Improving the accuracy of transformer dissolved gas analysis is always an important demand for power ***,the requirement for large numbers of fault samples becomes an obstacle to this *** article creatively uses a lar...
详细信息
Improving the accuracy of transformer dissolved gas analysis is always an important demand for power ***,the requirement for large numbers of fault samples becomes an obstacle to this *** article creatively uses a large number of health data,which is much easier to obtain by power companies,to improve diagnosis *** investigations from the view of both data set and methodology to deal with this problem are presented.A data set consists of 9595 health samples and 993 fault samples is used for *** characteristics of the data set and the influence of the health data on diagnostic accuracy are *** performance of many state‐of‐art algorithms that handle the imbalanced prob-lem is ***,an efficient fault diagnosis algorithm named self‐paced ensemble(SPE)is *** SPE,classification hardness is proposed to include the data characteristic in the *** method can guarantee the diversity of the data set and keep high *** to the experiment results,the superior of SPE is confirmed and also proves that involving more health samples can improve transformer diagnosis when fault data are limited.
Applications such as online virtual fitting room for clothes demand massive computing powers in order to obtain real-time and high fidelity *** cluster provides the infrastructure and solution to solve large scale, co...
详细信息
Applications such as online virtual fitting room for clothes demand massive computing powers in order to obtain real-time and high fidelity *** cluster provides the infrastructure and solution to solve large scale, computation intensive and high throughput problems like fine-grained cloth simulation. In this paper, some key techniques for cluster computing based online virtual fitting room are discussed and a prototype system is implemented. The experiment results show that the proposed architecture can achieve real-time, high-fidelity cloth simulation and provide encouraging online virtual fitting experiences.
暂无评论