市场环境下基于随机规划的主动配电网运行交易二层优化模型  

A Bi-level Stochastic Programming Approach for Strategic Active Distribution Network Operation in Electricity Market

作　　者：蔡宇[1] 林今[1] 万灿[1] 宋永华[1] 

机构地区：[1]电力系统及发电设备控制和仿真国家重点实验室(清华大学电机系),北京市海淀区100084

出　　处：《中国电机工程学报》2016年第20期5391-5402,5715,共12页Proceedings of the CSEE

基　　金：国家863高技术基金项目(2014AA051901);国家自然科学基金项目(51577096;51207077;51261130472)~~

摘　　要：随着主动配电网技术和需求侧响应的大力发展,主动配电网能更主动地参与电力市场交易,影响输电网的运行。当主动配电网容量足够大时,其可作为大用户购电或供电商参与市场,通过有策略地竞价影响市场的出清结果。该文研究了主动配电网充分利用内部灵活可控负荷和分布式电源,作为市场主体有策略地参与日前市场,并基于随机规划建立了二层优化的竞价模型。其中上层模型以主动配电网收益最大为目标,提出了主动配电网安全运行模型,考虑其容量和电压安全约束;下层模型以社会效益最大为目标,考虑输电网运行及电力市场清算,并通过基于场景集的随机规划模型考虑了参与市场竞争者竞价的不确定性。上层模型通过市场竞价决策传递给下层模型影响日前市场出清,下层模型通过反馈市场出清价格给上层模型影响主动配电网内部运行决策。基于互补理论和线性化技术将二层规划模型最终转换为单层的混合整数线性规划模型进行求解。算例分析验证了文中所提模型的正确性和有效性。The rapid development of active distribution network （ADN） and demand response technologies are paving their way into today＇s electricity market, enabling the obvious effects of the ADN economic operation towards the main transmission power system. Acting as a consumer or a producer with large enough capacity, an active distribution company who operates the ADN has the ability to participate into the electricity market, strategically bidding with the purpose of influencing the market clearing prices to maximize its individual profits. This paper proposed a stochastic bi-level model that allowed the ADN strategically bid in electricity market with considerations of not only its own operations with flexible loads （FLs） and distribution generations （DGs）, but also the day-ahead market clearings. The upper level problem represents the ADN, i.e. the active distribution company, to maximize its individual profits, and the lower level problem represents the clearing of the market to maximize social welfare. In the upper level problem, we achieved the security operation of ADN by proposing an optimal security- constrained operation model. A stochastic programming （scenario-based） was introduced to solve the uncertainties pertaining to the behaviors of other rivals in the market. The day-ahead market clearing is manipulated in the lower level model via the bidding decisions from the upper level model, while the lower level model returns the market clearing price to the upper level model to influences the ADN operation decisions in the upper level model. Based on the complementarity theory, the proposed bi-level model is able to be transformed into a MILP problem which could be solved easily and more efficiently. Case studies demonstrate the accuracy and effectiveness of the proposed model.

关 键 词：主动配电网 电力市场 竞价策略 互补理论 灵活负荷 

分 类 号：TM732[电气工程—电力系统及自动化]