电工技术学报2024，Vol.39Issue(9)：2652-2666,15.DOI:10.19595/j.cnki.1000-6753.tces.231369
基于随机-Nash-Harsanyi讨价还价博弈的光伏-氢储能多主体分散协同调度
Photovoltaic-Hydrogen Energy Storage Multi-Agent Decentralized Cooperative Scheduling Based on Stochastic-Nash-Harsanyi Bargaining Game
摘要
Abstract
Integrating photovoltaic(PV)and hydrogen energy storage(HES)into a PV-HES joint system proves to be an effective strategy in boosting local consumption of renewable energy and facilitating the low-carbon transformation of energy.The system's internal photovoltaic output and load exhibit considerable randomness,influencing the system's stable operation.Additionally,PV and HES,being typically associated with different stakeholders,each prioritize maximizing their own interests,which limits the effective operation of the PV-HES joint system.Current research often overlooks the bargaining power of diverse entities within the system,failing to fully portray each entity's actual contributions in cooperative efforts amidst uncertainties.In response to these problems,the relationships between entities in the PV-HES combined system should be coordinated and multiple uncertainties should be considered to ensure the stable operation of the PV-HES combined system. Initially,utilizing Nash-Harsanyi bargaining game theory,the study formulates a multi-agent cooperative operational model for PV-HES.This model equivalently addresses the determination of power transaction volume and power transaction cost as distinct subproblems.The Nash-Harsanyi bargaining game extends the Nash bargaining game into an asymmetric game model.This model accounts for participants'bargaining power and their contributions in cooperative efforts,ensuring a fair distribution of benefits.Secondly,stochastic optimal operating models for PV and HES systems were created using probabilistic scenarios,considering uncertainties in both energy supply and load aspects.These models aim to minimize cooperative operational costs.Thirdly,to safeguard the privacy of each participant,the improvement of alternating direction multiplier method(ADMM)is employed to address the aforementioned two subproblems.This method utilizes a two-stage dynamic step size correction approach. To validate the efficacy of the proposed decentralized collaborative scheduling model and distributed solving algorithm,simulations were performed on a standard PV-HES joint system."The simulation results indicate a reduction in operating costs for HES entities by 11.67%and 10.48%respectively,alongside a 15.72%increase in profit for PV aggregators.Furthermore,the improvement of ADMM method significantly boosts system efficiency,reducing solving time by 70%compared to the traditional ADMM algorithm.The calculation outcomes validate the model's advantages in reducing cooperative operating costs,enhancing solving efficiency,and fostering the utilization of renewable energy. The simulation analysis yields the following conclusions:(1)The proposed multi-agent cooperative operation model of PV-HES strikes a balance between individual rationality and alliance rationality.Unlike traditional cooperative approaches,this model comprehensively accounts for various uncertainties,including energy trading,entities'bargaining power,information privacy,and source load.(2)Employing the enhanced alternating direction multiplier method(ADMM)to resolve the electricity trading quantum problem and its associated cost exhibits favorable convergence characteristics.While safeguarding the confidentiality of individual data,this improved ADMM algorithm effectively addresses the multi-agent collaborative operation problem within PV-HES.关键词
光伏系统/氢储能/PV-HES联合系统/Nash-Harsanyi讨价还价/改进交替方向乘子法/分散协同调度Key words
PV system/Hydrogen energy storage/PV-HES joint system/Nash-Harsanyi bargaining game/improved alternating direction multiplier method/decentralized cooperative dispatching model分类
信息技术与安全科学引用本文复制引用
王学杰,赵会茹,曹艺琼,赵名锐..基于随机-Nash-Harsanyi讨价还价博弈的光伏-氢储能多主体分散协同调度[J].电工技术学报,2024,39(9):2652-2666,15.基金项目
国家自然科学基金(72101089)和中央高校基金(2022MS026)资助项目. （72101089）
