考虑样本类别不平衡的电网故障事件智能识别方法

电网中不同设备的故障概率存在差异,影响智能诊断技术的准确性.为解决此问题,提出了一种基于代价敏感学习和模型自适应选择融合的电网故障事件智能识别方法.首先,利用Word2vec模型将预处理后的电网告警信息向量化,并搭建2个双向长短期记忆网络作为基础分类器;然后,设计代价敏感损失函数,将交叉熵损失函数与代价敏感损失函数分别应用于2个分类器中;最后,提出一种模型自适应选择融合法,融合上述分类器,得到故障事件识别结果.实际数据测试表明,所提方法能够有效降低故障事件识别中样本类别不平衡的影响.     

基于改进自注意力机制生成对抗网络的智能电网GPS欺骗攻击防御方法

为了避免全球定位系统欺骗攻击(GSA)对相量测量装置造成的危害,提出了一种基于改进自注意力机制生成对抗网络(SAGAN)的智能电网GSA防御方法.首先,通过引入深度学习参数,构建了改进网络-物理模型,利用历史数据计算得到当前时刻的量测值.然后,在SAGAN的生成器和判别器网络中分别融入一个时间注意力模块,提出了一种用于实现网络-物理模型的改进SAGAN防御方法.通过训练改进SAGAN,得到一对判别器和生成器,利用判别器检测采集的量测值是否遭受GSA,当检测到攻击时,利用生成器生成的数据替换欺骗数据,从而实现智能电网对GSA的主动防御.最后,基于IEEE 14节点和IEEE 118节点系统进行仿真测试,结果验证了所提方法的可行性和有效性.     

冲压成形仿真过程中有限元网格模型的建立

网格模型对冲压成形模拟的精度和效率影响极大 ,文中阐述了有限元网格模型建立的方法 ,并从单元尺寸、单元类型、自适应网格再划分技术等方面论述了如何解决精度与效率的问题。     

Dynamic replica placement and selection strategies in data grids— A comprehensive survey

Data replication techniques are used in data grid to reduce makespan, storage consumption, access latency and network bandwidth. Data replication enhances data availability and thereby increases the system reliability. There are two steps involved in data replication, namely, replica placement and replica selection. Replica placement involves identifying the best possible node to duplicate data based on network latency and user request. Replica selection involves selecting the best replica location to access the data for job execution in the data grid. Various replica placement and selection algorithms are available in the literature. These algorithms measure and analyze different parameters such as bandwidth consumption, access cost, scalability, execution time, storage consumption and makespan. In this paper, various replica placement and selection strategies along with their merits and demerits are discussed. This paper also analyses the performance of various strategies with respect to the parameters mentioned above. In particular, this paper focuses on the dynamic replica placement and selection strategies in the data grid environment.     

Smart pricing scheme: A multi-layered scoring rule application

Defining appropriate pricing strategy for smart environment is important and complex task at the same time. It holds the primal fraction in Demand Response (DR) program. In our work, we devise an incentive based smart dynamic pricing scheme for consumers facilitating a multi-layered scoring rule. The proposed strategy characterizes both incentive based DR and price based DR programs facilities. This mechanism is applied between consumer agents (CA) to electricity provider agent (EP) and EP to Generation Company (GENCO). Based on the Continuous Ranked Probability Score (CRPS), a hierarchical scoring system is formed among these entities, CA–EP–GENCO. As CA receives the dynamic day-ahead pricing signal from EP, it will schedule the household appliances to lower price period and report the prediction in a form of a probability distribution function to EP. EP, in similar way reports the aggregated demand prediction to GENCO. Finally, GENCO computes the base discount after running a cost-optimization problem. GENCO will reward EP with a fraction of discount based on their prediction accuracy. EP will do the same to CA based on how truthful they were reporting their intentions on device scheduling. The method is tested on real data provided by Ontario Power Company and we show that this scheme is capable to reduce energy consumption and consumers’ payment.     

一种电网综合展现可视化系统的设计与应用

笔者在分析一种电网信息综合展现可视化系统建设目标和设计思路的基础上,详细阐述该系统的总体架构、业务架构、数据架构、技术架构、部署架构设计。该系统由智能信息集成模块、视频监控模块和综合展现可视化模块三部分组成,基于先进的多层架构体系和SOA构架模型,以ETL、数据挖掘、视频、GIS、图形图像技术为依托,实现电网信息视频、温湿度、风速、雨量气象环境、负荷、潮流等电网运行信息以及安防消防等各类辅助决策信息有逻辑的、互联互通的、交互的和实时的可视化展现,满足业务运行和对外展示的可视化需求,建成电网公司“汇集信息、全面监控、专业联动、统一展现”的电网信息综合展现可视化系统。该系统支持视频、GIS专题图和图形图表等多种展现方式,适用于不同的需求场景,在海南电网公司的实际应用结果证明了其可行性。     

智能电网隐私保护技术的分析研究

融合各种新兴的技术,智能电网的出现给电力系统带来了很大的变革。但和所有新兴事物一样,智能电网也将面临新的风险,尤其是用户侧的风险。面对智能电网用户侧风险,如何保护用户隐私权,是亟待解决的问题,也是文中研究的重点。首先分析了智能电网带来的风险,尤其是用户侧面临的风险;接着,介绍了智能电网隐私保护,对智能电网隐私保护特点进行了详细的解读;最后,结合电力系统,对我国智能电网隐私保护进行了思考,对未来的工作给出了建议。     

一种安全的随机网格视觉密码方案

随机网格视觉密码体制与传统视觉密码体制相比,其数据量得到极大的降低.但这种视觉密码的安全性能却不理想.在传统随机网格体制基础上,提出了一种新的随机网格视觉密码方案.该方案与传统随机网格视觉密码相比,只增加了很少的数据量,却极为有效地提高了随机网格视觉密码的安全性能.     

一种可搜索无限个邻域的改进A*算法

传统A*算法在栅格地图上进行路径规划时,求解得到的路径长度不是最短并且转折点较多.针对这些不足,提出了一种改进A*算法,将传统A*算法的可搜索邻域个数从离散的8个拓展为无限个,可以沿任意方向进行搜索.这样不仅求解出来的路径长度更短,并且大大降低了其转折点的个数.该算法被应用于自主研发的"智能先锋"号系列无人驾驶车辆上,实车试验以及它们在"中国智能车未来挑战赛"中的优异表现证明该方法能够在栅格地图中求解出一条更优的可行驶路径,可以显著提升无人驾驶车辆行驶的效率和平稳性.     

DAMPED OSCILLATIONS IN A BINGHAM PLASTIC FLUID

Stokes' second problem, the propagation and damping of waves into a semi-infinite fluid generated by harmonic oscillations of a flat pate on the surface, is solved for the simple Bingham Theological constitutive model. The solution reveals the existence of “windows” in the distance-time-stress space in which shearing is possible whereas outside these restricted regions no shearing can occur. Within these restricted regions the wave forms developed are exponentially damped, traveling waves which propagate from the excitation plane into the fluid and disappear along definitely prescribed boundaries determined by the yield condition. The most significant consequence of the existence of these “windows” of shear is that even very small yield stresses will radically modify the induced velocity wave patterns from that which would be expected based upon the classical Newtonian fluid solution of Stokes' second problem. At least in this physical setting, it is not necessary for shear to occur globally for motion to occur anywhere, as has been postulated in some recent studies of complex motions. Thus, the motion is consistent with a simple Bingham model which does possess a yield stress.