电力市场模拟中的不确定性机组自组合问题  

PROBABILISTIC SELF-UNIT COMMITMENT OF GENCO AGENT IN ELECTRICITY MARKET SIMULATION

作　　者：江健健[1] 康重庆[1] 夏清[1] 

机构地区：[1]清华大学电机工程与应用电子技术系,北京市海淀区100084

出　　处：《电网技术》2005年第15期34-39,60,共7页Power System Technology

基　　金：国家自然科学基金资助项目(50377016)~~

摘　　要：在电力市场环境下,发电商需要在不确定信息下,考虑机组的最小开、停机时间,确定各自机组期望的最优运行状态,并进而优化各自的报价。为了使电力市场模拟中的发电商决策模型更合理,作者基于报价中标概率函数,建立了考虑机组开、停机时间约束的报价决策模型,对机组自组合状态优化和报价决策行为进行研究和模拟。为了求解该不确定性混合整数问题,将其转化成为Markov过程,并提出相应算法。算例表明:考虑机组最小开、停机时间约束后,发电商会为了增加连续开机的可能性而降低在谷荷时段的报价,而这种报价策略会进一步加大市场中的峰谷电价差。该研究为电力市场模拟中发电商的决策提供了新的思路。In the electricity market, GenCos need to commit their own generation units and make bid decision with uncertain market information, in which the constraints of minimum start up/shut down times should be considered to determine their own expected optimal operation states and then to make bids more rational and feasible. Based on Bid Acceptance Probability （BAP） indicating the uncertain operation state in future, the authors establish the bid decision model, in which the constraints of minimum start up/shut down times are considered, to study and simulate the optimal self-unit commitment and bid decision problems of GenCos Agent. In order to solve the self-unit commitment problem, this uncertain and mixed integer optimal problem is transformed into a Markov chain and an efficient algorithm is proposed. Numeric results show that GenCos Agent might reduce the bidding prices in the load-valley hours to increase the possibility of non-stop operation of their generator units. And this bidding strategy will surely lower the electricity price in load-valley hours and increase the difference between the electricity price at load-peak hours and load-valley hours.

关 键 词：电力市场模拟 报价决策 机组自组合 MARKOV链 

分 类 号：TM73[电气工程—电力系统及自动化] F123.9[经济管理—世界经济]