<style type="text/css">.wpb_animate_when_almost_visible{opacity:1}</style>

Session 10: Key Technologies for Multi-Agent Synergy-Driven Global Demand Response in Integrated Energy Systems

“多智能体协同驱动综合能源全域需求响应关键技术”

Session 10

Key Technologies for Multi-Agent Synergy-Driven Global Demand Response in Integrated Energy Systems
“多智能体协同驱动综合能源全域需求响应关键技术”

In integrated energy systems, distributed multi-energy coordinated stations, large-scale electric vehicle charging networks, and intelligent computing data centers are three types of highly flexible emerging loads that are achieving rapid large-scale deployment. The traditional centralized demand response mode, which relies on human experience and top-down unified instructions, can no longer adapt to the distributed collaboration requirements of massive heterogeneous resources. Multi-agent synergy technology, with large language models as its cognitive core, is bringing a brand-new technical paradigm to global load management and demand response systems: by constructing dedicated agents with independent perception, reasoning and decision-making capabilities for different types of load entities, it can achieve second-level collaborative scheduling of cross-scenario distributed resources while strictly safeguarding the data security and independent operation rights of each stakeholder, thereby substantially improving the response speed, regulation accuracy and overall resource utilization efficiency of demand response.
This special session closely follows the core technical line of artificial intelligence and multi-agent synergy for multi-energy coordination, vehicle–grid synergy, and computing power–grid integration, and systematically integrates the global demand response requirements of these three core scenarios into a unified technical framework. It focuses on common core issues such as accurate calculation and analysis of multi-energy flows, whole-chain collaborative optimization of computing power, electricity and carbon, role division of heterogeneous agents, low-latency collaborative communication mechanisms, and distributed optimization decision-making. It gathers core university research teams together with frontline technical personnel from power grid dispatching institutions, integrated energy service providers, computing power operators and new energy vehicle enterprises, to share cutting-edge research results with substantive original contributions, verified pilot implementation experience and replicable commercial operation models, so as to promote the rapid maturity and large-scale deployment of the new-generation global demand response technology system.

综合能源系统中,多能互补分布式站点、大规模电动汽车充电网络、智算数据中心三类高灵活性新兴负荷正快速实现规模化部署,传统依赖人工经验、自上而下统一指令下发的集中式需求响应模式,已无法适配海量异质资源的分布式协同运行要求。以大语言模型为认知中枢的多智能体协同技术,正在为全域负荷管理与需求响应体系带来全新的技术范式:通过为不同类型的负荷主体构建具备独立感知、推理、决策能力的专属智能体,可在严格保障各主体数据安全与自主运营权益的前提下,实现跨场景分散资源的秒级协同调度,大幅提升需求响应的响应速度、调节精度与全系统资源利用效率。
本专题紧扣人工智能与多智能体协同的核心技术主线,将多能互补、车网协同、算电融合三大核心场景的全域需求响应需求统一纳入技术框架下展开系统性研讨,聚焦多能流精准计算分析、算电碳全链条协同优化、异质智能体角色划分、低时延协同通信机制、分布式优化决策等共性核心问题,定向汇聚高校核心科研团队与电网调度机构、综合能源服务商、算力运营商、新能源车企的一线技术人员,共同分享具备实质性原创贡献的前沿研究成果、已完成验证的试点落地经验与可复制推广的商业化运营模式,推动新一代全域需求响应技术体系的快速成熟与规模化落地。

Topics (Including but not limited to)

Multi-Agent Demand Response Technologies in Multi-Energy Complementarity Scenarios
多能互补场景下的多智能体需求响应技术

  • Dedicated agent modeling and response capability self-learning methods for the multiple flexible cooling, heating and power loads of integrated energy stations
    面向综合能源站冷热电多柔性负荷的专属智能体建模与响应能力自学习方法
  • Real-time optimal scheduling algorithms for multi-level demand response based on multi-agent distributed collaboration in complex multi-energy flow scenarios
    多能流复杂场景下多智能体分布式协同的多层级需求响应实时优化调度算法
  • Dynamic linkage and collaboration mechanisms between multi-energy complementary agents and generation-side renewable energy output forecasting agents
    多能互补智能体与源侧新能源出力预测智能体的动态联动协同机制

Multi-Agent Demand Response Technologies in Vehicle-Grid Synergy Scenarios
车网协同场景下的多智能体需求响应技术

