One of the essential points of the direct-method single-wavelength anomalous diffraction （SAD） phasing for proteins is to express the bimodal SAD phase distribution by the sum of two Gaussian functions peaked respec...One of the essential points of the direct-method single-wavelength anomalous diffraction （SAD） phasing for proteins is to express the bimodal SAD phase distribution by the sum of two Gaussian functions peaked respectively at φh″＋｜△φh｜ and φh″-｜△φh｜. The probability for △φh being positive （P＋） can be derived based on the Cochran distribution in direct methods. Hence the SAD phase ambiguity can be resolved by multiplying the Gaussian function peaked at φh″＋｜△φh｜ with P＋ and multiplying the Gaussian function peaked at φh″-｜△φh｜ with P_ （=1- P＋）. The direct-method SAD h phasing has been proved powerful in breaking SAD phase ambiguities, in particular when anomalous-scattering signals are weak. However, the approximation of bimodal phase distributions by the sum of two Gaussian functions introduces considerable errors. In this paper we show that a much better approximation can be achieved by replacing the two Gaussian functions with two von Mises distributions. Test results showed that this leads to significant improvement on the efficiency of direct-method SAD-phasing.展开更多
为应对配电网络低碳化运行的挑战,并充分挖掘系统中分布式资源的灵活调控潜力,文中构建一种基于电碳综合边际定价的虚拟电厂(virtual power plant,VPP)双层点对点(peer-to-peer,P2P)交易模型。上层由配电网运营商(distribution system o...为应对配电网络低碳化运行的挑战,并充分挖掘系统中分布式资源的灵活调控潜力,文中构建一种基于电碳综合边际定价的虚拟电厂(virtual power plant,VPP)双层点对点(peer-to-peer,P2P)交易模型。上层由配电网运营商(distribution system operator,DSO)建立基于碳排放流(carbon emission flow,CEF)技术的碳感知最优潮流模型,在此基础上计算出电碳综合边际价格,DSO可利用该价格信号协调VPP低碳调度。下层组建多VPP合作联盟,可将电动汽车规模化整合并引入碳信号引导的电动汽车灵活调度机制,建立基于贡献度的非对称纳什议价交易模型,各VPP在价格信号的引导下平衡个体与联盟利益,制定生产与交易的最优策略。然后,采用自适应交替方向乘子法(adaptive-scaling alternating direction method of multipliers,AS-ADMM)对模型进行求解,解决变量耦合导致的收敛速度问题。最后,采用改进的IEEE 33节点配电系统进行仿真验证。案例分析结果表明,所提交易模型可以通过提高分布式能源利用效率并优化负荷分布,在降低VPP运营成本的同时减少配电网碳排放。展开更多
随着电动汽车和分布式电源接入电网的比例不断提升,虚拟电厂(virtual power plant,VPP)为有效解决电动汽车、分布式电源并网提供了新思路。针对VPP独立运行时面临的运行成本高、电价和源荷不确定性大等挑战,文中提出了一种基于纳什三阶...随着电动汽车和分布式电源接入电网的比例不断提升,虚拟电厂(virtual power plant,VPP)为有效解决电动汽车、分布式电源并网提供了新思路。针对VPP独立运行时面临的运行成本高、电价和源荷不确定性大等挑战,文中提出了一种基于纳什三阶段鲁棒优化的多VPP协同运行的优化方法。为协调VPP运营商与电动汽车用户的经济利益冲突,采用主从博弈理论刻画VPP运营商和电动汽车上下层之间的互动关系,上层VPP运营商充分考虑到电力市场购售电价以及源荷功率波动带来的不确定性影响,由三阶段鲁棒优化模型构造上层主体,三阶段鲁棒优化模型较以往的传统模型不同,文中采用了min-maxmin-maxmin的构造刻画模型内部关系;构建了基于纳什谈判理论的多VPP协同优化模型,为解决复杂非凸非线性优化的求解问题,将模型等效转化为多VPP合作成本最小化和电能谈判支付两个子问题;考虑到各VPP间信息隐私安全,采用交替方向乘子法(alternating direction method of multipliers,ADMM)对上述两个子问题进行分布式求解。算例验证表明,所提方法不仅在多重不确定性影响的情况下为参与合作的各VPP提供了可行且鲁棒性强的调度方案,而且为各VPP制定了合理的能源交互策略和利益分配方案,参与合作的各VPP均实现了经济效益的提升。展开更多
