原文信息：Data-driven distributionally robust scheduling of community integratedenergy systems with uncertain renewable generations considering integrateddemandresponse原文链接：https://www.sciencedirect.com/science/article/pii/S0306261923001137摘要区域综合能源系统（CIES）是能源互联网和智慧城市在地理和功能方面的重要载体。它的出现为能源利用和环境污染问题提供了新的解决方案。为了协调可再生能源发电（RGs）的综合需求响应和不确定性，构建了一个数据驱动的两阶段分布式鲁棒优化（DRO）模型。使用由1-范数和∞-范数组成的综合范数作为不确定性概率分布信息集，从而避免了复杂的概率密度信息。为了解决RGs的多个不确定性，提出了一种基于带有梯度惩罚的Wasserstein距离的生成对抗网络来生成RG场景，该网络具有广泛的适用性。为了进一步挖掘需求响应的潜力，作者考虑了人体热舒适性和建筑热惯性的模糊性，制定了一个有效促进可再生能源消纳的综合需求响应机制。该方法在华北地区的实际CIES中进行了计算。与传统的随机规划和鲁棒优化相比，所提出的DRO模型在表现出更强的适应性的同时，适当地平衡了经济运行和鲁棒性之间的关系。此外，所提方法优于其他常用的DRO方法，具有更好的运行经济性、更低的可再生能源削减率和更高的计算效率。更多关于"Integrateddemandresponse"的研究请见：https://www.sciencedirect.com/search?qs=Integrated%20demand%20response&pub;=Applied%20Energy&cid;=271429AbstractAcommunity integrated energy system (CIES) is an important carrier of theenergy internet and smart city in geographical and functional terms. Itsemergence provides a new solution to the problems of energy utilization andenvironmental pollution. To coordinate the integrated demand response anduncertainty of renewable energy generation (RGs), a data-driven two-stagedistributionally robust optimization (DRO) model is constructed. Acomprehensive norm consisting of the 1-norm and ∞-norm is used as theuncertainty probability distribution information set, thereby avoiding complexprobability density information. To address multiple uncertainties of RGs, agenerative adversarial network based on the Wasserstein distance with gradientpenalty is proposed to generate RG scenarios, which has wide applicability. Tofurther tap the potential of the demand response, we take into account theambiguity of human thermal comfort and the thermal inertia of buildings. Thus,an integrated demand response mechanism is developed that effectively promotesthe consumption of renewable energy. The proposed method is simulated in anactual CIES in North China. In comparison with traditional stochasticprogramming and robust optimization, it is verified that the proposed DROmodel properly balances the relationship between economical operation androbustness while exhibiting stronger adaptability. Furthermore, our approachoutperforms other commonly used DRO methods with better operational economy,lower renewable power curtailment rate, and higher computationalefficiency.KeywordsCommunity integrated energy systemDistributionally robustoptimizationUncertainty modelingIntegrated demand responseRenewableenergyScenario generationGraphicsFig. 1. Basic structure of a GAN.Fig. 2.Schematic diagram of the proposed CIES.Fig. 4. CCG solving process.Fig. 8.Results of CIES electrical scheduling in typical scenarios.Fig. 10. Results ofIDR scheduling in typical RG output scenarios.关于Applied Energy本期小编：王琼审核人：赵林川《Applied Energy》是世界能源领域著名学术期刊，在全球出版巨头爱思唯尔 (Elsevier)旗下，1975年创刊，影响因子11.446，CiteScore20.4，高被引论文ESI全球工程期刊排名第4，谷歌学术全球学术期刊第50，本刊旨在为清洁能源转换技术、能源过程和系统优化、能源效率、智慧能源、环境污染物及温室气体减排、能源与其他学科交叉融合、以及能源可持续发展等领域提供交流分享和合作的平台。开源（OpenAccess）姊妹新刊《Advances in Applied Energy》现已正式上线。在《AppliedEnergy》的成功经验基础上，致力于发表应用能源领域顶尖科研成果，并为广大科研人员提供一个快速权威的学术交流和发表平台，欢迎关注！公众号团队小编招募长期开放，欢迎发送自我简介（含教育背景、研究方向等内容）至wechat@applied-energy.org点击“阅读原文”喜欢我们的内容？点个“赞”或者“再看”支持下吧！