电力系统运行与规划中需重点关注到楼宇空调负荷的不确定性,可将楼宇空调负荷变化的不确性场景转化为多个确定性场景的场景生成问题。提出了楼宇空调负荷场景生成问题的基本分析框架,深入分析了楼宇空调负荷的用能特征,挖掘了楼宇空调...电力系统运行与规划中需重点关注到楼宇空调负荷的不确定性,可将楼宇空调负荷变化的不确性场景转化为多个确定性场景的场景生成问题。提出了楼宇空调负荷场景生成问题的基本分析框架,深入分析了楼宇空调负荷的用能特征,挖掘了楼宇空调负荷用能时序序列数据所蕴含的动静态特征。将楼宇空调负荷数据的动静态特征作为条件监督项,将无监督对抗训练与监督训练相结合,设计了联合训练损失函数与全局优化损失函数,并在此基础上提出了一种基于条件时序生成对抗网络(time series generative adversarial nets,TimeGAN)的楼宇空调负荷场景生成方法。最后,通过算例验证了所提方法的可行性与有效性。研究成果对提高楼宇空调负荷主动参与电力系统的运行规划有积极的意义。展开更多
Adopting an elastic-viscoplastic, the asymptotic problem of mode I propagat ing crack-tip field is investigated. Various asymptotic solutions resulting from the analysis of crack growing programs are presented. The an...Adopting an elastic-viscoplastic, the asymptotic problem of mode I propagat ing crack-tip field is investigated. Various asymptotic solutions resulting from the analysis of crack growing programs are presented. The analysis results show that the quasi-statically growing crack solutions are the special case of the dynamic propagating solutions. Therefore these two asymptotic solutions can be unified.展开更多
文摘电力系统运行与规划中需重点关注到楼宇空调负荷的不确定性,可将楼宇空调负荷变化的不确性场景转化为多个确定性场景的场景生成问题。提出了楼宇空调负荷场景生成问题的基本分析框架,深入分析了楼宇空调负荷的用能特征,挖掘了楼宇空调负荷用能时序序列数据所蕴含的动静态特征。将楼宇空调负荷数据的动静态特征作为条件监督项,将无监督对抗训练与监督训练相结合,设计了联合训练损失函数与全局优化损失函数,并在此基础上提出了一种基于条件时序生成对抗网络(time series generative adversarial nets,TimeGAN)的楼宇空调负荷场景生成方法。最后,通过算例验证了所提方法的可行性与有效性。研究成果对提高楼宇空调负荷主动参与电力系统的运行规划有积极的意义。
文摘Adopting an elastic-viscoplastic, the asymptotic problem of mode I propagat ing crack-tip field is investigated. Various asymptotic solutions resulting from the analysis of crack growing programs are presented. The analysis results show that the quasi-statically growing crack solutions are the special case of the dynamic propagating solutions. Therefore these two asymptotic solutions can be unified.