基于信息差距决策理论的虚拟电厂报价策略OACSTPCD
Virtual Power Plant Quotation Strategy Based on Information Gap Decision Theory
为进一步提升分布式能源的调节潜力,基于信息差距决策理论,将探讨虚拟电厂(virtual power plant,VPP)在参与需求响应(demand response,DR)策略时的竞价方式分为平衡型、保守型和进取型 3 种策略模型,并为每种策略设计鲁棒函数和机会函数,分别实现对不同类型决策的优化.同时,设置ε约束模型,考虑了碳排放和利润的权衡关系.采用IEEE 18节点系统作为仿真环境,验证了所提方法的优点和必要性.仿真结果表明,保守型VPP能够保证在未来价格落入最大鲁棒性区间时获得最小关键利润;进取型VPP能够从意外的价格波动中获益,并实现期望的利润.
To further enhance the regulatory potential of distributed energy resource(DER),based on the information gap decision theory(IGDT),the bidding methods for virtual power plants(VPPs)participating in demand response(DR)strategies are divided into three strategy models:balanced,conservative and aggressive,and the robust and opportunity functions are designed for each strategy to optimize different types of decisions.Meanwhile,a ε-constraint model is set with consideration of the trade-off between carbon emissions and profits.The advantages and necessity of the proposed method were verified using an IEEE18 node system as the simulation environment.The simulation results show that the conservative VPP can ensure the minimum critical profit when the future price falls into the maximum robustness range;the progressive VPP can benefit from unexpected price fluctuations and achieve expected profits.
谢蒙飞;马高权;刘斌;潘振宁;商云峰
昆明电力交易中心有限责任公司,云南昆明 650011华南理工大学电力学院,广东广州 510641国家电投山东能源发展有限公司鲁东分公司,山东烟台 264000
虚拟电厂信息差距决策理论鲁棒性机会函数
virtual power plantinformation gap decision theoryrobustnessopportunity function deposition
《中国电力》 2024 (001)
40-50 / 11
国家自然科学基金资助项目(电力系统智能调度的高泛化性策略模型与元强化学习方法研究,52207105).This work is supported by National Natural Science Foundation of China(Research on High Generalization Strategy Model and Meta Reinforcement Learning Method for Intelligent Dispatching of Power Systems,No.52207105).
评论