative compared to traditional robust optimization.(3)The effectiveness of the proposed energy trading model as well as the efficiency of the ADMM and C&CG algorithms are demonstrated by corresponding simulation results.
作者
王波
王蔚
马恒瑞
李天格
姚良忠
Wang Bo;Wang Wei;Ma Hengrui;Li Tiange;Yao Liangzhong(School of Electrical Engineering and Automation,Wuhan University,Wuhan,430072,China)
出处
《电工技术学报》
北大核心
2025年第17期5553-5570,共18页
Transactions of China Electrotechnical Society
基金
国家重点研发计划资助项目(2021YFB2401300)。
关键词
双层优化
两阶段分布鲁棒优化
多能微网
合作博弈
分布式算法
Bilevel programming
two stage distributionally robust optimization
multi-energy microgrid
cooperation game theory
distributed algorithm
