芯片条状并行测试研究.pdf

中国科技论文在线 智能电网中基于吉布斯采样的联合用户调 度和需求响应控制算法# ** 杨怿，吴远，何艳飞，陈佳超 5 （浙江工业大学信息工程学院，杭州310023 ） 摘要：在智能电网中，采用需求响应能够有效地削减电力需求赤字、保持电网稳定工作。现 有关于需求响应的研究工作往往假设所有用户都将自愿参加需求响应。但是在实际过程中， 削减用户能耗和调度用户参加需求响应都将会对用户产生不利影响，因此，对于电力供电商 10 来说，在给定赤字的条件下，如何调度参与需求响应的用户，同时对调度的用户进行能耗削 减，仍然是一个开放的问题。为此，本文提出一种针对智能电网的联合用户调度与需求响应 控制模型，该模型的目标是在满足削减一定电力赤字的条件下，最小化包括用户调度代价和 因为削减能耗而产生的用户不满意度在内的系统总代价。针对这一模型，本文提出了一种带 有变温过程的吉布斯采样联合用户调度和需求响应控制算法，该算法能够计算出最佳用户调 15 度和与之相应的需求响应控制（能耗削减方案）。数值仿真结果表明该算法可获得与全局搜 索算法相近的性能，即满足削减一定电力赤字的要求下，保证最小化包括用户调度和不满意 度在内的系统总代价，同时该算法与全局搜索算法相比具有较低的复杂度。 关键词：智能电网；需求响应；用户调度；吉布斯采样 中图分类号：TP399 20 A Joint User-Scheduling and Demand Response Control Algorithm based on Gibbs Sampling in Smart Grid Yang Yi, Wu Yuan, He Yanfei, Chen Jiachao (College of Information Engineering, Zhejiang University of Technology, Hangzhou 310023) 25 Abstract: Demand response management is an efficient approach to reduce electricity deficit and guarantee the stability for the furture grid. Most of the current works on demand response assumed that all users were willing to join the process of demand response. However, from the practical point of view, not only shedding the users loads, but also scheduling them to join the demand response incur operational cost. Therefore, given the certain deficit, how does the supplier-side 30 schedule the appropriate set of users to join the demand response while reducing its electricity deficit is still an open question. To investigate this question, this paper proposed a joint user-