ID 原文 译文
58538 针对当前电力数据网风险预警模型无法有效预测网络风险的现状，提出一种基于熵权－灰色模型的电力数据网风险预测机制，重点解决网络整体风险的预测问题. Aiming at the present situation that risk prediction model of power data network cannot effec?tively predict the risk，a risk prediction mechanism of power data network based on entropy weight-graymodel is proposed. This paper focuses on risk prediction of the entire network.
58539 首先利用灰色模型对电力数据网的风险指标进行预测，确定单项风险指标值; Firstly，the gray model isused to predict the risk indexes of power data network，and the individual risk index value is determined.
58540 然后采用熵权法得出每一项指标的动态权重; Then，the dynamic weight of each index is obtained by entropy weight method.
58541 最后根据风险指标值和权重得出网络整体的预测风险值. Finally，it can calculatethe risk value of the network according to the risk index value and the weight.
58542 仿真验证结果表明，该模型可保证动态网络下实时预测的准确度. The simulation results showthat the proposed model can guarantee predictive accuracy of dynamic real-time network.
58543 提出一种基于生成对抗网络的遮挡图像修复算法，能够在大量像素缺失的场景下复原出图像的本来面目. A masked image inpainting algorithm based on generative adversarial nets was proposed，which can restore the original image from the lacking of a large number of pixels.
58544 该算法不同于其他的样本块搜索复原算法，可直接生成并且填充可能的缺失元素，改进了生成对抗网络生成模型的结构和生成损失的计算方法，具有半监督学习的特点. Unlike other blocksearch restoration algorithms，the algorithm proposed directly generates possible missing elements and re?store them. Due to the improved structure of generated model and the calculation method of generatingloss on generative adversarial nets，this article has the characteristics of semi-supervised learning.
58545 实验结果表明，在满足图像整体轮廓的前提下，新算法优于其他算法. Exper?iments show that the proposed method outperforms the existing one on the premise of satisfying the overallcontour of the image.
58546 针对联合采用超时机制和通配符匹配机制的规则放置问题，提出了基于时空联合的规则放置算法 TSＲPM，综合考虑了规则在流表中的逗留时间和规则的匹配空间，以确定规则的放置方案. Time-space based rules placement method ( TSＲPM) is proposed for solving the problem ofrules placement utilizing both timeout and wildcard match schemes. The sojourn time and space of a ruleare both considered to decide rules placement approach in TSＲPM.
58547 实验结果表明，所提算法产生的规则放置方案能够有效提高规则命中率，降低分组拒绝率. The results of simulation show thatTSＲPM can improve the match rate and reduce the packet refused rate.