基金Project supported by the Innovation Foundation of the Chinese Academy of Sciences and by the National Basic Research Program of China(Grant No.2002CB713801)
文摘One of the essential points of the direct-method single-wavelength anomalous diffraction （SAD） phasing for proteins is to express the bimodal SAD phase distribution by the sum of two Gaussian functions peaked respectively at φh″＋｜△φh｜ and φh″-｜△φh｜. The probability for △φh being positive （P＋） can be derived based on the Cochran distribution in direct methods. Hence the SAD phase ambiguity can be resolved by multiplying the Gaussian function peaked at φh″＋｜△φh｜ with P＋ and multiplying the Gaussian function peaked at φh″-｜△φh｜ with P_ （=1- P＋）. The direct-method SAD h phasing has been proved powerful in breaking SAD phase ambiguities, in particular when anomalous-scattering signals are weak. However, the approximation of bimodal phase distributions by the sum of two Gaussian functions introduces considerable errors. In this paper we show that a much better approximation can be achieved by replacing the two Gaussian functions with two von Mises distributions. Test results showed that this leads to significant improvement on the efficiency of direct-method SAD-phasing.
文摘为应对配电网络低碳化运行的挑战,并充分挖掘系统中分布式资源的灵活调控潜力,文中构建一种基于电碳综合边际定价的虚拟电厂(virtual power plant,VPP)双层点对点(peer-to-peer,P2P)交易模型。上层由配电网运营商(distribution system operator,DSO)建立基于碳排放流(carbon emission flow,CEF)技术的碳感知最优潮流模型,在此基础上计算出电碳综合边际价格,DSO可利用该价格信号协调VPP低碳调度。下层组建多VPP合作联盟,可将电动汽车规模化整合并引入碳信号引导的电动汽车灵活调度机制,建立基于贡献度的非对称纳什议价交易模型,各VPP在价格信号的引导下平衡个体与联盟利益,制定生产与交易的最优策略。然后,采用自适应交替方向乘子法(adaptive-scaling alternating direction method of multipliers,AS-ADMM)对模型进行求解,解决变量耦合导致的收敛速度问题。最后,采用改进的IEEE 33节点配电系统进行仿真验证。案例分析结果表明,所提交易模型可以通过提高分布式能源利用效率并优化负荷分布,在降低VPP运营成本的同时减少配电网碳排放。
文摘随着电动汽车和分布式电源接入电网的比例不断提升,虚拟电厂(virtual power plant,VPP)为有效解决电动汽车、分布式电源并网提供了新思路。针对VPP独立运行时面临的运行成本高、电价和源荷不确定性大等挑战,文中提出了一种基于纳什三阶段鲁棒优化的多VPP协同运行的优化方法。为协调VPP运营商与电动汽车用户的经济利益冲突,采用主从博弈理论刻画VPP运营商和电动汽车上下层之间的互动关系,上层VPP运营商充分考虑到电力市场购售电价以及源荷功率波动带来的不确定性影响,由三阶段鲁棒优化模型构造上层主体,三阶段鲁棒优化模型较以往的传统模型不同,文中采用了min-maxmin-maxmin的构造刻画模型内部关系;构建了基于纳什谈判理论的多VPP协同优化模型,为解决复杂非凸非线性优化的求解问题,将模型等效转化为多VPP合作成本最小化和电能谈判支付两个子问题;考虑到各VPP间信息隐私安全,采用交替方向乘子法(alternating direction method of multipliers,ADMM)对上述两个子问题进行分布式求解。算例验证表明,所提方法不仅在多重不确定性影响的情况下为参与合作的各VPP提供了可行且鲁棒性强的调度方案,而且为各VPP制定了合理的能源交互策略和利益分配方案,参与合作的各VPP均实现了经济效益的提升。