  • Distributed demand response aggregation methods for massive electric vehicle user-side agents and regional charging platform agents
    海量电动汽车用户侧智能体与区域充电平台智能体的分布式需求响应聚合方法
  • Multi-agent second-level precise demand response control and dynamic path planning technologies for ordered charging and V2G scenarios
    面向有序充电、V2G场景的多智能体秒级需求响应精准控制与动态路径规划技术
  • Market game and optimized bidding strategies for urban vehicle-grid interaction cluster agents participating in grid peak regulation and renewable energy consumption
    城市车网互动集群智能体参与电网尖峰调节、新能源消纳的市场博弈与优化报价策略

Multi-Agent Demand Response Technologies in Computing Power–Grid Integration Scenarios
算电融合场景下的多智能体需求响应技术

  • Autonomous identification and flexible regulation technologies for the interruptible and shiftable load demand response potential of data center computing power cluster agents
    数据中心算力集群智能体的可中断、可平移负荷需求响应潜力自主识别与柔性调节技术
  • Two-way collaborative demand response architecture and algorithm implementation between intelligent computing center computing power scheduling agents and power grid dispatching agents
    智算中心算力调度智能体与电网调度智能体的双向协同需求响应架构与算法实现
  • Linked operation optimization strategies integrating computing power demand response, green power consumption and carbon asset value increment under multi-agent collaboration
    多智能体协同下算电需求响应、绿电消纳与碳资产增值的联动运营优化策略

General Supporting Technologies for Global Multi-Agent Demand Response
全域多智能体需求响应共性支撑技术

  • Large language model driven causal reasoning and fault-tolerant decision-making technologies for multi-agent demand response under extreme abnormal operating conditions
    大语言模型驱动的多智能体需求响应因果推理与极端异常工况下容错决策技术
  • Blockchain-enabled trusted settlement and automatic benefit distribution mechanisms for distributed multi-agent global demand response
    基于区块链的全域多智能体分布式需求响应可信结算与权益自动分配机制
  • Federated learning empowered cross-stakeholder data privacy protection and distributed joint optimization technologies for multi-agent demand response models
    联邦学习赋能的多智能体跨主体数据隐私保护与需求响应模型分布式联合优化技术

Chair: Dr. Haoran Li, Shandong University of Finance and Economics, China

Haoran Li is an Associate Professor at Shandong University of Finance and Economics and the Shandong Provincial Key Laboratory of Blockchain Finance. After obtaining his Ph.D. in Control Theory and Control Engineering from Shandong University in 2022, he completed his postdoctoral research at the School of Electrical Engineering in 2024. His research interest is the energy system of smart cities, specifically encompassing urban energy system analysis and optimization, the transformation of energy data into factors of production, carbon emission accounting and trajectory analysis, as well as energy market economic policies.

 

Co-chair: Dr. Changlong Li, Shandong University, China

Changlong Li received the Ph.D. degree in electrical engineering from Shandong University, Jinan, China, in 2022. He is currently an Associate Professor with the School of Control Science and Engineering, Shandong University. His research interests include modeling and management of lithium-ion batteries in electric vehicles and energy storage systems.

 

Co-chair: Dr. Hang Tian, University of Jinan, China

Hang Tian is currently a Lecturer at the School of Electrical Engineering, University of Jinan, China. He received his M.E. degree in Electrical Engineering from the University of New South Wales, Australia, in 2017, and the Ph.D. degree in Electrical Engineering from Shandong University, China, in 2024. His research interests include simulation, optimization, and uncertainty analysis of power systems and integrated energy systems.

 

Co-chair: Dr. Lizhi Zhang, Shandong Police College, China

Lizhi Zhang received the B.S. degree in automation from Qingdao University, Qingdao, China, in 2016, and the M.S. degree in control engineering and the Ph.D. degree in control theory and control engineering from Shandong University, Jinan, China, in 2019 and 2024, respectively. He is currently with the Shandong Police College, Jinan, China. His research interests include optimal planning and security defense of multi-agent energy systems.

Call for Papers Timeline / 征稿时间

  • Submission of Full Paper: November 30th, 2026
    投稿截止日: 2026年11月30日 

  • Notification Deadline: December 30th, 2026
    通知书发送: 2026年12月30日 

  • Registration Deadline: January 20th, 2027
    注册截止日期: 2027年1月20日